Quality of Health Care for Medicare Beneficiaries: A Chartbook Focusing on the Elderly Living in the Community S H E I L A L E AT H E R M A N A N D D O U G L A S M C C A R T H Y U N C P R O G R A M O N H E A LT H O U T C O M E S , S C H O O L O F P U B L I C H E A LT H T H E U N I V E R S I T Y O F N O R T H C A R O L I N A AT C H A P E L H I L L The Commonwealth Fund One East 75th Street New York, NY 10021-2692 Telephone 2 1 2 . 6 0 6 . 3 8 0 0 Facsimile 2 1 2 . 6 0 6 . 3 5 0 0 E-mail c mw f @ c mw f. o rg Web w w w. c mw f. o rg M AY 2 0 0 5 Quality of Health Care for Medicare Beneficiaries: A Chartbook Focusing on the Elderly Living in the Community S H E I L A L E AT H E R M A N A N D D O U G L A S M C C A R T H Y U N C P R O G R A M O N H E A LT H O U T C O M E S , S C H O O L O F P U B L I C H E A LT H T H E U N I V E R S I T Y O F N O R T H C A R O L I N A AT C H A P E L H I L L M AY 2 0 0 5 Support for this research was provided by The Commonwealth Fund. The views presented here are those of the authors and should not be attributed to The University of North Carolina or The Commonwealth Fund or its directors, officers, or staff. Copies of this report (#815) are available on The Commonwealth Fund Web site at: www.cmwf.org. Authors Sheila Leatherman Research Professor, School of Public Health, and Senior Investigator, UNC Program on Health Outcomes, The University of North Carolina at Chapel Hill; and Senior Associate, Judge Institute, University of Cambridge, England Douglas McCarthy President, Issues Research, Inc., Durango, Colo. (under contract to The University of North Carolina at Chapel Hill) Expert Consultants Christine Cassel, M.D. President, American Board of Internal Medicine, Philadelphia, Pa. John Rother Director of Policy and Strategy, AARP, Washington, D.C. Paul Shekelle, M.D., Ph.D. Director, Southern California Evidence-Based Practice Center, RAND, Santa Monica, Calif. Editorial and Scientific Consultants Kathleen Lohr, Ph.D. Research Professor, School of Public Health, and Senior Investigator, UNC Program on Health Outcomes, The University of North Carolina at Chapel Hill; and Distinguished Fellow, Research Triangle Institute Joanne Garrett, Ph.D. Professor, Department of Medicine, The University of North Carolina at Chapel Hill Advisory Board Anne-Marie Audet, M.D., M.Sc. Assistant Vice President, The Commonwealth Fund, New York, N.Y. Donald Berwick, M.D., M.P.P. President and CEO, Institute for Healthcare Improvement, Cambridge, Mass. Robert Galvin, M.D. Director, Global Health Care, General Electric Company, Fairfield, Conn. Judith Hibbard, Dr.P.H. Professor, Department of Planning, Public Policy, and Management, University of Oregon, Eugene Elizabeth McGlynn, Ph.D. Associate Director, RAND Health, Santa Monica, Calif. William Roper, M.D., M.P.H. Dean, School of Medicine, and Vice Chancellor for Medical Affairs, The University of North Carolina at Chapel Hill; and Chief Executive Officer, UNC Health Care System Project Administration Sue Tolleson-Rinehart, Ph.D. Administrator, UNC Program on Health Outcomes, The University of North Carolina at Chapel Hill Sara Massie Research Assistant, UNC Program on Health Outcomes, The University of North Carolina at Chapel Hill Design Jim Walden Walden Creative, LLC, Bayfield, Colo. Table of Contents Acknowledgments • page 4 Table of Charts • page 5 Summary and Highlights • page  Introduction • page 7 Methods • page 22 Terminology • page 23 Charts Effectiveness • page 25 Patient Safety • page 73 Access and Timeliness • page 87 Patient and Family Centeredness • page 0 Equity • page 5 Capacity to Improve • page 35 Appendices • page 53 References • page 75 About the Authors • page 82 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 3 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 4 Acknowledgments This chartbook is a compendium of data representing the efforts of many researchers, Carol Roth, RAND to whom we are indebted for their contributions. We especially wish to thank the Bob Schlenker, University of Colorado Health Sciences Center individuals who kindly provided data, sources, information, or clarification of results: Ravi Sharma, Westat Bernard Shulamit, RTI International Steven Asch, RAND Judy Strait-Jones, State of Hawaii Dept. of Health Katherine Baicker, Dartmouth College Bruce Stuart, University of Maryland School of Pharmacy Gerrie Barosso, University of Minnesota, Research Data Assistance Center Judith Swan, National Cancer Institute Sharon Bee Cheng, Medicare Payment Advisory Commission Sally Trude, Center for Studying Health System Change Jacqueline Bender, Center for Medicare Advocacy Neil Wenger, University of California-Los Angeles Christina Boccuti, Medicare Payment Advisory Commission Alan White, Abt Associates Dale Bratzler, Oklahoma Foundation for Medical Quality Ira Byock, Dartmouth Medical School We gratefully acknowledge the expert consultants and advisory board members Olivio Clay, University of Alabama at Birmingham (listed on the credits page) who kindly provided comments, advice, and assistance to Karyn Collins, University of Montana improve and disseminate the chartbook. We thank Kathy Lohr and Joanne Garrett Jalpa Doshi, University of Pennsylvania for exceptional effort in reviewing our work and Paul Shekelle for facilitating use of Susan Enguidanos, Partners in Care Foundation baseline data from the ACOVE-2 study. Margaret Fang, University of California-San Francisco We thank the staff at The Commonwealth Fund and especially Anne-Marie Audet, Elizabeth Goldstein, Centers for Medicare and Medicaid Services Karen Davis, Cathy Schoen, and Stephen Schoenbaum for their advice and support, and Edwin Huff, Centers for Medicare and Medicaid Services Paul Frame, Chris Hollander, Mary Mahon, Bill Silberg, and the communications team David Hunt, Centers for Medicare and Medicaid Services for support in production and dissemination. We are grateful to Sue Tolleson-Rinehart Dana Gelb Safran, Tufts-New England Medical Center and Sara Massie at the UNC Program on Health Outcomes for administrative support. Paul Ginsburg, Center for Studying Health System Change We acknowledge Kara Nyberg and Elizabeth Staton for assistance in drafting Peter Groeneveld, VA Medical Center, Philadelphia narratives, and Myoung Fry for assistance with the literature search. Jim Walden Wanda Johnson, Oklahoma Foundation for Medical Quality demonstrated great flexibility in working with us under a demanding design schedule. Edward Kelley, Agency for Healthcare Research and Quality The authors retain sole responsibility for any errors, omissions, or other Russell Mardon, National Committee for Quality Assurance shortcomings in the content of the chartbook. We regret that we were not able to Karen Milgate, Medicare Payment Advisory Commission include all the information that was provided us. Mary Mittelman, New York University School of Medicine Cover photo: Andersen Ross / Photodisc Red / Getty Images Greg O’Neill, National Academy on an Aging Society Table of Charts # SUBJECT/PERSPECTIVE YEAR(S) P O P U L AT I O N SOURCE PAG E 1: EFFECTIVENESS S TAY I N G H E A LT H Y : P R E V E N T I O N A N D H E A LT H P R O M O T I O N 1:1 Immunization of Elderly Adults 1989 to 2003 (national) Community-dwelling elderly adults (ages National Health Interview Survey; Behavioral 27 and 2003 (states) 65+) Risk Factor Surveillance System 1:2 Breast Cancer Screening 1987 to 2000 (national) Community-dwelling elderly v. middle-age National Health Interview Survey; Behavioral 29 and 2002 (states) adults (ages 50–64, 65+) Risk Factor Surveillance System 1:3 Colorectal Cancer Screening 2000 (national) and Community-dwelling elderly v. middle-age National Health Interview Survey; Behavioral 31 2002 (states) adults (ages 50–64, 65+) Risk Factor Surveillance System 1:4 Osteoporosis Counseling and Screening 2000 Community-dwelling female elderly Medicare Medicare Current Beneficiary Survey 33 beneficiaries (ages 65+) 1:5 Falls and Instability: Screening and 2000–2001 At-risk older patients of two medical groups Medical records (Assessing Care of Vulnerable 35 Management (ages 75+) Elders study) GET TING BET TER: ACUTE CARE FOR ILLNESS OR INJURY 1:6 Inappropriate Use of Antibiotics for the 1997–1998 v. 2000–2001 Community-dwelling elderly v. middle-age National Ambulatory Medical Care Survey; 37 Common Cold adults (ages 45–64, 65+) National Hospital Ambulatory Medical Care Survey 1:7 Hospital Treatment for Pneumonia 2002 (national and states) Hospitalized Medicare fee-for-service Medicare Quality Improvement Organization 39 beneficiaries (all ages) program data 1:8 Hospital Treatment for Heart Attack 2002 Hospitalized Medicare fee-for-service Medicare Quality Improvement Organization 41 beneficiaries (ages 65–74, 75–84, 85+) program data 1:9 Time to Reperfusion for Heart Attack 2000–2001 Hospitalized Medicare fee-for-service Medicare Quality Improvement Organization 43 beneficiaries (all ages) program data 1:10 Hospital Mortality 1995 v. 2002 Hospitalized Medicare fee-for-service Medicare administrative data (AHRQ Inpatient 45 beneficiaries (all ages) Quality Indicators) 1:11 Hospitalizations for Ambulatory Care 1995 v. 2002 Hospitalized Medicare fee-for-service Medicare administrative data (AHRQ 47 Sensitive Conditions beneficiaries (all ages) Prevention Quality Indicators) AHRQ = Agency for Healthcare Research and Quality Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 5 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 6 # SUBJECT/PERSPECTIVE YEAR(S) P O P U L AT I O N SOURCE PAG E LIVING WITH ILLNESS: ONGOING CARE FOR CHRONIC CONDITIONS 1:12 High Blood Pressure Awareness 1988–1994 v. 1999–2000 Community-dwelling elderly v. middle-age National Health Interview Survey; National 49 and Control adults (ages 45–64, 65+) Health and Nutrition Examination Survey 1:13 High Cholesterol Screening, Awareness, 1999–2000 Community-dwelling elderly v. middle-age National Health and Nutrition 51 Treatment, and Control adults (ages 45–64, 65+) Examination Survey 1:14 Cardiovascular Care and Outcomes in 2000 v. 2003 Medicare managed care plans v. employer- HEDIS 53 Managed Care Plans sponsored plans (all ages) 1:15 Stroke Prevention for Outpatients with 1991–1992 v. 1999–2000 Community-dwelling elderly v. nonelderly National Ambulatory Medical Care Survey 55 Atrial Fibrillation adults (ages 18–64, 65–79, 80+) 1:16 Diabetes Management 2001 Community-dwelling elderly v. middle-age Medical Expenditure Panel Survey 57 adults (ages 45–64, 65+) 1:17 Osteoarthritis: Evaluation and Treatment 2002–2003 At-risk older patients of two medical groups Patient interviews and medical records 59 (ages 75+) (Assessing Care of Vulnerable Elders study) 1:18 Urinary Incontinence: Screening and 2000–2001 At-risk older patients of two medical groups Medical records (Assessing Care of Vulnerable 61 Management (ages 75+) Elders study) 1:19 Treatment for Depression 1999–2001 Elderly v. near-elderly patients of 18 clinics Patient interviews 63 (ages 60–64, 65–74, 75+) 1:20 Mental Health Care in Managed Care Plans 2000 or 2001 v. 2003 Medicare managed care plans v. employer- HEDIS 65 sponsored plans (all ages) CHANGING NEEDS 1:21 Home Health Care Outcomes 2002 v. 2004 Adult patients of Medicare-certified home Outcome and Assessment Information Set 67 health care agencies (ages 18+) S U M M A R Y P E R F O R M A N C E A C R O S S M U LT I P L E C O N D I T I O N S 1:22 State-Level Performance on Medicare 1998–1999 v. 2000–2001 Medicare fee-for-service beneficiaries Medicare Quality Improvement Organization 69 Quality Indicators (all ages) program data 1:23 Assessing Care of Vulnerable Elders 1998–1999 Community-dwelling vulnerable elderly Patient interviews and medical records 71 (ACOVE) members of two health plans (ages 65+) (Assessing Care of Vulnerable Elders study) # SUBJECT/PERSPECTIVE YEAR(S) P O P U L AT I O N SOURCE PAG E 2 : PAT I E N T S A F E T Y 2:1 Adverse Events and Postoperative 2002 Hospitalized Medicare fee-for-service Medicare Patient Safety Monitoring System 75 Complications of Care beneficiaries (all ages) (medical records) 2:2 Trends in Adverse Events and 1995 v. 2002 Hospitalized Medicare fee-for-service Medicare administrative data (AHRQ Patient 77 Complications of Care in the Hospital beneficiaries (all ages) Safety Indicators) 2:3 Adverse Events and Complications of Care 2001 Hospitalized elderly v. middle-age adults Health Care Utilization Project, 79 in the Hospital, by Patient Age (ages 45–64 and 65+) National Inpatient Sample (AHRQ Patient Safety Indicators) 2:4 Appropriate Use of Antibiotics to Prevent 2001 Hospitalized Medicare fee-for-service Medicare National Surgical Infection 81 Surgical Infections beneficiaries (all ages) Prevention Project (medical records) 2:5 Potentially Inappropriate Prescribing for 1996 v. 1998 v. 2000 Community-dwelling elderly adults Medical Expenditure Panel Survey 83 the Elderly (ages 65+) 2:6 Preventable Adverse Drug Events in 1999–2000 Medicare beneficiaries seen in a large multi- Incident reports, hospital discharge 85 Ambulatory Care specialty group practice summaries, and medical records 3: ACCESS AND TIMELINESS 3:1 Unmet Need and Delay in Seeking Care 2002 Community-dwelling elderly v. nonelderly National Health Interview Survey 89 adults (ages 45–64, 65+) 3:2 Financial Barriers to Prescription 2001 Community-dwelling elderly in 8 states Kaiser/Commonwealth Fund/Tufts-New 91 Adherence (ages 65+) England Medical Center, Survey of Seniors 3:3 Financial Barriers to Access 1996–1999 Elderly Medicare fee-for-service beneficiaries Medicare Current Beneficiary Survey (Access 93 (ages 65+) to Care for the Elderly Project indicators) 3:4 No Usual Source of Health Care 1993 v. 1997 v. 2002 Community-dwelling elderly v. middle-age National Health Interview Survey 95 adults (ages 45–64, 65+) 3:5 Waiting Time for Physician Visits for a 1997 v. 2003 Community-dwelling elderly v. near-elderly Center for Studying Health System Change, 97 Specific Illness adults (ages 55–64, 65+) Community Tracking Study 3:6 Use of Hospice at End of Life 1998 v. 2002 Medicare fee-for-service beneficiaries v. Medicare administrative data 99 Medicare managed care plan members (all ages and ages 65+) Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 7 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 8 # SUBJECT/PERSPECTIVE YEAR(S) P O P U L AT I O N SOURCE PA G E 4 : PAT I E N T A N D FA M I LY C E N T E R E D N E S S 4:1 Experiences with Insurance and Care 2001 Community-dwelling elderly v. nonelderly Commonwealth Fund Survey of Health 103 adults (ages 19–65, 65+) Insurance 4:2 Consumer Assessment of Health Plans 2003 Medicare fee-for-service beneficiaries v. Medicare Consumer Assessment of Health 105 Medicare managed care plan members Plans Survey (CAHPS) (all ages) 4:3 Interpersonal Quality of Care 2001 Community-dwelling elderly v. middle-age Medical Expenditure Panel Survey 107 adults (ages 45–64, 65+) 4:4 Beneficiary Knowledge of Medicare and 1998 v. 2002 Community-dwelling elderly Medicare Medicare Current Beneficiary Survey; 109 Accuracy of Medicare Information beneficiaries (ages 65+) Government Accountability Office 4:5 to Family Ratings of Quality of Care at End of 2000 U.S. adults who died of a chronic illness and Nationally representative mortality follow- 111 to 4:7 Life: Parts I, II, and III used health care at end of life (avg. age 74) back telephone survey 113 # SUBJECT/PERSPECTIVE YEAR(S) P O P U L AT I O N SOURCE PAG E 5: EQUITY P AT I E N T S A F E T Y 5:1 Racial and Ethnic Disparities in Adverse 2001 Hospitalized elderly adults (ages 65+) Health Care Utilization Project, State Inpatient 117 Events and Complications of Care Database (AHRQ Patient Safety Indicators) S TAY I N G H E A LT H Y : P R E V E N T I O N A N D H E A LT H P R O M O T I O N 5:2 Disparities in Preventive Care 1998, 2000, 2001 Community-dwelling elderly adults (ages 65+) National Health Interview Survey 119 by Race and Ethnicity 5:3 Disparities in Preventive Care: 1998, 2000, 2001 Community-dwelling elderly adults (ages 65+) National Health Interview Survey 120 by Family Income Level 5:4 Disparities in Preventive Care: 1998, 2000, 2001 Community-dwelling elderly adults (ages 65+) National Health Interview Survey 121 by Type of Coverage 5:5 Impact of Medicare Coverage on Receipt of 1996 v. 2000 Near-elderly adults in 1996 who became Health and Retirement Study 123 Preventive Care eligible for Medicare in 2000 (ages 60–64, 65+) LIVING WITH ILLNESS: ONGOING CARE FOR CHRONIC CONDITIONS 5:6 Racial and Ethnic Disparities in Chronic 1999 Medicare managed care plan members HEDIS 125 Care Management (all ages) C A R E AT E N D O F L I F E 5:7 Racial and Ethnic Disparities in Use of 1998 v. 2002 Medicare fee-for-service beneficiaries who Medicare administrative data 127 Hospice at End of Life died (all ages) 5:8 Unexplained Variation in Care at End of Life 1999–2000 Medicare fee-for-service beneficiaries treated Medicare administrative data 129 at 77 U.S. hospitals during their last six months of life S U M M A R Y P E R F O R M A N C E A C R O S S M U LT I P L E C O N D I T I O N S 5:9 Relationship Between Medicare Spending 2000–2001 Medicare fee-for-service beneficiaries Medicare administrative data and Quality 131 and Quality of Care (all ages) Improvement Organization program data 5:10 Physician Perceptions of Quality of Care for 2000–2001 Elderly white and black Medicare fee-for- Community Tracking Study Physician Survey 133 White and Black Patients service beneficiaries (ages 65+) and Medicare administrative data Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 9 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 10 # SUBJECT/PERSPECTIVE YEAR(S) P O P U L AT I O N SOURCE PA G E 6: C A PAC I T Y TO I M P R O V E S TAY I N G H E A LT H Y : P R E V E N T I O N A N D H E A LT H P R O M O T I O N 6:1 Improving Pneumococcal 2000–2001 v. 2001–2002 Elderly v. nonelderly patients with pneumonia Medical records (Louisiana State University, 137 Vaccination Screening in the Hospital at a Louisiana teaching hospital (all ages and Internal Medicine inpatient service) ages 65+) GET TING BET TER: ACUTE CARE FOR ILLNESS OR INJURY 6:2 Improving Hospital Treatment for 1998–1999; 2000 Medicare heart attack patients in 10 Michigan Medical records (American College 139 Heart Attack hospitals (all ages) of Cardiology Guidelines Applied in Practice Initiative) LIVING WITH ILLNESS: ONGOING CARE FOR CHRONIC CONDITIONS 6:3 Reducing Rehospitalization for Congestive 1997–2001 Elderly congestive heart failure patients Patient interviews and medical records 141 Heart Failure treated at six Philadelphia hospitals (ages 65+) (Quality-Cost Model of Advanced Practice Nursing Transitional Care) 6:4 Improving Depression Treatment 1999–2001 Elderly and near-elderly patients with Patient interviews (Improving 143 and Outcomes depression or dysthymia in eight health care Mood: Promoting Access to Collaborative organizations (ages 60+) Treatment program) 6:5 Supporting Caregivers of Patients with 1987–1996 Spouse-caregivers receiving support services Caregiver interviews (New York University 145 Alzheimer’s Disease at a university-affiliated Alzheimer’s center Spouse-Caregiver Intervention Study) (avg. age 71) CHANGING NEEDS 6:6 Improving Home Health Care Outcomes 1995–1999 Patients of 54 Medicare-certified home health Outcome and Assessment Information Set 147 care agencies in 27 states (predominantly (Outcome-Based Quality Improvement older adults) Demonstration) 6:7 Program of All-Inclusive Care for the 1995–1997 Frail elders at risk of nursing home placement Patient interviews (Evaluation of the PACE 149 Elderly (PACE) at 11 demonstration sites (avg. age 78) Demonstration) C A R E AT E N D O F L I F E 6:8 Expanding Palliative Care Options at the 1999–2000 HMO patients with life-threatening chronic Patient interviews and administrative data 151 End of Life illnesses (avg. age 70) (Kaiser Permanente Palliative Care Program) Summary and Highlights We conducted a broad review of recently published studies and as screening and treatment for those with urinary incontinence or at reports to present a coherent picture of the quality of health care for risk of falls. Up to half of family members report concerns with the elderly Medicare beneficiaries living in the community. We included care provided to a relative at the end of life. findings for all Medicare beneficiaries when available data were not Disparities and unjustified variations in care appear to be no limited to the community-dwelling elderly population. less an issue for the elderly, despite near-universal coverage by The results, displayed in 60 charts, reveal many signs of progress, Medicare. Minority and low-income elders and those without any especially in areas that have been targeted as national priorities. supplemental insurance coverage are less likely to get recommended However, there are also significant gaps and deficiencies in care and preventive care. Minorities, especially blacks, are more likely to wide variation in quality across the country. While Medicare appears experience certain preventable adverse events or complications of to be working well as an insurance program in providing the elderly care in the hospital. The physicians of elderly black patients are more with access to needed care, there must be increased efforts to assure likely to report barriers to providing high-quality care. The amount systematic and predictable improvements in the quality of care. of care received at the end of life varies dramatically depending on On the positive side, improvements can be seen in the provision where one receives it. of preventive services such as mammography, in hospital treatment The good news is that change is possible with concerted effort. of heart attack, and in outpatient care for chronic conditions such The chartbook highlights eight exemplary quality improvement as diabetes. Fewer elderly patients are receiving inappropriate interventions that offer promising approaches for critical needs, medications and fewer are dying in the hospital after being treated such as reducing repeat hospitalizations for patients with heart for heart failure, stroke, pneumonia, and other conditions. More failure, supporting spouses caring for patients with Alzheimer’s seniors have a usual source of care, an important predictor of getting disease, helping frail elderly maintain their independence at home, preventive care and having health care needs met. and providing palliative care at the end of life. The Medicare On the negative side, large gaps need to be addressed in screening program has several initiatives under way that hold the promise for colorectal cancer, treatment for depression, and control of high of encouraging improvements in access to and quality of care. Yet, blood pressure and high cholesterol among the elderly. Potentially greater effort is needed to assure that all Medicare beneficiaries preventable hospitalizations have increased for certain conditions, consistently receive the best care that the American health care as have recorded rates of adverse events or complications of care in system has to offer and to reliably expand capacity for continually the hospital. Pilot studies suggest that many vulnerable elderly are improving quality over time. not receiving care that is important to well-being in later life, such Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 11 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 12 Chartbook Highlights: Signs of Progress in Improving Quality Since 989, the rate of influenza vaccination has doubled and the The proportion of Medicare beneficiaries using hospice care at the rate of pneumococcal vaccination has quadrupled among the elderly. end of life increased by 9 percentage points from 998 to 2002, and However, one-third did not get an annual flu shot in 2003; almost the use of hospice was nearly equalized among all age groups in one-half had never received the pneumococcal vaccine (Chart :). contrast to a marked age disparity in rates of use in 998 (Chart 3:6). Risk-adjusted hospital mortality rates decreased by 0 percent More seniors reported that they had a regular place to go for health to 3 percent from 995 to 2002 among Medicare beneficiaries care in 2002 than in 993, an important determinant of getting hospitalized for eight conditions and procedures, such as coronary recommended preventive care (Chart 3:4). artery bypass surgery. In contrast, risk-adjusted mortality rates measured 30 days after hospitalization worsened from 2000 to 2002 Beneficiary knowledge about the Medicare program increased for six of the same eight conditions or procedures (Chart :0). from 998 to 2002, although more than half indicated they didn’t have the information they needed in 2002. Only six of 0 calls to Physicians more often prescribed anticoagulant medication to the Medicare information line were answered correctly in a 2004 help prevent strokes among their highest risk elderly patients government audit (Chart 4:4). with atrial fibrillation (irregular heart beat), but about one-half of patients still did not receive these potentially life-saving drugs Compared to privately insured nonelderly adults in 200, elderly during 999–2000 (Chart :5). Medicare beneficiaries were more likely to rate their insurance highly and to be satisfied with their care. Moreover, they were The proportion of seniors who were taking potentially less likely to report problems with coverage and access to care inappropriate medications declined by 37 percent from 996 (Chart 4:). to 2000 (Chart 2:5). The rate at which seniors were prescribed antibiotics for the common cold decreased 44 percent from 997– One study found that after older adults became eligible for Medicare 998 to 2000–200 (Chart :6). at age 65, preexisting disparities in screening were reduced between those who were insured and those who were uninsured before Functional outcomes for patients of Medicare-certified home health enrolling in Medicare (Chart 5:5). care agencies improved by  to 5 percentage points from 2002 to 2004 across nine indicators of quality (Chart :2). Looking across 22 indicators of quality of care for Medicare beneficiaries, performance in the median state improved from 69.5 percent during 998–999 to 73.4 percent during 2000–200 (Chart :22). Chartbook Highlights: Examples of Deficiencies in Quality Less than one-third of depressed elderly patients in one study In one pilot study, the quality of care delivered to vulnerable received potentially effective treatment during 999–200 (Chart elderly—those at high risk of declines in health—met expert :9). Only 60 percent of Medicare managed care plan members standards only a little more than half the time. The greatest gaps hospitalized for mental illness in 2003 received recommended occurred in the care of geriatric conditions, such as screening and follow-up care within one month of leaving the hospital (Chart :20). management of falls and urinary incontinence (Chart :23). From 995 to 2002, rates of potentially preventable hospitalizations Among elderly adults in 2000, one-half had not received a among Medicare beneficiaries increased for seven of 2 conditions, colorectal cancer screening test as recommended (Chart :3). such as a 24 percent increase in the rate of hospitalization due to Similarly in 2000, only one-half of elderly women had ever talked to bacterial pneumonia (Chart :). their doctor about osteoporosis (Chart :4). Risk-adjusted rates of potentially preventable adverse events or Although hospital treatment of Medicare pneumonia patients complications of care increased for nine of 3 indicators from complied with one of three evidence-based standards 63 percent 995 to 2002, as recorded in hospital billing records for Medicare to 8 percent of the time in 2002, only 30 percent received care beneficiaries (Chart 2:2). consistent with all three recommended care standards (Chart :7). About half of Medicare patients undergoing selected surgeries in High blood pressure and high cholesterol are two major, 200 did not receive prophylactic antibiotics in a timely manner modifiable risk factors for heart disease. Only one-quarter of elderly consistent with evidence about how most effectively to prevent adults whom researchers determined had high blood pressure had postoperative infections (Chart 2:4). it under control during 999–2000 (Chart :2). Likewise, only 8 percent of those that researchers determined had high cholesterol Among those who died of a chronic condition in 2000 and received had it controlled (Chart :3). care at the end of life, 5 percent to 50 percent of their family members expressed concerns about some aspects of the care delivered at the end of life (Charts 4:5 to 4:7). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 13 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 14 Chartbook Highlights: Examples of Disparities and Unjustified Variations in Care Among Medicare beneficiaries enrolled in managed care plans during 999, blacks were less likely than whites to receive recommended Minority elderly patients were more likely than white elderly chronic care services, such as beta-blocker medications after a heart patients to suffer certain potentially preventable adverse events attack or blood tests to check on control of diabetes. Hispanics, or complications of care among those hospitalized during 200. Asian Americans, and Native Americans were less likely than whites For example, black patients were 2.3 times more likely than white to receive some services but equally or more likely to receive other patients to suffer a pressure sore during a hospital stay of five days or services or to achieve good outcomes (Chart 5:6). longer (Chart 5:). The amount of care provided to chronically ill Medicare In national surveys conducted among community-dwelling elderly beneficiaries during the last six months of their lives varied adults during 998, 2000, and 200: greatly (three-fold to 4-fold difference in rates from highest to lowest) among 77 hospitals during 999–2000, suggesting that where · Minorities were less likely than whites to receive some preventive one receives care—rather than individual medical need—determines services. For example, Asian Americans were half as likely to have the amount of care that is provided (Chart 5:8). ever received a pneumococcal vaccination as of 200; Hispanics were almost one-third less likely to have ever received sigmoidoscopy or States with higher spending per Medicare beneficiary tended to rank colonoscopy as of 2000 (Chart 5:2). lower on 22 quality of care indicators. This inverse relationship might reflect medical practice patterns that favor intensive, costly · Those with lower income were less likely than those with higher care rather than the effective care measured by these indicators income to receive most preventive services studied. In 2000, for (Chart 5:9). example, only 56 percent of poor elderly women had received a mammogram in the past two years as compared to 83 percent of During 2000–200, physicians visited predominantly by black high-income elderly women (Chart 5:3). Medicare patients were less likely than physicians visited · Seniors with private supplemental coverage (such as retiree coverage predominantly by white Medicare patients to report that they can or a Medigap plan) were more likely to receive the preventive deliver and obtain access to high-quality care for their patients services studied than were low-income seniors who are dually (Chart 5:0). eligible for Medicare and Medicaid or those seniors without any supplemental coverage (Chart 5:4). Chartbook Highlights: Family members who care for a relative with Alzheimer’s disease Eight Exemplary Interventions to Improve Quality of Care often experience psychological distress. Providing spouse- for Medicare Beneficiaries caregivers with intensive counseling and ongoing support reduced their burden of depression compared to the burden in a control An educational intervention for internal medicine physicians at a group. Alzheimer’s patients whose spouses received enhanced teaching hospital resulted in a 72 percentage point improvement in services were cared for at home nearly a year longer before being the proportion of elderly pneumonia patients screened to determine institutionalized (Chart 6:5). whether they needed the pneumococcal vaccine and a 34 percentage point increase in those given the vaccine when needed (Chart 6:). The hospitalization rate fell by 22 percent over three years among home health care agencies that used regular reports on their Medicare patients at 0 southeastern Michigan hospitals were more patients’ outcomes to plan and make improvements in care as part of likely to receive evidence-based treatment when caregivers used a national demonstration program (Chart 6:6). customized, guideline-oriented tools, such as standard admission orders, clinical pathways, and standard discharge forms, as part of a PACE (Program of All-Inclusive Care for the Elderly) serves frail structured intervention to improve heart attack treatment (Chart 6:2). elders eligible for Medicare and Medicaid who are at risk of nursing home placement. An interdisciplinary team based at an adult Hospital readmissions were reduced by 36 percent when elderly day care center provides health care and supportive services. patients with heart failure received individualized transitional care Participants enrolled in PACE demonstrations in  cities spent from an advanced practice nurse who provided needs assessment, fewer days in a hospital or nursing home, had equal or better care planning, patient education, and therapeutic support through outcomes, were less likely to die during the demonstration, and had discharge planning and home follow-up visits. Implementing such a lower Medicare costs per participant than those in a comparison program nationally for all Medicare beneficiaries could prevent up group (Chart 6:7). to 84,000 hospital readmissions each year (Chart 6:3). Some people with life-threatening chronic illnesses do not qualify Older adults with depression were more likely to receive for hospice care because of uncertain prognosis or because they treatment and to achieve better outcomes when a trained nurse wish to continue receiving some curative care. A palliative care or psychologist collaborated with the patient and primary care program that allowed participants to receive gradually more physician to support medication management and/or provide brief supportive services at home enabled more of them to die at home, psychotherapy under supervision of a psychiatrist and primary care with increased satisfaction and at lower cost than for a comparison expert (Chart 6:4). group (Chart 6:8). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 15 Introduction The federal government’s financial responsibility to provide problems in health care quality in the United States as a access to health care for the nation’s 4 million Medicare “chasm,” requiring new approaches to quality improvement beneficiaries implies a concomitant obligation to assure that at both the local and national levels (IOM 200a, 2004). funds spent on behalf of taxpayers achieve the overarching The IOM envisioned a redesign of local health care delivery goals of the health care system: “to continually reduce the systems accompanied by new policies to promote high-quality burden of illness, injury, and disability, and to improve the health care through regulatory and payment incentives health and functioning of the people of the United States” and the application of health information technologies that (IOM 200a). These goals are realized more specifically by can positively influence the way in which physicians and assuring that the 280 billion spent for Medicare health care organizations work. services are delivered to beneficiaries in a safe, effective, Quality of Health Care for Medicare Beneficiaries is the timely, patient-centered, equitable, and efficient manner. third in a series of chartbooks intended to help achieve these The Medicare program has taken great strides in its goals by providing a common understanding of the magnitude capacity to influence the quality of health care since Congress and scope of quality problems among the many stakeholders first created the Professional Standards Review Organizations interested in improving the performance of the American in 972. The Institute of Medicine’s landmark 990 report on health care system. It presents 60 charts portraying the state quality assurance in Medicare (IOM 990) was instrumental of health care quality in the Medicare program, focusing in shifting the focus from retrospective case review to a more primarily on quality of care delivered to the 35 million elderly systematic and proactive approach. Medicare launched its beneficiaries (ages 65 and older) living in the community Health Care Quality Improvement Program in the 990s to who constitute the great majority of the Medicare program.* promote the wider adoption of professionally developed, The final section profiles some examples of promising quality evidence-based standards of care. The Peer Review improvement initiatives to illustrate that significant change is Organizations have been renamed Quality Improvement indeed possible, even if often difficult to replicate and sustain. Organizations (QIOs) and tasked to work cooperatively with * This chartbook does not address specific quality of care issues for disabled Medicare local health care providers on statewide quality improvement beneficiaries, patients in the Medicare End Stage Renal Disease program, or elderly projects that will advance the national Medicare quality nursing home residents. These population groups have special needs and concerns that deserve attention for quality measurement and improvement, but which we agenda (Sprague 2002). were unable to include within the necessarily limited scope of this project. These These efforts have taken on greater urgency since a series individuals, however, are included in data depicting quality of care for Medicare beneficiaries in general (see Table of Charts). of recent Institute of Medicine reports characterized pervasive Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 17 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 18 O V E R V I E W: P R O G R E S S , G A P S , A N D VA R I AT I O N S little more than half of vulnerable elders receive the care that experts believe is important to the elderly (Wenger et al. 2003) Like its predecessors, this chartbook presents examples both of (see Chart :23). progress in improving health care quality and of deficiencies that point to the need for further action (see Summary and Overuse Highlights). Most indicators that have been measured over Researchers at RAND published a series of studies time show movement in the right direction, with some documenting that about one-third of surgical procedures notable exceptions, such as increasing rates of potentially were performed for inappropriate reasons or had questionable preventable hospitalizations and adverse events. The pace of benefits for Medicare beneficiaries during the 980s and change is often slow, however, and appears to have reached a early 990s (McGlynn and Brook 200). The need to refrain plateau in some areas, such as adult immunizations. Although from providing services to those not likely to benefit (IOM improvements offer inspiration and potential lessons for 200b) is now receiving renewed attention as concerns about application to other areas, large gaps from optimal care remain the affordability and safety of health care are increasingly in too many areas. Even where a relatively high level of quality acknowledged. One of the few indicators of overuse in this has been achieved, constant effort will be needed, as new chartbook suggests limited improvement in that the elderly medical therapies and health care approaches are developed are less often receiving antibiotics for the common cold. and proven, to continuously incorporate the best standards Several charts in the Capacity to Improve section illustrate care and clinical practices into health care delivery. management approaches that reduce costs and risk to patients through the avoidance of hospital admissions and nursing Underuse home stays. In contrast to the Institute of Medicine’s 990 report, which found the problem of underuse hard to document, quality Misuse measurement systems have advanced to the point that we now The Institute of Medicine’s 999 report, To Err is Human, have many good examples of the failure to provide services galvanized national attention to this area of quality, defined based on scientific evidence to all who could benefit (IOM as avoidable complications of appropriate health care 200b). Average performance is approaching the 75 percent (Chassin 99). New measures and sources of data, such as range on widely accepted standards of care represented in the the Patient Safety Indicators developed by the Agency for Medicare Quality Improvement Organization program (Jencks Healthcare Research and Quality and the Medicare Patient et al. 2003) (see Chart :22 and Appendix Table ). However, Safety Monitoring System, are providing useful data to study variation in performance on measures of clinical effectiveness this problem. The major challenge lies in identifying the represented in this chartbook is quite wide, ranging from underlying causes behind misuse and actions that can be taken 0 percent to 90 percent of optimal care. Moreover, smaller to prevent adverse events and harm to patients. studies that focus on geriatric needs and conditions find that Variations in use By race, ethnicity, and income: Racial and ethnicity The data presented in the chartbook suggest that variations disparities in care are pervasive but not monolithic or and disparities in care are wider for services that are relatively consistent from condition to condition or from measure to new or are generally underused. This variation may lessen over measure. This suggests that the determinants of disparities— time as the specific health care services become more widely and by implication the actions likely to reverse disparities—are used and accepted generally. specific to the particular context, although some factors are By patient age: The clinical quality and outcomes of health undoubtedly correlated across conditions. Socioeconomic care for the elderly are sometimes similar to that of nonelderly factors may have a larger influence on disparities in the receipt adults but also varies in both directions—better and worse. of preventive care than race or ethnicity alone. For both elderly and nonelderly, however, quality is typically By geography: State-level variations in compliance with far from optimal. These variations must be interpreted in standards for preventive care can be wide. Variations appear to the context of characteristics of the elderly population (see be narrower for measures, such as mammography, with higher next section for further discussion). The elderly generally overall compliance than for measures with lower overall rates, experience more adverse events or complications of care, for such as colorectal cancer screening. Comparing state quality example, in part because of their more vulnerable physical performance to Medicare spending suggests that high-quality condition. By contrast, elderly Medicare beneficiaries generally health care need not cost more, at least as measured by these report fewer problems with their coverage and access to care indicators (see Chart 5:9). The relationship between quality and they give higher ratings to their health plan and the and cost is complex and likely to be influenced by structural patient-centeredness of their care than nonelderly adults. factors such as physician supply (Baicker and Chandra 2004). By type of coverage: Although Medicare beneficiaries with A growing body of research is finding that racial and ethnic supplemental coverage are more likely to receive high-quality disparities nationally can be attributed in part to geographic care, Medicare as a whole offers important advantages in variations in care because minorities tend to live in areas with meeting the health care needs and expectations of the elderly lower overall use of particular services (Skinner et al. 2003; across all types of coverage. Comparisons by type of coverage Groeneveld et al. 2005). Hence, eliminating disparities cannot must be interpreted with caution, because those who are on depend solely on equalizing care locally. Rather, it requires Medicaid or who do not have supplemental coverage generally achieving the appropriate amount of care for everyone in have lower incomes than those with private supplemental similar need, regardless of where they live. coverage, and low income is an independent risk factor for access barriers. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 19 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 20 CHALLENGES IN MEASURING extrapolate the likely benefits of treatment from studies done Q UA L I T Y F O R T H E E L D E R LY P O P U L AT I O N in a general population of healthy adults. In some cases, a lower amount of care might not represent poor quality when The elderly have different population health characteristics evidence or consensus is lacking on the efficacy of treatment and make more intensive use of health care compared to the or when patients and their physicians might decide that the general population—factors that call for special attention in potential risks of treatment outweigh the potential benefits in understanding health care quality and how to improve it for particular circumstances. Yet, one cannot assume that well- the elderly (Reuben et al. 2003a). Although the burden of informed decision-making is universally practiced and entirely disease generally increases and health status generally declines accountable for gaps in compliance to guidelines. with age, the elderly are not a homogenous group. Many The population data used for many charts do not permit elderly individuals remain in relatively good health; others one to determine the appropriateness of care or to report suffer various degrees of functional impairment or disability rates in accordance with evidence about the ages for which (Kane et al. 2004). This heterogeneity calls for individualized care is known to be effective. Even where data are reported approaches to health maintenance and treatment that can by age ranges, interpreting the appropriateness of a service challenge quality measurement to account for justifiable may be difficult without additional information. For example, variations in care. information on health status is needed to estimate life Two-thirds (65%) of the elderly have multiple chronic expectancy and the likely benefit of mammography among conditions (Wolff et al. 2002). Appropriate care for these women ages 75 and older (see narrative accompanying Chart individuals represents one of the more pressing challenges :2 for further discussion). Hence, this chartbook should be for clinical practice and quality measurement. Evidence- considered a preliminary approach at examining many topics. based standards of care captured in guidelines and quality More detailed research would be helpful to examine quality for measures (and in many of the charts shown in this chartbook) particular conditions in more depth. focus on single diseases as if these were treated in isolation. Yet, medications that are known to be effective to treat THE CHALLENGE AHEAD: specific diseases may be less beneficial in combinations that I M P R O V I N G Q UA L I T Y O F C A R E F O R T H E E L D E R LY increase risks for poor adherence, side effects, and drug-drug interactions. These concerns “raise the question of whether The elderly population is expected to double in size in the next what is good for the disease is always best for the patient” 25 years, from 35 million today to 7 million people ages 65 and (Tinetti et al. 2004). older by the year 2030 (CDC 2003c). The impending retirement The elderly—and especially the oldest old—are often in a of the baby boomer generation represents both a challenge for situation analogous to that of children in that physicians must the financing of Medicare and an opportunity to consider the most effective way to organize and deliver health care for the • systematic evidence reviews for conditions affecting elderly. Many experts have noted that the unique and growing Medicare beneficiaries; needs of the elderly population demand improved training • public reporting of performance information on health in the principles of geriatric medicine for the nation’s health plans, hospitals, kidney dialysis facilities, and home health professionals (Hudson 2003; LaMascus et al. 2005). agencies; and Many Medicare beneficiaries have limited incomes, • demonstration programs for chronic disease management, decreased mobility, low health literacy, and impaired ability cancer prevention and treatment among ethnic and racial to use the telephone (Williams 2004). The elderly are less minorities, information technology in doctors’ offices, and likely than younger adults to use the Internet. They often pay for performance. need help to manage complex medication regimens and navigate the health care system. These individual challenges These efforts must be rigorously evaluated to learn in combination with the fragmented structure of a health whether they are effective in meeting goals and how they care system that is focused on acute care services may create might be improved over time to best meet the needs of formidable barriers to accessing and realizing the benefits of Medicare beneficiaries. appropriate care for the elderly. Systemic improvement in quality of care for the elderly CO N C LU S I O N as for other populations calls for an integrated strategy. The essential elements of such a strategy include setting national As the country’s only national social health insurance priorities, defining targets for achievement, providing support program, Medicare offers a reasonable model for the future through investment in information technology, and paying of health care coverage in America. The evidence that the for and monitoring performance (Leatherman and Sutherland elderly are more likely to have their health care needs met and 2003). Although the examples in the final section of the experience fewer problems with their insurance and health chartbook provide inspiration that improvement is possible in care than nonelderly adults is a testimony to Medicare’s success particular settings, achieving any measurable impact on the in achieving its founding goals. As an increasing proportion of health care and health of the elderly requires wider adoption of the nation’s economy is devoted to health care in the coming these kinds of interventions. years, with an increasing proportion paid for by the federal The Medicare program has several promising changes government, understanding the factors behind Medicare’s under way or forthcoming that may help realize these success and building on them to strengthen the Medicare improvement goals. They include: program for the future is crucial. • a new prescription drug benefit and expanded coverage for certain preventive care services; Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 21 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 22 Methods The charts selected for this chartbook are intended to represent We did not directly compare clinical quality of care a coherent sample of the best available published data in terms between the Medicare fee-for-service program and Medicare of relevance to policy, generalizability to important segments managed care plans, other than to report the rate of influenza of the population, scientific soundness of measures, balance vaccination from the CAHPS (Consumer Assessment of in depicting various aspects of quality, and feasibility for Health Plans Study) survey. Definitions and/or data sources presentation in chart format. Our process was as follows: used by the Centers for Medicare and Medicaid Services We reviewed the general literature on quality of care (CMS) to measure clinical quality for Medicare fee-for-service for Medicare beneficiaries and solicited feedback of expert beneficiaries are not fully equivalent to the HEDIS measures advisors on the project definition and scope. and/or data sources used by the National Committee for We conducted a literature review using PubMed and other Quality Assurance to report on quality of care for managed searches to identify potential studies of interest, focusing on care plans. For example, CMS diabetes measures are based data published since our first chartbook in 2002. on administrative claims data while HEDIS uses a hybrid of From about 400 studies identified, we selected a subset that administrative data and medical records that produces a more we judged most relevant and feasible for presentation. Our accurate but higher rate than administrative data alone. Work expert consultants and advisors prioritized this list based on should be undertaken to define and report on comparable the criteria described above and we made other adjustments measures. In the meantime, CAHPS provides the fairest direct based on considerations of balance. comparisons between Medicare fee-for-service and Medicare The final group of charts and narrative was reviewed managed care based on beneficiary perceptions (see Chart 4:2). by the members of the Chartbook Advisory Board, project We generally discuss differences in rates only when they consultants, and staff at The Commonwealth Fund, who are statistically significant (i.e., 95 percent confidence or suggested several additions and improvements. greater that differences are not due to random chance), where We preferred studies using recent and nationally significance has been reported or can be inferred based on representative data. Other strong data were considered when large sample size. We use the term “significant” only in this no national data were available to depict an important topic. context. In other cases, we describe what we considered to be Because we were limited by the availability of published data to meaningful differences. Percentages and rates generally are depict quality for seniors, the topics included in the chartbook rounded to the nearest whole number, except where rounding should not be considered an ideal quality measurement set. would mask significant or potentially meaningful differences. See the Technical Appendix for details on study methodologies. Terminology Elderly, elders, and seniors are used interchangeably to refer to Several different populations are described in the chartbook: adults ages 65 and older. We recognize that some people prefer to use the term “older adults” for this population but we found that this Medicare beneficiaries include those living in the community or in term could be confusing to some readers. institutions such as nursing homes. Age-specific data were available for elderly Medicare beneficiaries only for selected conditions, such Vulnerable elderly are a subset of the elderly at greater risk for as hospital treatment of heart attack. When care for the elderly did declines in health. not differ substantially from nonelderly beneficiaries, such as for hospital treatment of pneumonia, we reported overall results. Near-elderly is used in the way that was intended by the research being cited. In some cases, this term refers to adults ages 55 to 64 Medicare fee-for-service beneficiaries include all those who and in other cases it means adults ages 60 to 64. have their health care bills paid by Medicare’s traditional (original) Medicare fee-for-service program. Middle-age generally refers to adults ages 45 to 64, an age category frequently used for reporting on national survey data, but it Community-dwelling adults are civilian, noninstitutionalized also refers to adults ages 50 to 64 in some contexts such as when individuals, including a small number who are not Medicare describing colorectal cancer screening. beneficiaries. We often used data from national surveys of community-dwelling adults (rather than Medicare-specific The terms health professional, clinician, and practitioner refer data sources) to compare services received for both elderly and to individuals including physicians, nurses and nurse practitioners, nonelderly adults. Most age-specific comparisons focus on middle- and physician’s assistants. We generally reserve the use of the term age adults (rather than younger adults) because their health care health care provider to encompass a broader category including needs are more like those of the elderly. both individual professionals and institutions such as hospitals. Medicare managed care plan members are beneficiaries who have Race and ethnicity are reported generally following the terminology (e.g., joined private plans (primarily health maintenance organizations) that black or African American) used in the original survey or article. For contract with the federal government to provide Medicare-covered this reason, the usage may appear inconsistent from chart to chart. services. These plans were called Medicare+Choice plans but are now called Medicare Advantage plans. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 23 SECTION 1 Effectiveness Effectiveness means “providing services based on scientific knowledge to all who could benefit and refraining from providing services to those not likely to benefit (avoiding underuse and overuse).” – Institute of Medicine 2001a 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 26 E F F E C T I V E N E S S • S TAY I N G H E A LT H Y • C H A R T 1 : 1 Immunization of Elderly Adults Why is this important? Influenza and pneumonia are the Implications: The nation remains far from the Healthy fifth leading cause of death among adults ages 65 and People 200 goal of 90 percent coverage for these older in the United States (NCHS 2004a). Within this vaccines, even in the best-performing state. Common age group, complications from influenza lead to 32,000 reasons that seniors give for not getting vaccinated deaths annually (Thompson et al. 2003) and severe include not knowing the vaccines are needed, fearing pneumococcal infections (bacteremia and meningitis) that the vaccine will cause infection or side effects, not account for an additional 3,400 deaths each year believing that the vaccine will be effective, and simply (Robinson et al. 200). The Centers for Disease Control forgetting about it (CDC 999, 2004b). A delay in vaccine and Prevention’s Advisory Committee on Immunization supply was a factor in lower vaccination rates during the Practices recommends that adults ages 65 years and older 2000–200 flu season. The vaccine shortage during 2004– receive an annual influenza vaccination and a single 2005 highlights the need for a national strategy to assure pneumococcal vaccination, which can prevent many adequate vaccine supply (GAO 2004b). hospitalizations and premature deaths (CDC 997, 2000; The most effective interventions for increasing adult Harper et al. 2004). Medicare has paid for pneumococcal vaccination and other preventive care services involve vaccination of Medicare beneficiaries since 98 and for organizational changes, such as offering prevention influenza vaccination since 993 (GAO 2002a). clinics and planned preventive care visits, engaging in quality-improvement activities, and designating Findings: From 989 to 2003, the proportion of community- nonphysician staff to perform prevention activities. dwelling Americans ages 65 and older who reported Education and reminders for health care providers and receiving an influenza vaccination in the past year more patients also can be effective (Stone et al. 2002). than doubled, from 3 percent to 66 percent, while the proportion who reported ever receiving a pneumococcal vaccination quadrupled, from 4 percent to 56 percent (NCHS 2004a). There has been little or no substantial increase in the past few years, however. Minnesota achieved adult vaccination rates of 80 percent for influenza and 73 percent for pneumococcal disease, the best performance for any state in 2003 (CDC 2004a). E F F E C T I V E N E S S • S TAY I N G H E A LT H Y • C H A R T 1 : 1 Immunization of Elderly Adults The rate of influenza vaccination doubled and the rate of pneumococcal vaccination quadrupled among the elderly from 1989 to 1999, but rates have not increased proportionally since then. One-third to one-half of elderly adults were not immunized as recommended in 2003. The higher rates achieved in states such as Minnesota demonstrate that substantial improvement is possible assuming adequate vaccine supply. ����������������������������������������������������������������������������������������������������� ������������������������� ������������� ��� �������������� ��������������������������������� ������� ����������������������������� ������������� �� �� ������������� ������������� ������������� �� �� �� �� �� �� �� �������������� �� ������������ �� �� �� �� �� �� �� �� ���� ������������ �� �� �� �� �������������� ������� �� �� �� ������������� �� �� �� �� ������������� �� ������������� ������������� � �������������� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� ������������ Sources: National rates—National Health Interview Survey (NCHS 2004a). State rates—Behavioral Risk Factor Surveillance System (CDC 2004a). National and state rates are not comparable because of differences in survey methods. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 27 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 28 E F F E C T I V E N E S S • S TAY I N G H E A LT H Y • C H A R T 1 : 2 Breast Cancer Screening Why is this important? Breast cancer is the most prevalent Implications: A lower rate of screening among older women non-skin cancer among U.S. women, with 22,000 new might be appropriate if it reflected patient preferences cases and 40,000 deaths annually (Jemal et al. 2004). Breast based on individualized assessment of potential benefits cancer victims lose 9 years of life on average (Brown et al. and harms. Yet, older women are less likely than younger 200). Mammography—a low-dose X-ray of the breast that women to actively participate with their physician in the can detect breast cancer at its earliest and most treatable decision to be screened (Burack et al. 2000). Moreover, stage—reduces the risk of death from breast cancer by mammography use decreases by age independent of 23 percent in women ages 50 and older (USPSTF 2002a). self-reported health status and disease burden (Blustein Although evidence is strongest for screening women ages and Weiss 998; Burack et al. 998). A recent survey in 50 to 69, one randomized controlled trial that included California found that women ages 80 to 85 in the best women ages 70 to 74 reported benefit for this age group health were less likely to be screened than women ages 75 (Humphrey et al. 2002). Women ages 75 and older are to 79 in the worst health, even though the former group likely to benefit from screening and early detection if their were more likely to benefit from mammography (Walter life expectancy is not compromised by comorbid illness. et al. 2004). These findings indicate that better tools are Medicare has paid for screening mammography since 99, needed to promote more informed decision-making subject to a 20 percent patient copayment for physician about breast cancer screening among older women services (GAO 2002a). (Walter and Covinsky 200). Findings: The proportion of community-dwelling women ages 65 and older who reported having a mammogram in the past two years tripled from 987 to 2000. The rate for women ages 65 to 74 increased from 27 percent to 74 percent, while the rate for those ages 75 and older increased from 7 percent to 6 percent. This trend was similar to that among women ages 50 to 64, who are screened at a higher rate (NCHS 2004a). In 2002, screening rates among women ages 65 and older varied from a high of 86 percent in Rhode Island to a low of 68 percent in Arkansas and Oklahoma (CDC/MIAH 2004). E F F E C T I V E N E S S • S TAY I N G H E A LT H Y • C H A R T 1 : 2 Breast Cancer Screening The proportion of elderly women who reported having a recent mammogram tripled over the past decade. Although evidence is strongest for screening women ages 50 to 69, screening is likely to be beneficial for older women with life expectancies of five years or longer. Screening rates for elderly women varied among the states by 17 percentage points from lowest to highest. ����������������������������������������������������������������������������������������������� ������������������������� ��������������������������� ��� �������������������������������� ���������������� ���������������� �� �� �� �������������� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� ���������������� �� ��������������������������� ������������� � �� �� �� �� �� �� �� �� �� �� �� �� �� �� ������������� ������������������������������������ Sources: Natonal rates — National Health Interview Survey (NCHS 2004a). State rates — Behavioral Risk Factor Surveillance System (CDC/MIAH 2004). National and state rates are not comparable because of differences in survey methods. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 29 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 30 E F F E C T I V E N E S S • S TAY I N G H E A LT H Y • C H A R T 1 : 3 Colorectal Cancer Screening Why is this important? Colorectal (colon or rectum) higher among the elderly than middle-age adults cancer is the second most frequent cause of cancer death, (Seeff et al. 2004). Among the states, the proportion of claiming 56,000 lives annually and reducing lifespan by 3 seniors who reported ever receiving a sigmoidoscopy or years on average (Ries et al. 2000; Jemal et al. 2004). The colonoscopy ranged from 47 percent in Nebraska to 75 U.S. Preventive Services Task Force and other medical percent in Minnesota in 2002 (CDC/MIAH 2004). societies recommend regular screening for colorectal cancer among adults ages 50 and older to detect polyps or Implications: Although colorectal cancer screening has cancers at an earlier and more treatable stage (Pignone et nearly doubled compared to rates reported for 992 al. 2002; USPSTF 2002b; Winawer et al. 2003). Screening (Nadel et al. 2002), it remains widely underused. For options include the following: example, men are more likely to be screened for prostate • fecal occult blood test (done at home to detect blood in cancer than colorectal cancer, despite the proven benefit the stool) every year, and/or sigmoidoscopy (in which the of colorectal cancer screening and the uncertain benefit of doctor inserts a flexible, lighted tube to visually inspect prostate cancer screening (Sirovich et al. 2003). The most the rectum and lower large intestine) every five years, or common reasons cited for not receiving colorectal cancer • total colon examination by colonoscopy (in which the screening indicated lack of awareness (“didn’t think of doctor inserts a flexible, lighted tube to visually inspect it,” cited by one-half of those not screened) and lack of a the rectum and entire large intestine) every 0 years, or physician recommendation (cited by about one-quarter) by double-contrast barium enema (which is an x-ray (Seeff et al. 2004). Medicare began paying for these examination of the rectum and entire large intestine) tests for screening purposes in 998; sigmoidoscopy and every five years. colonoscopy are subject to a copayment and the Medicare Part B deductible (GAO 2002a). Findings: In 2000, only one-half of community-dwelling adults ages 65 and older reported performing a blood stool test at home in the past year or receiving a colorectal endoscopy test (including sigmoidoscopy or colonoscopy) in the past 0 years. Results were similar for endoscopy testing in the past five years (the survey did not ask about double-contrast barium enema). Screening was somewhat E F F E C T I V E N E S S • S TAY I N G H E A LT H Y • C H A R T 1 : 3 Colorectal Cancer Screening Elderly adults are more likely than middle-age adults to receive colorectal cancer screening tests, but one-half had not been screened as recommended in 2000. The proportion of seniors who had ever received sigmoidoscopy or colonoscopy varied among the states by 28 percentage points from lowest to highest in 2002. ��������������������������������� ������������������������������������ ������������������������������������������ ������������������������������������������� ��� ��������������������������� ������������������������������������������ ��������������������������������������� ���������� �������� �� �� �� �� �� �� �� �� �� �� �� � ���������������� ���������� �������� ����������� ���������������������������� �������������� ������������� ���������������� ������������� ������������� �������������� ��������������� ������������� ������������ ����������������� ������������ �������������� �������������������������� Source: National rates — National Health Interview Survey (Seeff et al. 2004). State rates — Behavioral Risk Factor Surveillance System (CDC/MIAH 2004). National and state rates are not comparable because of differences in measures and survey methods. *The national survey asked about most recent receipt of proctoscopy, sigmoidoscopy, or colonoscopy; the recommended time interval for colonoscopy is used but results were similar for a five -year interval. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 31 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 32 E F F E C T I V E N E S S • S TAY I N G H E A LT H Y • C H A R T 1 : 4 ( E M E R G I N G I S S U E ) Osteoporosis Counseling and Screening Why is this important? An estimated 0 million Americans Findings: One-half of elderly female Medicare beneficiaries have osteoporosis (“porous bone”), four of five of them (ages 65 and older) living in the community had not women, and another 34 million have low bone mass that talked to their physician about osteoporosis in 2000. puts them at risk of developing this skeletal disorder. About one-third had received a bone density test for Osteoporosis leads to bone fragility and an estimated .5 osteoporosis, but one-quarter had never heard of the test million fractures each year. Risk of osteoporosis increases (Adler and Shatto 2002). with age. Among those ages 50 and older, half of women and one-quarter of men will have an osteoporosis- Implications: Because osteoporosis was an emerging issue at related fracture during their lifetime, including vertebral the time of this survey, these results should be considered deformities that can lead to chronic pain and hip fractures a baseline for future improvement. Several studies have that increase the risk for nursing home admission and found that older men and women often are not screened death (DHHS 2004; NIH 2004; NOF 2004). or treated for osteoporosis after suffering a fracture, Physician counseling on osteoporosis should which represents a missed opportunity to prevent emphasize preventive measures for bone loss, including future fractures among those likely to be at high risk for weight-bearing exercise, adequate dietary intake of osteoporosis (Kamel et al. 2000; Andrade et al. 2003; calcium and Vitamin D, strategies to prevent falls, Solomon et al. 2003). One community increased rates and avoidance of tobacco and excessive alcohol use of bone density testing and osteoporosis treatment by (NOF 999, 2003). Osteoporosis screening became a educating patients about osteoporosis when they visited Medicare-covered benefit in 998, when the National hospital emergency departments (EDs) for wrist fractures Osteoporosis Foundation and other medical societies and by having the EDs fax a guideline-based reminder to recommended that all women ages 65 and older, and the patients’ primary care physicians to encourage follow- younger post-menopausal women with a fracture or risk up care (Majumdar et al. 2004). factors, have bone density measurement for osteoporosis. The U.S. Preventive Services Task Force made a similar recommendation in 2002. Those diagnosed with osteoporosis can be treated with medications that improve bone density and reduce the risk for fracture (USPSTF 2002d). E F F E C T I V E N E S S • S TAY I N G H E A LT H Y • C H A R T 1 : 4 ( E M E R G I N G I S S U E ) Osteoporosis Counseling and Screening Osteoporosis screening became a Medicare-covered benefit in 1998, when the National Osteoporosis Foundation first recommended it for elderly women. In 2000, one-half of elderly female Medicare beneficiaries (ages 65 and older) said that they had ever talked to their doctor about osteoporosis, and one-third had ever had a bone density test to check for osteoporosis. ���������������������������������������������������������� �������������������������������������������������� ��������������������� ����������� ����������������������� ���������������� ��������� ���������� ������������������ ������������������� ���������������������������� ������������������������� ��������� ����������������� ������������� ������������������ ������������ Source: 2000 Medicare Current Beneficiary Survey (Adler and Shatto 2002). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 33 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 34 E F F E C T I V E N E S S • S TAY I N G H E A LT H Y • C H A R T 1 : 5 Falls and Instability: Screening and Management Why is this important? Older adults are more prone to those who had fallen, just more than half were asked falls and mobility disorders than younger adults due about their fall history and less than a quarter had a to muscle weakness, gait and balance problems, visual basic fall examination. Although exercise programs and/or cognitive impairment, and medication side effects were offered to 69 percent of those identified as having (Rubenstein et al. 2004). About one-third of community- mobility problems, only 3 percent of patients with dwelling elderly fall at least once per year. Fall-related decreased balance were offered both an appropriate fractures and injuries among the elderly are the cause of exercise program and an evaluation for an assistive device .6 million visits to hospital emergency departments and (Wenger et al. 2005). 388,000 hospital admissions each year (NCIPC 2005). Falls often result in functional decline, disability, and Implications: Although this study was limited to a small fear of falling, leading to loss of independence and many population, it suggests that many falls and mobility nursing home admissions (Tinetti and Williams 997, disorders likely go undetected in the elderly and that 998; Bezon et al. 999). many opportunities for prevention are not being realized. Several intervention strategies are effective for The cost of fall-related injuries is substantial (Englander reducing falls and instability including risk factor et al. 996); Medicare spent 5.5 billion for treatment of assessment and targeted exercise programs (Chang et al. fractures among the elderly in 999 (Bishop et al. 2002). 2004). Evidence-based guidelines for the prevention and There is some evidence that interventions to prevent falls management of falls state that clinicians should regularly can be cost-saving (Rizzo et al. 996), suggesting that ask patients about falls and instability and use diagnostic Medicare reimbursement for fall prevention programs tests to identify causes and contributing factors, many might be cost-effective. Primary care physician education of which will respond to intervention (AGS/BGS/AAOS may be warranted to help elders avoid falls and resulting Panel on Falls Prevention 200). disability. * At-risk patients included those who screened positive for falls or fear of falling, Findings: Among at-risk* patients ages 75 and older bothersome incontinence, or memory impairment. treated in two medical groups participating in the Assessing Care of Vulnerable Elders (ACOVE-2) study during 2000–200, only 40 percent were asked at least annually about the occurrence of recent falls. Among E F F E C T I V E N E S S • S TAY I N G H E A LT H Y • C H A R T 1 : 5 Falls and Instability: Screening and Management Falls and mobility disorders are common in the elderly and often lead to functional decline and loss of independence. A pilot study found that older patients at risk for these conditions often did not have an adequate examination or an evaluation that led to diagnostic and treatment recommendations. Therapy was not always offered even when problems were diagnosed. ��������������������������������������������������������������������� ������������������������������������������������������������ � �� �� �� �� ��� �������������������������������������������������������� �� ����������������������������� �� ���������������������������������� �� ��������������������������������������������������������� �� �������������������������������������� � �� ������������������������������������������������������ �� ������������������������������������������������ � Source: Medical records from the Assessing Care of Vulnerable Elders (ACOVE-2) study ( Wenger et al. 2005). *At-risk patients screened positive for falls or fear of falling, bothersome incontinence, or memory impairment (N=644). **Among those with two or more falls in the past year, or a single fall with injury requiring treatment. ***Within three months among those who report or are found to have new or worsening difficulty with ambulation, balance, and/or mobility. †Among those with gait, strength, or endurance problems. ‡Among those with decreased balance and/or proprioception or increased postural sway. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 35 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 36 EFFEC TIVENESS • GET TING BET TER • CHART 1:6 Inappropriate Use of Antibiotics for the Common Cold Why is this important? Widespread over-prescribing of Implications: The improvement in this population-based antibiotics contributes to the emergence of antibiotic- rate may reflect two factors: ) patients may have resistant strains of bacteria (Lewis 995), which have been less often visited physicians with the cold, and/or 2) increasing in prevalence (Whitney et al. 2000). Antibiotic physicians may have less often prescribed antibiotics resistance threatens the effectiveness of the antibiotic when they did visit. This downward trend may reflect the arsenal for all patients. Moreover, antibiotic use puts an effects of an intensive educational campaign undertaken individual at risk for subsequent infection with antibiotic- by the Centers for Disease Control and Prevention in resistant bacteria (Dowell and Schwartz 997). Therefore, concert with state and local public health departments public health experts and medical societies recommend and medical societies (CDC 2005). Research in Finland careful antibiotic use for patients who are most likely to suggests that reducing the use of antibiotics can lead to a benefit (Gonzales et al. 200). The common cold is caused decrease in the prevalence of antibiotic-resistant bacteria by a virus, against which antibiotics are not effective and in the community (Seppala et al. 997). never indicated. A multifaceted educational intervention for patients and physicians, combined with performance feedback Findings: The population-based rate of antibiotic for physicians, safely reduced inappropriate antibiotic prescribing at visits to physician offices and hospital prescribing among non-elderly adults in one health outpatient clinics and emergency departments for plan (Gonzales et al. 999). Yet, no measurable effect patients diagnosed with the common cold decreased by was observed when the educational intervention was 44 percent among the elderly and by 33 percent among extended to include elderly patients. The authors of middle-age adults from 997–998 to 2000–200.* The the study speculated that “factors other than patient elderly were 29 percent more likely than middle-age expectations and demands may play a stronger role in adults to receive antibiotics for the common cold in antibiotic treatment decisions in elderly populations” 2000–200, as compared to 50 percent more likely in (Gonzales et al. 2004). 997–998 (AHRQ 2005b). * The numbers shown in the chart represent the rate of antibiotic prescribing within the entire civilian, noninstitutionalized population age-groups specified (e.g., 226 antibiotics prescribed per 10,000 community-dwelling elderly in 1997–1998 vs. 126 per 10,000 in 2000–2001). EFFEC TIVENESS • GET TING BET TER • CHART 1:6 Inappropriate Use of Antibiotics for the Common Cold Antibiotics are never appropriate treatment for the common cold. Elderly patients are more likely than middle-age adults to receive antibiotics for a cold, whether because they more often visit physicians with a cold or because physicians are more likely to prescribe antibiotics when they do visit. The inappropriate use of antibiotics decreased among both age groups from 1997–1998 to 2000–2001, with a 44 percent decline among elderly patients. ��������������������������������������������������������������������������������������� ��������������������������������������������������������������������������������������� ��� ��� ��� �������� ���������� ��� ��� ��� ��� �� �� � ��������� ��������� Source: National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey (AHRQ 2005b). *These data represent the rate of antibiotic prescribing within the entire civilian, noninstitutionalized population age -groups shown. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 37 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 38 EFFEC TIVENESS AND TIMELINESS • GET TING BET TER • CHART 1:7 Hospital Treatment for Pneumonia Why is this important? About 600,000 Medicare administered within four hours of hospital arrival, and beneficiaries are hospitalized with pneumonia each 68 percent were given an initial antibiotic consistent with year (CMS 2000b). Previous research found that giving current guideline recommendations. Only 30 percent antibiotics to pneumonia patients within eight hours of of these patients received care consistent with all three hospital arrival resulted in a lower death rate (Meehan et standards. Among the states, rates of timely antibiotic al. 997). In 2000–200, hospitals achieved this goal for administration varied by 3 percentage points from a low 85 percent of Medicare fee-for-service beneficiaries with of 46 percent in Delaware to a high of 77 percent in South pneumonia. Newer research has shown that antibiotic Dakota (AHRQ 2005b). administration within four hours of hospital arrival is associated with further reductions in death rates and Implications: Evolving standards present a challenge for hospital length of stay (Houck et al. 2004). This four- hospitals to continually improve quality of care. The rate hour standard was adopted for the Medicare Quality of appropriate antibiotic selection decreased in 2002 Improvement Organization program starting in 2002. from 84 percent in 2000–200, probably because of a “lag The American Thoracic Society and the Infectious time” for physicians to become aware of updated scientific Disease Society of America recommend that a blood guidelines for preferred antibiotic treatment (personal culture be drawn before antibiotics are administered communication with Edwin Huff 2005). so that treatment can be tailored to the specific form of Hospitals in one state were more likely to improve infection (Bartlett et al. 2000; Niederman et al. 200). evidence-based pneumonia treatment if they used Timely collection of blood cultures (within 24 hours a combination of quality improvement strategies of hospital arrival) and use of recommended antibiotic including clinical pathways, standing orders, physician combinations is associated with lower death rates champions, multidisciplinary teams, and case managers (Meehan et al. 997; Gleason et al. 999). (Tu et al. 2004). Many hospitalizations for pneumonia might be prevented altogether if more older adults were Findings: During 2002, 8 percent of Medicare fee-for- immunized as recommended and treated appropriately service beneficiaries hospitalized with pneumonia when they seek care in the outpatient setting (see Charts had a blood culture collected before an antibiotic : and :). was administered (among those for whom cultures were collected at all), 63 percent had an antibiotic EFFEC TIVENESS AND TIMELINESS • GET TING BET TER • CHART 1:7 Hospital Treatment for Pneumonia Evidence-based treatment of pneumonia is associated with reduced risk of death. Hospital treatment of Medicare fee-for-service beneficiaries with pneumonia met three guideline standards less than one-third of the time in 2002. Among the states, rates of timely antibiotic administration varied by 31 percentage points from lowest to highest. ������������������������������������������������������������������������������� ��������������������������������� ��������������������������������������������������������� ������������������������������������� ���������������������������������������������������� ������������������������ �� ������������������������� �� ������������������������� �� ������������������������� ������������������������� �� ��������������� ���������������� ��������������������������� ������������� ������������������������ �� ������������� �������������������������� Source: Centers for Medicare and Medicaid Services, review of medical records (AHRQ 2005b). *Among those for whom a blood culture was ordered. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 39 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 40 EFFEC TIVENESS • GET TING BET TER • CHART 1:8 Hospital Treatment for Heart Attack Why is this important? Coronary heart disease among from the analysis). Rates of treatment for those ages 85 Medicare beneficiaries manifests in more than 400,000 and older were 4 to 0 percentage points lower than for heart attacks requiring hospitalization each year. those ages 65 to 74. Rates of beta-blocker prescribing Evidence-based treatment guidelines recommend that increased from 2000-200 to 2002 (see Appendix Table heart attack patients receive certain medications early b), with the greatest increases occurring among those during hospitalization and/or afterwards as long-term ages 85 and older (not shown) (AHRQ 2005b). preventive therapy to reduce the risk of a recurrent heart attack and improve the likelihood of survival (CMS 2003; Implications: Performance on these measures has improved Antman et al. 2004). substantially from 994–995, when beta-blockers were • Aspirin helps prevent the blood from clotting. Early use given to only about half of Medicare fee-for-service of aspirin for heart attack victims reduces short-term beneficiaries hospitalized for heart attack and aspirin mortality by 23 percent. Long-term aspirin use after a was given to about three-quarters (Burwen et al. 2003). heart attack lowers mortality by 3 percent. Further improvement could save many more lives. The • Beta-blockers ease the heart’s pumping and reduce its need lower rate of treatment among the oldest elderly might for blood and oxygen. Early beta-blocker administration reflect misconceptions about the benefits of treatment in improves survival by 4 to 5 percent. Long-term use after a the elderly; in some cases, however, treatment might not heart attack improves survival by 23 percent. have been appropriate for very frail individuals. More • ACE (angiotensin-converting enzyme) inhibitors data are needed to understand patterns of care and guide increase the supply of blood and oxygen to the heart. treatment for the oldest elderly. Chart 6:2 illustrates Long-term use after a heart attack among patients with an intervention that improved heart attack treatment impaired left ventricle function reduces their mortality up for Medicare patients, with the greatest improvements to 27 percent. observed among the oldest elderly. Findings: From 63 percent to 9 percent of elderly Medicare Note: Rates of ACE inhibitor prescription at hospital discharge do not account for the substitution of newer medications called angiotensin receptor blockers, which fee-for-service beneficiaries hospitalized for heart attack may add up to 10 percentage points to the rate and will be counted for compliance in future years (personal communication with Edwin Huff 2005). in 2002 received or were prescribed medications when indicated, depending on the drug and the patient’s age (those with documented contraindications were excluded EFFEC TIVENESS • GET TING BET TER • CHART 1:8 Hospital Treatment for Heart Attack In 2002, 63 percent to 91 percent of elderly Medicare patients hospitalized for heart attack received or were prescribed recommended medications to prevent a second heart attack, depending on the drug and the patient’s age. Rates of treatment for those ages 85 and older were 4 to 10 percentage points lower than for those ages 65 to 74. ����������������������������������������������������������������� ������������������������������������������������������������������ ��� ���������� �� ���������� �� �� �� �������� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � ������� ������������ ������� ������������ ��������������� ����������������������������� �������������������������������� ��������������������� Source: Centers for Medicare and Medicaid Services, review of medical records (AHRQ 2005b). *Includes only patients with no documented contraindications to the medication. **Among those with left ventricular systolic dysfunction. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 41 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 42 EFFEC TIVENESS AND TIMELINESS • GET TING BET TER • CHART 1:9 Time to Reperfusion for Heart Attack Why is this important? Timely reperfusion improves Findings: Among Medicare fee-for-service beneficiaries survival among a subset of heart attack patients by hospitalized for heart attack during 2000–200 who were increasing blood supply to the heart muscle.* Depending eligible for and received reperfusion, the median time to on the patient’s risk factors and contraindications and start reperfusion (measured from hospital arrival) varied the hospital’s capabilities, reperfusion may be done with widely across states.* The median time to initiate PTCA clot-dissolving drugs (thrombolysis) or a minimally within a state ranged from 59 to 260 minutes and was 07 invasive surgical procedure (percutaneous transluminal minutes in the median state. Two-thirds of the states met coronary angioplasty or PTCA). Timely PTCA can be the door-to-balloon time goal of 20 minutes. The median more effective than thrombolysis when done in a capable time to initiate thrombolysis within a state ranged from facility, but many hospitals are not equipped for this 28 to 2 minutes and was 45 minutes in the median state. procedure. Expert guidelines provide criteria for selecting Only four states met the door-to-needle time goal of 30 the type of reperfusion and for transferring patients to minutes (AHRQ 2005b). capable facilities, but “appropriate and timely use of some reperfusion therapy is likely more important than the Implications: The guideline writers emphasize that choice of therapy” (Antman et al. 2004). timeliness goals “should not be perceived as an average The sooner reperfusion is started, the greater the performance standard but a goal of an early treatment benefit it confers (the benefits and risks may differ system that every hospital should seek for every for patients over the age of 75). Current guidelines appropriate patient...Systems that are able to achieve even recommend that thrombolysis be started in eligible more rapid times for patients should be encouraged” patients within 30 minutes of hospital arrival (door-to- (Antman et al. 2004). This implies an opportunity for needle time) and that PTCA should commence within improvement even in states where the median time met 90 minutes (door-to-balloon time). However, the the target (since half of the patients had longer times). door-to-balloon time target was 20 minutes during the Data from the National Registry of Myocardial Infarction time period shown in the chart; the Medicare program indicate that the elderly are more likely to experience uses this goal for quality evaluation purposes (personal delays in reperfusion compared to non-elderly heart communication with Edwin Huff 2005). attack victims (Angeja et al. 2002). * Patients eligible for reperfusion include those with ST-elevation myocardial infarction or left bundle branch block. EFFEC TIVENESS AND TIMELINESS • GET TING BET TER • CHART 1:9 Time to Reperfusion for Heart Attack Median time to reperfusion ��� for Medicare fee-for-service ��� beneficiaries in 2000–2001, by state* Reperfusion restores blood flow to the ��� heart after a heart attack, either through ��� a surgical procedure (angioplasty or PTCA) or clot-dissolving drugs ��� (thrombolysis). The faster reperfusion ��� is started, the greater the benefit. The ������������� median time from hospital arrival ��� ��������������������� �������� to start reperfusion varied widely �������������������������� ��� across the states. The median time to ������������������������� initiate PTCA met the national goal ��� of 120 minutes in two-thirds of the ������������ states. The median time to initiate �� ������� thrombolysis met the national goal �� of 30 minutes in only four states. �� ��������������������������� �� �������������������������� ������������������������ � � � � � � �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� ������������������������������������������������������������ �������������������������������� Source: Centers for Medicare and Medicaid Services, review of medical records (Jencks et al. 2003). States include Puerto Rico and the District of Columbia (DC). Data were not available for thrombolysis in DC and for PTCA in Maine. *Includes only patients with ST-elevation myocardial infarction or left bundle branch block who received reper fusion. **PTCA = percutaneous transluminal coronary angioplasty. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 43 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 44 EFFEC TIVENESS • GET TING BET TER • CHART 1:10 Hospital Mortality Why is this important? The Agency for Healthcare Research Implications: The improvement in hospital death rates and Quality has identified several conditions and surgical for these conditions and procedures is encouraging procedures for which there is evidence that hospital and suggests that patients are receiving higher quality mortality may be associated with the quality of care treatment in a timely manner while in the hospital, provided in the hospital (AHRQ 2002a). For example, although improvements in diagnostic and treatment timely and evidence-based hospital treatment of heart modalities also may play a role. The recent increase attack increases the likelihood of patient survival (see in 30-day mortality rates, however, provides reason Charts :8 and :9). Considering mortality measured 30 for concern. This trend might indicate a variety of days after hospitalization in conjunction with hospital problems, such as inadequate quality in skilled nursing mortality provides a more accurate picture of deaths that facilities or rehabilitation facilities to which hospital may be attributable to inpatient health care but that occur patients are discharged, or that hospitals are discharging soon after discharge from the hospital (MedPAC 2004c). patients without adequately educating patients and their caretakers on appropriate self-care, or that patients have Findings: Risk-adjusted* rates of in-hospital mortality inadequate follow-up care and support once they leave steadily declined from 995 to 2002 among Medicare fee- the hospital. Monitoring this trend and identifying its for-service beneficiaries hospitalized for eight conditions underlying causes will be critical for assuring effective or procedures studied during this seven-year period. The patient care. largest absolute reduction in mortality occurred for those * Mortality rates were risk-adjusted to account for differences in patients’ age, sex, being treated for a heart attack (36 deaths per 0,000 and severity of illness over time; however, some clinical risk factors may not be fully discharges). The rates of death 30 days after hospital accounted for using these methods and administrative data. admission also decreased for these eight conditions or procedures from 995 to 2000, with heart attack again showing the largest absolute reduction (272 deaths per 0,000 discharges) during this five-year period. However, 30-day mortality rates increased from 2000 to 2002 for six of the eight conditions or procedures, with mortality for stroke and pneumonia exhibiting the greatest absolute increases during this two-year period (87 and 80 deaths per 0,000 discharges) (MedPAC 2004c). EFFEC TIVENESS • GET TING BET TER • CHART 1:10 Hospital Mortality In-hospital death rates decreased between 1995 and 2002 among fee-for-service Medicare beneficiaries treated for eight conditions for which outcomes are related to the quality of hospital care. Rates of death within 30 days of hospital admission for these conditions decreased from 1995 to 2000 but increased for six of the conditions between 2000 and 2002. ��������������������������������������������������� ���������������������������������������������� ������������������������������������������������ ������������������������������������������������ ����� ����� ����� ����� ����� ����� ����� ����� ����� ����� ����� ����� ��� ��� ��� ��� ��� ��� ��� ��� � � � ����� ����� ����� ���� � ����� ����� ����� ���� ������������� ������ ������ ������ ����� ������������� ������ ������ ������ ����� ������� ������ ������ ������ ����� ������� ������ ������ ������ ����� ����������� ������ ������ ������ ����� ���������� ������ ������ ������ ����� ���������� ������ ������ ������ ��� ����������� ������ ������ ������ ����� ����������� ������ ���� ���� ��� ����������� ������ ������ ������ ����� �������������� ���� ���� ���� ��� �������������� ������ ������ ���� ��� ����� ���� ���� ���� ��� �������������� ���� ���� ���� ��� �������������� ���� ���� ���� ��� ����� ���� ���� ���� ��� Source: Medicare Payment Advisory Commission (2004c) analysis of Medicare administrative data using AHRQ Inpatient Quality Indicators, risk-adjusted for age, sex, and severity of illness. AAA=abdominal aortic aneurysm; CABG=coronary artery bypass graft surgery; GI=gastrointestinal. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 45 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 46 EFFEC TIVENESS AND TIMELINESS • GET TING BET TER/LIVING WITH ILLNESS • CHART 1:11 Hospitalizations for Ambulatory Care Sensitive Conditions Why is this important? Elderly Americans are more likely Implications: Reducing preventable hospitalizations could than any other age group to be hospitalized for conditions help to preserve Medicare funds for needed services while for which good ambulatory care is important, including concurrently improving patient health. Assuming that congestive heart failure, chronic obstructive pulmonary an average hospital stay costs 5,300 per admission, a 5 disease, and diabetes complications (Kruzikas et al. 2004). percent decrease in the 2,388,000 Medicare admissions Effective diagnosis, treatment, and patient education for these 2 conditions in 2000 would translate to 633 can help control the exacerbation of an illness and million in cost savings (see Technical Appendix). Two prevent or delay complications of chronic illness, thus examples of possible strategies for doing so include: ) reducing hospitalizations (Niefeld et al. 2003). Although promoting increased immunization among seniors to hospitalization rates are influenced by socioeconomic reduce admissions for pneumonia (see Chart 6:), and factors and patient behaviors, high rates of potentially 2) increasing the use of care coordination to reduce preventable hospitalizations might indicate suboptimal rehospitalizations among patients with congestive heart prevention, inadequate primary care, or barriers to failure (see Chart 6:3). obtaining timely and effective ambulatory care (Bindman Focusing attention on Medicare beneficiaries with et al. 995; AHRQ 2002b). multiple chronic conditions might yield the greatest benefits since the likelihood of being hospitalized for Findings: Among Medicare fee-for-service beneficiaries, an ambulatory care sensitive condition increases in rates of hospital admissions (age- and sex-adjusted) proportion with the number of chronic conditions that an increased from 995 to 2002 for seven of 2 ambulatory individual suffers (Wolff et al. 2002). Facilitating access care sensitive conditions studied (only the top 0 are to primary care in underserved geographic areas might shown on the chart). The rate of hospitalization due reduce the higher rates of preventable hospitalizations to bacterial pneumonia exhibited the largest absolute among vulnerable Medicare beneficiaries in those areas increase (38 per 0,000), while the rate for angina (chest (Parchman and Culler 999; Epstein 200). pain) without the performance of a cardiac procedure * The decrease in the rate of admissions for angina without procedure would not decreased by a similar amount (36 per 0,000).* In a indicate an improvement in ambulatory quality of care to the degree that it was similar example, the hospitalization rate for uncontrolled offset by any increase in admissions for angina with procedure. diabetes decreased by 4 per 0,000, while the rate for long-term complications of diabetes increased by 6 per 0,000 (MedPAC 2004c). EFFEC TIVENESS AND TIMELINESS • GET TING BET TER/LIVING WITH ILLNESS • CHART 1:11 Hospitalizations for Ambulatory Care Sensitive Conditions Some hospitalizations might be preventable when patients receive timely and appropriate ambulatory care (for an example, see Chart 6:3). Rates of hospitalization increased from 1995 to 2002 for five of the 10 conditions shown. ������������������������������������������������������������������������������������������������������������� � �� ��� ��� ��� ��� ��� ��� ������������������������ ��� ��� ������������������� ��� ��� ������������������������������������� ��� �� ����������������� �� �� ����������� �� ���� �� ������������������������������������� �� ���� �� ������������������������������� �� �� ������������ �� �� ���������������������������� �� �� ��������������������� � Source: Medicare Payment Advisory Commission (2004c) analysis of Medicare administrative data using AHRQ Prevention Quality Indicators (only 10 highest rates shown). *Rates are age - and sex-adjusted. **Among those with diabetes. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 47 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 48 EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:12 High Blood Pressure Awareness and Control Why is this important? Hypertension, or high blood was normal or high. Despite this high level of awareness, pressure, is a major risk factor for heart disease and stroke another national study found that many of those whom in the elderly and also can lead to kidney disease and researchers determined had high blood pressure (based vision problems when left untreated (Kilker et al. 2000). on objective measurement) did not have their condition The prevalence of this chronic condition increases with under control. Specifically, control of high blood pressure age. Roughly half of all elderly Medicare beneficiaries, improved from 33 percent to 40 percent of middle-age and two-thirds of elderly black adults, have high blood adults (ages 45 to 64) during the 990s, but remained pressure (FIFARS 2004). Those with hypertension are unchanged at 24 percent of elderly adults (ages 65 and generally less healthy and use more health care services older) (AHRQ 2005b). than those without this chronic condition. High blood pressure can be controlled with lifestyle Implications: The elderly population is lagging far behind modifications and/or medication. Many elderly have the national Healthy People 200 goal that high blood systolic hypertension, in which systolic pressure pressure will be controlled for at least half of Americans (the first number) is high but diastolic pressure (the with the condition (DHHS 2002a). Another analysis of second number) is low. In the past, this condition the same survey shown in the chart found that middle- was considered a normal part of aging (AMA 2003), age and older adults with high blood pressure were but research has shown that treatment reduces the equally likely to receive treatment during 999–2000 (63 incidence of stroke and cardiovascular disease in the percent of each age group), but that older adults were elderly (Chaudhry et al. 2004). Because evidence for less likely to achieve blood pressure control when treated treatment is less strong for the oldest patients, expert (44 percent of those ages 60 and older vs. 66 percent of guidelines emphasize the need to assess the potential those ages 40–59) (Hajjar and Kotchen 2003). A review of benefits and risks of treatment for elderly patients on an quality-improvement studies found that many strategies individualized basis (Chobanian et al. 2003). were effective for improving hypertension care and outcomes (Walsh et al. 2005). Even small improvements Findings: In a national survey conducted in 998, most in individual blood pressure control can have large health elderly and middle-age adults—92 percent—reported that effects when considered on a population-wide basis. their blood pressure had been measured within the past two years and could state whether their blood pressure EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:12 High Blood Pressure Awareness and Control Most adults say they know whether they have high blood pressure. Only one-quarter of elderly adults with high blood pressure (as determined by researchers) had it controlled during 1999–2000, which was lower than the rate among middle-age adults. The improvement in blood pressure control among middle-age adults was not matched by any improvement among elderly adults during the 1990s. �������������������������������������������� ��������������������������������������������� ������������������������������������������ ������������������������������������������������������ �������������������������������������������� ������������������������������������������� ���������������������������������� ��� ��� �� �� ��������� ��������� �� �� �� �� �� �� �� �� �� �� �� �� � � ���������� �������� ���������� �������� Sources: National Health Interview Survey (AHRQ 2003) and National Health and Nutrition Examination Survey (AHRQ 2005b). Blood pressure control defined as 140/90 mmHg or lower. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 49 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 50 EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:13 High Cholesterol Screening, Awareness, Treatment, and Control Why is this important? Coronary heart disease (CHD), Findings: A nationally representative study conducted in which is characterized by narrowing and blockage of 999–2000 found that the majority of elderly U.S. adults arteries that supply blood to the heart, is the number one (ages 65 and older) with high cholesterol (as determined cause of death among elderly Americans (AHA 2005b). by researchers) had their cholesterol checked in the past; The incidence of CHD and its manifestations, such as however, barely more than one-half knew that it was heart attacks, increase with age. Prevention of disease high. Among all elderly with high cholesterol, 30 percent “offers the greatest opportunity for reducing the burden were taking cholesterol-lowering medication but only 8 of CHD in the United States” (Grundy et al. 2004). Two percent had their high cholesterol controlled. The elderly major modifiable risk factors for CHD are hypertension were somewhat more likely than middle-age adults with (see Chart :2) and high cholesterol. high cholesterol to have been tested, exhibit awareness, National guidelines recommend that adults be take medications, and have their cholesterol controlled screened for high cholesterol and supported in making (Ford et al. 2003). lifestyle changes (e.g., diet, exercise, weight control) to reduce their risk for CHD, including high cholesterol Implications: The higher rates of cholesterol testing, (Pearson et al. 2002). For elderly adults at highest risk awareness, treatment, and control in elderly Americans of CHD, or in whom lifestyle change is not successful, is promising, but the control of high cholesterol in this cholesterol-lowering therapy should be considered population is still extremely low. The study did not based on individualized assessment of efficacy, safety, report the proportion of elderly seeking to make lifestyle tolerability, and patient preference. Although evidence changes such as a healthy diet and regular exercise. The is strongest for treating high cholesterol in the elderly American Heart Association guideline authors note that with known heart disease, treatment is also likely to be a “physician-patient partnership must be forged, on the effective in the elderly at risk for developing heart disease physician’s part by assessing and communicating risk and (NCEP 200; Grundy et al. 2004). Current treatment by co-developing with the patient a plan of preventive strategies focus on reducing high levels of “bad” action” (Pearson et al. 2002). cholesterol (low-density lipoprotein or LDL). EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:13 High Cholesterol Screening, Awareness, Treatment, and Control Most elderly adults whom researchers determined had high cholesterol reported that they had a cholesterol test in the past, but little more than half said they knew they had high cholesterol, less than one-third were using cholesterol-lowering medications, and few had achieved control over their high cholesterol. Rates were somewhat higher for elderly than middle-age adults. ��������������������������������������������������������������������������� �������������������������������������������������������������������������������������� ��� �� ���������� �������� �� �� �� �� �� �� �� �� �� �� � � ��������������� ������������������������ ����������������������� ��������������������� ������������������� ������������������������ ������������������� �������������������� �������������������� ���������������������� Source: National Health and Nutrition Examination Survey (Ford et al. 2003). Participants were classified as having high cholesterol if they reported using cholesterol-lowering medications or if a blood test showed total cholesterol of 200 mg/dL or higher. Cholesterol was classified as controlled if a blood test showed total cholesterol was under 200 mg/dL. All results are reported for the entire sample and are weighted to be nationally representative. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 51 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 52 EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:14 Cardiovascular Care and Outcomes in Managed Care Plans Why is this important? About 5 million Medicare with high cholesterol controlled (an intermediate beneficiaries are enrolled in managed care plans outcome measure) increased by 4 percentage points. The that contract with the Medicare program to provide proportion of those with diagnosed high blood pressure Medicare-covered services to their members. Under the who had it controlled also increased by 4 percentage Balanced Budget Act of 997, Congress required Medicare points. Medicare and employer plan members received managed care plans to use an established process for comparable care and achieved comparable outcomes improving quality of care delivered to Medicare enrollees (NCQA 2004). (MedPAC 2002b). Medicare plans also must report on clinical quality using a set of standard indicators (called Implications: Despite promising gains in the cardiovascular HEDIS) developed by the National Committee for care received by Medicare beneficiaries in managed care Quality Assurance (NCQA). The NCQA uses similar plans, adequate control of cardiovascular disease risk indicators to measure quality of care delivered to privately factors was not attained by roughly one-third. In 2003, the insured individuals in employer-sponsored health plans. NCQA began reporting on the proportion of managed Aggregate results are published by the NCQA in an care plan members who attained optimal cholesterol annual State of Health Care Quality report and Medicare control after a heart attack, and this was achieved by plan-specific results are published on the Centers for only one-half. If every Medicare beneficiary received Medicare and Medicaid Services Web site. HEDIS quality of care equivalent to that provided through the includes several measures of cardiovascular disease care best-performing health plans, thousands of heart attacks, (see Appendix Table 2 for other indicators). strokes, and deaths could be prevented annually (NCQA 2004). Reporting publicly on performance may provide Findings: Managed care plans and their participating an important incentive for improvement. providers improved cardiovascular care for adults in Note: The results shown here are not measured in the same way as, and therefore are both Medicare and employer-sponsored plans from not directly comparable to, those reported on Charts 1:8, 1:12 and 1:13. 2000 to 2003. Among Medicare beneficiaries who suffered a heart attack, the proportion who received beta-blocker medication increased by 4 percentage points, the proportion who had their cholesterol checked increased by 0 percentage points, and the proportion EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:14 Cardiovascular Care and Outcomes in Managed Care Plans Managed care plans and their affiliated providers achieved improvements in cardiovascular care and outcomes for adults in both Medicare and employer plans from 2000 to 2003. Medicare plan members were about equally as likely as employer plan members to receive recommended treatment and both groups achieved comparable outcomes in 2003. �������������������������������������������������������������������������������������������������������� ������������������������������������� ��������������������������� ��� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � ���� ���� ���� ���� ���� ���� ���� ���� ���� ����������������������� ��������������������� ����������� ������� ���������� ������������������ �������������������� ����������� ���������������������� ��������� ����������������������������������������� Source: HEDIS (NCQA 2004). *LDL-C <130 mg/dl. **LDL-C <100 mg/dl. ***Blood pressure of 140/90 mmHg or less. These data are not directly comparable to data reported in Charts 1:8, 1:12, and 1:13 because of differences in measures and methods. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 53 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 54 EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:15 Stroke Prevention for Outpatients with Atrial Fibrillation Why is this important? Atrial fibrillation (AF), points among those ages 80 and older. In contrast, there characterized by a rapid and irregular heart beat, affects was little change in the prescription of anticoagulants up to 5 percent of the elderly ages 65 and older, and up for nonelderly adults with AF. Among visits by patients to 0 percent of those ages 80 and older (Go et al. 200; at the highest risk for stroke, the proportion at which AHA 2005a). In a person with AF, the heart’s upper anticoagulants were prescribed increased by 22 percentage chambers quiver and do not completely pump out blood. points from 99 to 2000, whereas the prescription of As a result, blood may pool and clot. These blood clots aspirin remained relatively constant (Fang et al. 2004). can dislodge and travel to the brain, causing a stroke. Fifteen percent of strokes occur in persons with AF, and Implications: Increased anticoagulation for AF during about half of all AF-related strokes occur in persons older the 990s probably reflects the influence of evidence- than age 75 (AHA 2005a). based treatment recommendations. Anticoagulation The American Heart Association, American College is not always optimally managed among patients who of Cardiology, and the American College of Chest do receive it (Samsa et al. 2000). Some studies report Physicians recommend that persons with AF who are improved medication management and fewer adverse at high risk for stroke should use an adjusted-dose oral events when patients are assigned to an anticoagulation anticoagulant (“blood thinner”) such as warfarin, which clinic or service in which a pharmacist helps manage can reduce the risk of stroke by up to 60 percent in these warfarin therapy (Wilt et al. 995; Chiquette et al. 998; individuals (Fuster et al. 200; Singer et al. 2004). This Wilson et al. 2003). Patient self-management education medication requires frequent monitoring and careful and home self-monitoring has been shown to improve dosing to minimize the risk of abnormal bleeding. anticoagulation control and safety (Siebenhofer et Patients who are at low risk for stroke or who cannot al. 2004; Menendez-Jandula et al. 2005), but lack of safely take anticoagulants should take aspirin. insurance coverage is perceived as a barrier to wider adoption in the United States (Wittkowsky et al. 2005). Findings: In a national sample of visits to physicians by patients with diagnosed AF and no documented contraindications, the prescription of anticoagulant medication increased from 99 to 2000 by 0 percentage points among those ages 65 to 79 and by 34 percentage EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:15 Stroke Prevention for Outpatients with Atrial Fibrillation More elderly patients with a rapid and irregular heart beat are being prescribed recommended blood thinning drugs to reduce their risk of stroke, especially the oldest who are at higher risk for stroke, but many more patients could probably benefit. ������������������������������������������� �������������������������������������������������� ������������������������������������������������������ ����������������������������������������������������� ��������������������������������������� ������������������������������������� ��� ��� �������� ���������� �������� ������� �� �� ������������������ �� �� �� �� � �� �� �� �� �� �� �� � �� �� �� �� �� � � ��������� ��������� ��������� ��������� Source: 1991–2000 National Ambulatory Medical Care Surveys (Fang et al. 2004). *Aspirin use is not shown but was fairly constant at 10 percent to 11 percent of visits from 1991–1992 to 1999–2000 for all patients with atrial fibrillation (AF). **Patients at highest risk for stroke are defined as those with AF who were older than age 75 or who had a prior diagnosis of transient ischemic attack or stroke (excluding intracranial hemmorrhages), valvular heart disease, hypertension, or heart failure. These data represent civilian, noninstitutionalized individuals. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 55 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 56 EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:16 Diabetes Management Why is this important? Diabetes is associated with lipid levels, which can lead to complications of heart increased functional disability and premature death disease. However, one-quarter (25%) did not have the among the elderly and is a risk factor for heart disease. recommended annual dilated eye examination to check Diabetes prevalence increases with age. Fourteen percent for signs of retinopathy, an eye disease that can lead to of elderly white adults and almost one-quarter of elderly blindness, and three of 0 (3%) did not have their feet black and elderly Hispanic adults report that they have checked for signs of nerve damage. Compared to middle- diabetes (FIFARS 2004). age adults (ages 45–64), the elderly were somewhat less Research in nonelderly populations has found that the likely to receive a hemoglobin test but were more likely to development and progression of diabetes complications receive an eye exam (AHRQ 2005b). can be reduced through control of blood sugar, blood pressure, and blood lipids. Complications of diabetes Implications: Diabetes management requires a collaboration include blindness, kidney failure, and cardiovascular between health care professionals and their patients, disease resulting in heart attacks, strokes, and often involving the expertise of a multidisciplinary amputations. Intensive diabetes management requires care team (Jack et al. 2004). In 998, Medicare began a number of years to produce benefits and may reduce covering the cost of a diabetes education program, quality of life in the short term. Therefore, treatment for blood glucose monitors, and testing strips, which can diabetes must be customized to the needs of the elderly help patients manage and control their diabetes (ADA individual considering life expectancy and disease 2004). Longitudinal data from the Medicare Quality comorbidities, with regular monitoring to adjust therapy Improvement Organization program (Jencks et al. 2003), and goals as appropriate (Brown et al. 2003). Medicare managed care plans (NCQA 2004), and state- level surveys (CDC 2002) indicate that diabetes care Findings: Among community-dwelling elderly Americans has been improving among all adults with diabetes, (ages 65 and older) with diabetes in 200, nine of 0 including the elderly. A review of quality improvement (89%) reported that they had received a glycosylated studies found that multifaceted approaches involving hemoglobin test in the past year, which provides a three- organizational change, patient education, and/or provider month average reading of blood sugar control so that education can be effective in improving clinician the doctor can adjust medications and recommend diet compliance with guidelines and patient outcomes of care and exercise changes. Likewise, most (95%) had their (Shojania et al. 2004). blood lipids checked to monitor control of abnormal EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:16 Diabetes Management Most elderly Americans with diabetes report that they are receiving recommended tests to monitor their blood sugar and lipids. One-quarter did not have an eye exam and three of 10 did not have their feet checked for signs of diabetes complications. Compared to middle-age adults, the elderly were somewhat less likely to receive a hemoglobin A1c test but were more likely to receive an eye exam. ����������������������������������������������������������� �������������������������������������������������������� ��� �� �� ���������� �������� �� �� �� �� �� �� �� �� �� �� � ���������������������� ��������������������� ����������������������� ���������������������� �������������������������� ���������������������������� ���������������������� ������������������������� ������������� �������������������� ������������ ��������������������� Source: Medical Expenditure Panel Survey (AHRQ 2005b). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 57 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 58 EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:17 Osteoarthritis: Evaluation and Treatment Why is this important? Osteoarthritis, a degenerative for pain at least annually. More than two-thirds were condition in which the joints become stiff, swollen, and offered education regarding the treatment and self- painful, is prevalent among the elderly, affecting more management of their condition, but less than half of than 50 percent of individuals ages 65 years and older and those eligible were prescribed a strengthening or exercise more than 85 percent of those ages 75 and older. Although program. Acetaminophen was the first-line therapy there is no cure for osteoarthritis, several forms of among only three of five using drug therapy. Patients treatment can reduce pain, limit functional impairment, prescribed NSAIDs were often not warned of the risks and maintain or improve joint mobility (AHRQ 2002d). associated with these drugs or offered prophylaxis for The American College of Rheumatology (ACR 2000), gastrointestinal bleeding when they were at potential American Academy of Orthopaedic Surgeons (AAOS risk for these side effects. Almost three-quarters of those 2003), and American Geriatrics Society (AGS 998, 200) eligible were offered a referral for surgical evaluation have published guidelines that emphasize the importance (Wenger et al. 2005). of pain assessment and patient education and self- management. The guidelines recommend a combination Implications: These exploratory findings, based on a of nonpharmacologic therapy and drug therapy. Research limited population, suggest that treatment practices for shows that exercise can reduce or eliminate many of the osteoarthritis in older individuals may not be in accord major risk factors for osteoarthritis, including obesity, with evidence-based expert recommendations. Wider muscle weakness, inactivity, and poor joint biomechanics. use of standard quality measures for osteoarthritis A recent review of evidence supports recommendations would enable better understanding and improvement that acetaminophen be considered as initial therapy of osteoarthritis care practices. Recent media coverage for mild to moderate joint pain, because it causes fewer surrounding the market withdrawal of a popular adverse reactions than nonsteroidal anti-inflammatory prescription painkiller might be prompting wider drugs (NSAIDs) (Wegman et al. 2004). discussion of the risks posed by NSAIDs than was common at the time of this study. Findings: Among at-risk* patients ages 75 and older with * At-risk patients included those who screened positive for falls or fear of falling, osteoarthritis treated in two medical groups participating bothersome incontinence, or memory impairment. in the Assessing Care of Vulnerable Elders (ACOVE-2) study during 2002–2003, only three of five were evaluated EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:17 Osteoarthritis: Evaluation and Treatment Osteoarthritis is the most prevalent chronic disease among older adults, causing up to half of all disability among seniors. A pilot study found that older patients with osteoarthritis were often not evaluated for pain, provided patient education, or prescribed recommended therapies. Almost three-quarters were referred for surgical evaluation when appropriate. ����������������������������������������������������������������������������������������� ������������������������������������������������������������ � �� �� �� �� ��� ������������������������������������������������������ �� ���������������������������������������������������������������� �� �������������������������������������������������������������������� �������������������������������������������� �� ������������������������������������������������ ����������������������������������������������������������������� �� ������������������������������������������������������� ����������������������������������������������������������������� �� ������������������������������������������������������ �������������������������������������������������������� �� ������������������������������������������� ���������������������������������������������������������������������������� �� ���������������������������������������������������������������������� Source: Medical records and patient interviews from the Assessing Care of Vulnerable Elders (ACOVE-2) study ( Wenger et al. 2005). *At-risk patients screened positive for falls or fear of falling, bothersome incontinence, or memory impairment (N = 644). **Except when contraindicated. ***Among all ACOVE-2 patients treated with NSAIDs. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 59 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 60 EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:18 Urinary Incontinence: Screening and Management Why is this important? Urinary incontinence (UI) affects Although physicians discussed treatment options almost one-quarter (22%) of community-dwelling with most individuals with UI, behavioral treatment Medicare beneficiaries ages 65 and older and more was offered to only 5 percent of patients who could than one-third (35%) of those ages 85 and older (NCHS potentially benefit (Wenger et al. 2005). 2005). This problem can reduce an individual’s quality of life, leading to social isolation, loss of self-esteem, Implications: The findings of this study are consistent with and depression. Dependence on caregivers increases as other research indicating that UI in the elderly often incontinence symptoms worsen. UI is one of the major goes undetected and is undertreated by primary care causes of institutionalization of the elderly, prevalent physicians (AHRQ 996). A prior study investigating in more than 50 percent of the individuals in nursing why physicians do not ask older patients about UI facilities (Gnanadesigan et al. 2004). found a lack of time and patient embarrassment to Treatment options for the management of UI in adults be the most frequently reported reasons. Moreover, may include behavior therapy, medication, and surgery nearly three-quarters of physicians underestimated the (AHRQ 996). When offered a choice, most patients proportion of older patients who could benefit from prefer behavioral therapy, which is effective in reducing therapy and half said they did not feel prepared to UI for up to 80 percent of ambulatory and mentally treat this condition (CDC 995). High priority should competent adults (Diokno and Yuhico 995; Burgio et al. be placed on research to test and identify effective 998; Teunissen et al. 2004). interventions that will help physicians improve their ability to detect and treat this problem, given its Findings: Among at-risk* patients ages 75 and older treated prevalence and consequences for the elderly. in two medical groups participating in the Assessing * At-risk patients included those who screened positive for falls or fear of falling, Care of Vulnerable Elders (ACOVE-2) study during bothersome incontinence, or memory impairment. 2000–200, fewer than two of five were screened by their doctors to determine if they had UI. Physicians treating those with UI obtained a complete history or performed a physical exam for only about half of the patients, and recommended lab work was often not performed. EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:18 Urinary Incontinence: Screening and Management Urinary incontinence (UI) affects many seniors and can lead to activity limitations, social isolation, and depression. Even when physicians recognized a patient as having UI, they often did not perform a complete history and physical exam or order recommended lab work. Although treatment options were often discussed, behavioral therapy was seldom offered. ��������������������������������������������������������������������� ������������������������������������������������������������ � �� �� �� �� ��� ���������������������������������������������� �� ��������������������������� �� ���������������������������������� �� ������������������������������ �� ������������������������������������������������������������ �� ����������������������������� �� ��������������������������������������������� �� Source: Medical records from the Assessing Care of Vulnerable Elders (ACOVE-2) study ( Wenger et al. 2005). *At-risk patients screened positive for falls or fear of falling, bothersome incontinence, or memory impairment (N = 644). **Among those with new or worsening urinary incontinence (UI) that persists for over one month or UI at the time of new evaluation. ***Among cognitively intact patients who are capable of independent toileting and have documented stress, urge, or mixed incontinence without evidence of hematuria or high post-void residual. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 61 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 62 EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:19 Treatment for Depression Why is this important? An estimated 2 million elderly mood). Between 999 and 200, only 27 percent of Americans, or 6 percent of those ages 65 and older, suffer those ages 60 to 64, 32 percent of those ages 65 to 74, from a depressive illness and another 5 million, or 5 and 26 percent of those ages 75 and older had received percent, experience depressive symptoms (NIMH 2003). potentially effective recent treatment for depression (at Up to 25 percent of those with chronic illness suffer least two months of antidepressant medication or four comorbid depression (DHHS 2000). Late-life depression or more psychotherapy or counseling sessions within the is associated with increased use of health care and an past three months). Men, African Americans, Latinos, increased risk of medical illness and suicide (Unutzer et and those who preferred psychotherapy to medication al. 997; Katon et al. 2003). Depressed elderly adults are reported significantly lower rates of recent depression less likely than younger or middle-age depressed adults to care (data not shown) (Unutzer et al. 2003). perceive that they need mental health care or to receive any specialty mental health care (Klap et al. 2003). Implications: These data are consistent with other research In recognition of the significant public health problem indicating that depression is undertreated in the elderly posed by depression in older adults, a National Institutes (DHHS 2000). For example, only two-thirds of Medicare of Mental Health Consensus Panel recommended beneficiaries diagnosed with depression between 992 aggressive approaches to recognize, diagnose, and treat and 998 received any treatment for depression in the elderly individuals suffering from late life depression year that they were diagnosed (Crystal et al. 2003). (NIH 99; Lebowitz et al. 997). The U.S. Preventive Interventions that support effective depression treatment Services Task Force recommends depression screening through primary care may be more acceptable to elderly for all adults in the primary care setting coupled patients than those that seek to facilitate referral to with systematic depression treatment, including specialty care (Bartels et al. 2004). (See Chart 6:4 for a antidepressants and/or psychotherapy (USPSTF 2002c). description of the intervention phase of this study, which substantially improved depression care and outcomes for Findings: As part of a quality-improvement intervention these patients.) at 8 primary care clinics across the United States, researchers identified and interviewed a sample of ,80 adults ages 60 and older who met diagnostic criteria for major depression or dysthymia (chronic depressed EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:19 Treatment for Depression Among older patients of 18 clinics whom researchers determined had current major depression or dysthymia (chronic depressed mood), less than one-third had recently received treatment that experts would consider effective. ������������������������������������������������������������������������������� ����������������������������������������������������������������������������������� �� �� �� �� �� �� � ���������� ���������� �������� Source: Patient interviews (Unutzer et al. 2003). *Potentially effective recent treatment means at least two months of antidepressant medication or four or more counseling or psychotherapy sessions for depression in past three months. Results may not be nationally representative. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 63 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 64 EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:20 Mental Health Care in Managed Care Plans Why is this important? Medical management of depression prescribed an antidepressant medication. In 2003, little is often not optimal, whether patients are seen in primary more than half (53%) remained on their antidepressant care or by mental health specialists (Katz et al. 998; during the 2-week acute treatment phase and less than Simon et al. 200). Many patients who are started on an two of five (39%) completed six-months of continuation antidepressant medication do not complete therapy and phase treatment. Only one of 0 (%) had at least three do not have adequate follow-up with their physician to follow-up visits with their physician during the acute monitor medication safety and effectiveness. Response to treatment phase. Similarly, only two of five (39%) of those medication may be slower in the elderly, requiring up to hospitalized for a mental illness had follow-up within a 2 weeks of therapy to achieve maximum effect and six week and only three of five (60%) within 30 days of being months’ continuation to prevent remission (NIH 99; hospitalized. These rates of treatment did not improve DHHS 2000). Patients who are treated in accordance much from 200 to 2003 and were worse for Medicare with guidelines are less likely to experience a relapse in than employer plan members (NCQA 2004). depression (Sood et al. 2000). When an individual requires hospitalization for Implications: There is no representative data such as this for mental illness, it is important to provide follow-up care to patients outside of managed care plans and limited studies support the transition back home and assure continued suggest that care is unlikely to be better. The differences improvement (NCQA 2002). Because some individuals between Medicare and employer plans probably reflects do not seek follow-up care on their own, reminder poorer mental health care for the elderly in general. The systems may be needed to proactively schedule such NCQA identifies mental health care as a weak spot that visits. The National Committee for Quality Assurance remains an exception to improvement seen in other areas (NCQA) developed several measures of mental health of quality measurement and reporting by managed care care quality that are used by health care purchasers and plans (NCQA 2004). Some interventions have improved regulators to monitor the performance of managed care treatment adherence and patient outcomes and reduced plans on these topics. relapse among patients who received telephonic and/or in-person support from an intermediate-level practitioner Findings: Medicare beneficiaries in managed care plans (Tutty et al. 2000; Katon et al. 200). often do not receive recommended outpatient medication management when diagnosed with depression and EFFEC TIVENESS • LIVING WITH ILLNESS • CHART 1:20 Mental Health Care in Managed Care Plans Medicare beneficiaries in managed care plans often do not receive recommended medication management when they have been diagnosed with depression and prescribed an antidepressant. Many do not receive timely follow-up after a hospitalization for mental illness. Rates did not improve much from 2000 or 2001 to 2003 and were worse for Medicare than employer plan members. �������������������������������������������������������������������������������������������������������� ��� ������������������������������������� ��������������������������� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � ���� ���� ���� ���� ���� ���� ���� ���� ���� ���� ��������������� ��������������������� �������������������� ������������� �������������� ���������������� ���������������� ���������� ��������������������������������������������������������� �������������������������������� ����������������������������������������������������� ������������������ Source: HEDIS (NCQA 2004). *Those who continued using an antidepressant for 12 weeks after diagnosis (acute phase) or for six months after diagnosis (continuation phase). **At least three follow-up contacts during the 12-week acute phase. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 65 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 66 EFFEC TIVENESS • CHANGING NEEDS • CHART 1:21 Home Health Care Outcomes Why is this important? In 2000, Medicare spent 8.7 billion 2002 to 2004 (AHRQ 2005b; CMS 2005a).* Home health on home health care services for 2.5 million home- care agencies were more successful at improving some bound Medicare beneficiaries (GAO 2002b). Medicare patient outcomes, such as healing wounds and pressure reimburses home health agencies (HHAs) for episodes sores, than others such as medication management and of care lasting no more than 60 days. Services must be urinary incontinence. provided in accordance with a physician’s care plan and may include skilled nursing and aide services, physical Implications: Because Medicare pays for much of the and occupational therapy, speech pathology services, home health care delivered in the United States, it plays and medical social work. The goals of home health care an important role in setting standards and expectations are to “enable individuals to remain as functional and affecting quality. Home health quality assessment is independent as possible in their own homes, thereby challenging given that there are no accepted standards avoiding institutional long-term care” (CMS 2003). for the processes of care that should be delivered and the As a condition of participation in Medicare, the fact that different care providers may work independently 6,900 HHAs that contract with Medicare must undergo within each patient’s home. Outcomes assessment and periodic quality assurance surveys, develop continuous reporting provides one way to monitor and encourage quality-improvement programs, and collect standard improvements in the quality of home health care and patient assessment data called the Outcome Assessment may be especially important given the incentives for and Information Set (OASIS). The Centers for Medicare undertreatment inherent in Medicare’s prospective and Medicaid Services (CMS) collects OASIS data in payment system (MedPAC 2004a). (See Chart 6:6 for an a national repository, which it uses to generate reports example and discussion of Medicare’s Outcome-Based for home health agencies to use in quality-improvement Quality Improvement system for home health care.) activities. CMS publishes a subset of these outcomes on * The data shown in the chart represent adult patients who received skilled care from its Web site. Medicare-certified home health agencies and whose care was paid for by Medicare or Medicaid. It does not include patients who received maternity services or who received only personal care. Findings: Functional outcomes for adult patients (ages 8 and older) served by Medicare-certified home health care agencies improved by  to 5 percentage points across nine publicly reported quality measures from EFFEC TIVENESS • CHANGING NEEDS • CHART 1:21 Home Health Care Outcomes There was a trend toward improvement in functional outcomes for home health care patients across multiple measures from 2002 to 2004. Patients tend to improve more for some types of outcomes, such as toileting, than for others, such as ambulation. ��������������������������������������������������������������������������������������������� � �� �� �� �� ��� �� ��������������������������������������������� �� �� ����������������������������� �� �� �������������������������������������������� �� �� ��������������������� �� �� ���� �������������������������������� �� ���� �� ��������������������������������������� �� �� ������������������������ �� �� ������������������������������������������������� �� �� �������������������������������������� �� Source: Outcome and Assessment Information Set (OASIS). Data for 2002 were reported by the Agency for Healthcare Research and Quality (2005b). Data for 2004 were reported by the Centers for Medicare and Medicaid Services (2005a) Home Health Compare Web site. Measures shown represent a subset of functional outcomes measured for which results were publicly reported for both 2002 and 2004. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 67 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 68 E F F E C T I V E N E S S • M U LT I P E R S P E C T I V E • C H A R T 1 : 2 2 State-Level Performance on Medicare Quality Indicators Why is this important? Medicare launched a Health Care Implications: The median relative improvement represents Quality Improvement Program in the 990s to promote one measure of the degree to which the quality gap—the the widespread adoption of professionally developed, difference between actual and ideal performance—was evidence-based standards of care (Jencks and Wilensky reduced. Twelve of 3 states in the highest quartile of 992). The Medicare program contracts with Quality performance in 2000–200 also ranked higher (first or Improvement Organizations (QIOs), formerly called second quartile) in relative improvement. In contrast, Peer Review Organizations, that work with health care 0 of 2 states in the lowest quartile of performance in providers in every state “to systematically promote 2000–200 also ranked lower (third or fourth quartile) in improved performance on the quality measures tracked relative improvement. With leadership and commitment, under this program using a voluntary, collaborative, and health care providers in poorly performing states may be nonpunitive educational strategy” (Jencks et al. 2000). able to emulate methods used in better-performing states QIOs provide quality improvement strategies, pretested to develop a stronger infrastructure for improvement. educational materials, a forum for collaboration, The federal government has proposed that QIOs work and customized technical assistance free of charge to more intensively with Medicare providers to achieve participating Medicare providers (CMS 2000b). (See the significant performance improvement in several areas Introduction for additional background on this topic.) during the next three years (CMS 2004c). More than 4,000 U.S. hospitals have volunteered to participate in the Findings: During 2000–200, northern and less populous Hospital Quality Alliance, a public-private partnership states tended to perform better across 22 indicators of the that is reporting participants’ performance on 7 quality effectiveness of care delivered to Medicare beneficiaries, indicators, which may provide further incentive for including preventive care and/or treatment for heart improvement (CMS 2005b). attack, heart failure, stroke, pneumonia, influenza, * Relative improvement was measured as absolute change / (100 - baseline). The diabetes, and breast cancer (see Appendix Table a for Medicare Quality Improvement Organization program included 24 quality indicators a list of indicators and national rates of performance but two indicators measuring time to reperfusion were excluded from the state rankings described in this chart. on each indicator). From 998–999 to 2000–200, the median state’s performance across the 22 indicators improved from 69.5 percent to 73.4 percent, representing a 2.8 percent relative improvement* (Jencks et al. 2003). E F F E C T I V E N E S S • M U LT I P L E C O N D I T I O N S • C H A R T 1 : 2 2 State-Level Performance on Medicare Quality Indicators Average state performance on provision of effective care to Medicare fee-for-service beneficiaries, by quartile rank, 2000–2001 �� Northern and less populous states tended to perform better across 22 indicators of the quality of care ������������� delivered to Medicare beneficiaries, including ��������� preventive care and/or treatment for heart attack, � heart failure, stroke, pneumonia, influenza, diabetes, � and breast cancer (see Appendix Table 1a for a list �������� of the indicators included in this ranking). Median relative improvement* in the provision of effective care to Medicare fee-for-service beneficiaries, by quartile rank �� From 1998–1999 to 2000–2001, the median state’s ������������� performance across the 22 quality indicators improved ������� from 69.5 percent to 73.4 percent, representing ��������������� a 12.8 percent relative improvement.* This is a ��������������� measure of the degree to which the gap between ��������������� actual and ideal performance was reduced. ������������� Source: Center for Medicare and Medicaid Services, Medicare Quality Improvement Organization program (Jencks et al. 2003). Adapted and used with permission from: Journal of the American Medical Association, Jan. 15, 2003, 289: 310-11. Copyrighted © 2003, American Medical Association. All Rights reserved. *Relative improvement was defined as absolute change / (100 - baseline). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 69 1: EFFECTIVENESS 1: EFFECTIVENESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 70 E F F E C T I V E N E S S • M U LT I P L E C O N D I T I O N S • C H A R T 1 : 2 3 Assessing Care of Vulnerable Elders Why is this important? More than one-third (36%) of Although treatment-related quality was high (8% of national health care expenditures goes toward the care of indicators met), recommended processes of care related the elderly (Keehan et al. 2004), yet little is known about to prevention and diagnosis were attained less than half of the quality of care that these patients receive for common the time (43% and 46%, respectively) (Wenger et al. 2003). geriatric problems. As older persons begin to decline functionally, they typically place the greatest priority Implications: Although this study was limited to a small on maintaining functional ability and quality of life population, it provides a caution that the quality of care (Phillips et al. 996), yet studies of this population tend to for vulnerable elders is often suboptimal. The finding that focus on general medical conditions and longevity. The quality of care for geriatric conditions, such as dementia Assessing Care of Vulnerable Elders (ACOVE) project and urinary incontinence, is poorer than care for general used an expert-consensus process to develop a system medical conditions calls into question whether the quality for comprehensively assessing quality of care across 22 indicators in widespread use today are sufficient to assure target conditions that are important to the well-being of high-quality care for the elderly. Based on these findings, vulnerable elders in the community. This system uses 236 a series of interventions is being tested in community expert-validated quality indicators covering screening medical groups to improve performance on a subset and prevention, diagnosis, treatment, and follow-up and of the worst-performing geriatric conditions (urinary continuity of care (Wenger et al. 2003). incontinence, falls, and dementia). Results from these interventions will help guide the development of physician Findings: Two managed care organizations participated in and patient educational materials and tools designed an ACOVE pilot study that assessed the care provided to facilitate better care of the growing vulnerable older during 998–999 to vulnerable elders ages 65 and older population (Reuben et al. 2003b; Wenger et al. 2005). at risk for functional decline or death. The care provided to these patients met only 55 percent of the 236 quality indicators and varied widely across conditions, ranging from a high of 82 percent for stroke care to a low of 9 percent for care at the end of life. Quality indicators for geriatric conditions were met less frequently than those for general medical conditions (3% vs. 52%). E F F E C T I V E N E S S • M U LT I P L E C O N D I T I O N S • C H A R T 1 : 2 3 Assessing Care of Vulnerable Elders In a pilot study in two health plans, the quality of care provided to vulnerable elders—those at higher risk for functional decline or death—met expert standards only a little more than half the time. The greatest gaps in quality occurred in the care of geriatric conditions and in preventive care. ��������������������������������������������������������������������� ����������������������������������������������������������� ��� �� �� �� �� �� �� �� �� �� �� �� �� �� � � ������ ������� ����������� ������� ��������� ������������ ���������� ��������� ��������� ��������� ����� ���������� ���� ������� ��������� ���������� �������� ����������������� �������������������� ����������������� Source: Assessing Care of Vulnerable Elders (ACOVE-1) indicators applied to patient interviews or medical records ( Wenger et al. 2003). Vulnerable elders were defined as community-dwelling persons ages 65+ who have four times the risk for functional decline or death over the next two years (N = 420). *Stroke care category also includes care for patients with atrial fibrillation. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 71 1: EFFECTIVENESS SECTION 2 Patient Safety Patient safety means “avoiding injuries to patients from the care that is intended to help them.” – Institute of Medicine 2001a 2: SAFETY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 74 PAT I E N T S A F E T Y • C H A R T 2 : 1 Adverse Events and Postoperative Complications of Care Why is this important? The Institute of Medicine reported of hip joint replacement surgery (when done following in 999 that thousands of Americans are harmed each a fracture) to a low of 0.4 percent of all hospital patients year from the health care that is intended to help who experienced a hospital-acquired bloodstream them (IOM 999). The IOM called upon all concerned infection (AHRQ 2005b).* stakeholders to take specific actions to improve patient safety. Other high-risk industries, such as aviation and Implications: A patient-oriented approach to patient safety nuclear power, do not depend on human perfection to recognizes that the patient’s ultimate concern is freedom achieve high levels of safety. Rather, they design “fault- from harm. “Physicians and organizations should strive tolerant” systems that prevent harm by taking account of to prevent or mitigate situations that actually cause harm the human propensity for error (Reason 997). Such an to patients, whether the harm is caused by an error or a approach seeks information on system vulnerabilities so faulty or inefficient process,” says patient safety expert that they can be proactively mitigated and eliminated. Roger Resar, M.D. (quoted by Neveleff 2003). The The Medicare Patient Safety Monitoring System occurrence of some adverse events is related, at least (MPSMS) is a nationwide surveillance program intended in part, to patients’ underlying conditions and the risk to help achieve this goal for the Medicare program. The inherent in some treatments. Nevertheless, the experience MPSMS uses explicit (structured and objective) review of anesthesiology and of several health care organizations of hospital medical records and administrative data to shows that adverse events can be dramatically reduced determine rates of specific adverse events of importance and in some cases eliminated by creating systems that to the Medicare population. An adverse event is defined reliably provide evidence-based treatment, encourage as an “unintended harm, injury, or loss that is more likely proactive nursing care, and promote good teamwork and associated with [the patient’s] interaction with the health communication (Gaba 2000; Khuri et al. 2002; Schoeni care delivery system than from an attendant disease 2002; Bellomo et al. 2003; Pronovost and Berenholtz process” (Hunt et al. 2004). 2004). * These rates of adverse events are not directly comparable to those reported in Findings: Among Medicare fee-for-service beneficiaries Charts 2:2 and 2:3 because of differences in methods and data sources. hospitalized during 2002, the rates of  high-priority adverse events ranged from a high of 22.6 percent of patients who experienced postoperative complications PAT I E N T S A F E T Y • C H A R T 2 : 1 Adverse Events and Postoperative Complications of Care Medicare began a national program of monitoring adverse events and complications of hospital care in 2002. Although the occurrence of these events is related in part to patients’ underlying conditions, many might be preventable with good medical and nursing care. �������������������������������������������������������������������� �������������������������������������������������������������������� � � �� �� �� �� ������������������������������������������������������ ���� ������������������������������������������������������������������� ��� ���������������������������������������� ��� ���������������������������������������� ��� ������������������������������������������������� ��� �������������������������������������� � ��� ������������������������������������������������� ��� ������������������������������ � ��� ���������������������������������������������� ��� ������������������������������������������ � ��� ���������������������������������������� � ��� Source: Medicare Patient Safety Monitoring System (AHRQ 2005b). These data are based on explicit reviews of medical records and are not directly comparable to data shown in Charts 2:2 and 2:3 because of differences in methods and sources. *Among patients who had the indicated surgery. **Among patients who were on a ventilator. ***Among patients in whom a central venous catheter (CVC) was inserted. †Among all surgical patients. ‡Among all hospital patients. See Technical Appendix for definitions. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 75 2: SAFETY 2: SAFETY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 76 PAT I E N T S A F E T Y • C H A R T 2 : 2 Trends in Adverse Events and Complications of Care in the Hospital Why is this important? The Institute of Medicine’s 999 Implications: These findings suggest that adverse events report, To Err is Human, prompted national efforts to might be increasing or that they are more often diminish threats to patient safety (IOM 999). Before the being recorded in hospital billing records. Another IOM report, few tools and very little data were available analysis using 8 Patient Safety Indicators applied to to understand and monitor the scope and types of safety hospitalizations for the entire U.S. population estimated problems affecting patients. As one response, the federal that potentially preventable adverse events accounted for Agency for Healthcare Research and Quality created 2.4 million additional hospital days, 32,600 deaths, and Patient Safety Indicators, which use hospital billing 4.6 billion in additional cost to the health care system records to “screen for problems that patients experience (Zhan and Miller 2003). as a result of exposure to the healthcare system, and The findings shown in the chart are not definitive that are likely amenable to prevention by changes at the because of the limitations of the administrative data on system or provider level” (AHRQ 2003b). which they are based. The increase in some rates could be caused, in part, by improved accuracy or changes in Findings: Medicare fee-for-service beneficiaries hospitalized coding, such as the introduction in 998 of a new code during 2000 experienced more than 325,000 potentially for acute and respiratory failure. However, experts told preventable adverse events and complications of care staff of the Medicare Payment Advisory Commission identified by 3 Patient Safety Indicators. The risk-adjusted that changes in coding were unlikely to account for other rate of adverse events increased for nine of the 3 indicators observed increases (MedPAC 2004c). from 995 to 2002 (only the top 0 are shown on the chart, The trends identified in this analysis deserve ongoing excluding “failure to rescue”). For example, the rate of monitoring and further investigation. Individual hospitals pressure sores during a hospital stay of five days or longer might use such information to help identify areas where increased by 35 percent (absolute increase of 82 per 0,000 process improvement is needed and develop effective discharges). On the other hand, the rate of hip fractures strategies that promote better organization, training, following surgery decreased by 28 percent (absolute procedures, teamwork, and communication to increase decrease of 5 per 0,000 discharges) (MedPAC 2004c). patient safety. PAT I E N T S A F E T Y • C H A R T 2 : 2 Trends in Adverse Events and Complications of Care in the Hospital Rates of several potentially preventable adverse events and complications of hospital care increased from 1995 to 2002 among Medicare fee-for-service beneficiaries, and/or they were more often recorded in hospital billing records. ��������������������������������������������������������������������������������������� �������������������������������������������������������������������������������� � �� ��� ��� ��� ��� ��� ��� ��� ����������������������������������� ��� �� ������������������������������������������������������������ ��� �� ������������������������ ��� �� ������������������������������������� �� �� ��������������������������������������������������� �� ���� �� ���������������������������������� �� ���� �� ���������������������������������� �� �� ��������������������������������� �� �� ������������������������������ �� �� ������������������������������������������������������� �� Source: Medicare Payment Advisory Commission (2004c) analysis of Medicare administrative data using AHRQ Patient Safety Indicators. Only the 10 highest rates (other than “failure to rescue”) are shown. Rates exclude complications present on admission and are adjusted for age, gender, age -gender interactions, comorbidities, and diagnosis-related group clusters. *See Technical Appendix for footnotes defining the population at risk for each measure. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 77 2: SAFETY 2: SAFETY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 78 PAT I E N T S A F E T Y • C H A R T 2 : 3 Adverse Events and Complications of Care in the Hospital, by Patient Age Why is this important? Because the elderly generally use Findings: According to an analysis that applied Patient more health care services than other age groups, they Safety Indicators to a national sample of hospitalizations are more often exposed to potential patient safety threats during 200, the oldest old, as compared to middle-age (Thomas and Brennan 2000). Adverse events such as adults, were less likely to develop infections attributable falls are more frequent and their consequences may be to intravenous lines or catheters, almost twice as likely more severe among the elderly (Rothschild et al. 2000). to experience deep vein thrombosis or pulmonary Understanding the incidence of adverse events in the elderly embolism, and over three times more likely to suffer from might help hospitals appreciate the scope of these problems pressure sores during long hospital stays (AHRQ 2005b). and investigate how they can reliably employ preventive strategies to help reduce their occurrence. For example: Implications: A review of research concluded that “[t]he • Infections associated with intravenous lines and catheters main cause of these increased risks [to the elderly] can be reduced or prevented by using simple tools such appears to be the diminished physiological reserve of as checklists and a standardized supply cart to ensure elderly patients; however, age alone is a less important compliance with the CDC’s infection control guidelines predictor of adverse events than comorbidities and and by asking daily during patient rounds whether functional status” (Rothschild et al. 2000). Patient safety catheters can be removed (Berenholtz et al. 2004). experts recommend that hospitals promote a culture of • Formation of blood clots in the leg (deep vein safety, apply human factors principles to minimize error thrombosis), which may travel to and become lodged in through work design, consider the potential benefits of the lungs (pulmonary embolism), is often preventable geriatric specialists and geriatric care units to improve if providers follow recommendations of the American care, and perform a comprehensive geriatric assessment College of Chest Physicians for use of anticoagulants, at admission to predict risk of complications. compression stockings, and pneumatic compression devices (Geerts et al. 2004). • Pressure sores may be preventable with interventions such as regular skin assessments, turning schedules, pressure reduction devices, and nutritional supplements (AHRQ 992). Using prevention protocols and making system improvements reduced the incidence of pressure sores by up to two-thirds in some studies (Bergstrom 997). PAT I E N T S A F E T Y • C H A R T 2 : 3 Adverse Events and Complications of Care in the Hospital, by Patient Age This chart focuses on three adverse events or complications of care that can often be prevented with good medical and nursing care. The oldest old, as compared to middle-age adults, are less likely to develop infections attributable to intravenous lines or catheters in the hospital, almost twice as likely to have postoperative blood clots form in their legs and/or travel to their lungs, and over three times more likely to experience pressure sores during long hospital stays. ����������������������������������������������������������������� ����������������������������������������������������������������������� ���������� ���������� ���������� ���������� ���������� �������� ��� ��� ��� ��� ��� ��� ��� ��� ��� ��� ��� �� �� �� �� �� �� �� �� �� �� � � ������������������������������� �������������������������������� ������������������������������������ ������������������������� Source: Patient Safety Indicators applied to Health Care Utilization Project Nationwide Inpatient Sample (AHRQ 2005b). Rates exclude complications present on admission and are adjusted for gender, comorbidities, and diagnosis- related group clusters. *Infections primarily related to intravenous lines and catheters. **Among surgical patients. ***Among patients with hospital stays of five days or longer. See the Technical Appendix for specific exclusions. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 79 2: SAFETY 2: SAFETY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 80 PAT I E N T S A F E T Y A N D T I M E L I N E S S • C H A R T 2 : 4 Appropriate Use of Antibiotics to Prevent Surgical Infections Why is this important? About three of every 00 operations only a little more than half (56%) of patients were given performed in the United States are complicated by antibiotics within one hour before surgical incision, and surgical site infections (SSIs) (Gaynes et al. 200). less than half (4%) had prophylaxis discontinued within Compared with uninfected patients, patients developing 24 hours after surgery. Nearly 0 percent of patients SSIs are twice as likely to die, are over five times received their first antibiotic dose more than four hours more likely to be readmitted to the hospital, spend an after surgical incision, which is too late to provide a additional 6.5 days in the hospital, and incur more than benefit. The median time to antibiotic discontinuation 3,000 in excess direct health care costs (Kirkland et al. was more than 40 hours versus the recommended 24 999). The Centers for Medicare and Medicaid Services hours (Bratzler et al. 2005). and the Centers for Disease Control and Prevention partnered in 2002 to form the National Surgical Infection Implications: Improvement is needed in SSI prevention Prevention Project. The project promotes evidence-based practices for about half of patients undergoing major use of prophylactic antibiotics to reduce the risks of SSI, surgery. Higher rates of timely antibiotic administration antibiotic complications, and bacterial drug resistance before surgery might be associated with more frequent (Bratzler and Houck 2004). These practices include use of preprinted care plan forms that include antibiotic treating patients with appropriate (relatively narrow- protocols. The Surgical Care Improvement Project, a spectrum) antibiotic drugs, giving antibiotics within one national partnership of organizations seeking to improve hour of surgical incision, and discontinuing antibiotic surgical care by reducing postoperative complications, is prophylaxis within 24 hours after surgery (CMS 2003; launching a five-year campaign to reduce the incidence Bratzler et al. 2005). of surgical complications by 25 percent by the year 200. A collaboration among Medicare Quality Improvement Findings: A nationwide review of the medical records of Organizations at 56 medical centers across the country fee-for-service Medicare patients who underwent one achieved a 27 percent reduction in SSIs using evidence- of five types of major surgery in 200 found that nearly based practices such as these, according to preliminary all patients received prophylactic antibiotics. Most findings (SCIP 2004). (93%) of these patients received a relatively narrow- spectrum antibiotic drug consistent with guidelines. The appropriate timing of antibiotic administration was poor: PAT I E N T S A F E T Y A N D T I M E L I N E S S • C H A R T 2 : 4 Appropriate Use of Antibiotics to Prevent Surgical Infections Surgical site infections substantially increase the use of health care resources and the risk of patient death. Many Medicare patients undergoing surgery in 2001 did not receive antibiotics in a manner consistent with evidence about how to effectively prevent postoperative infections while minimizing the risk of spreading antibiotic resistance. ������������������������������������������������������������������������������������������� ���������������������������������������������������������������������������� ������������������� �������������������������������������� �������������������������� ��������������� �������������������������������������� ������������������������ ��� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � ������������������� ������� �������� �������� ����� ������������ ����������������� Source: Centers for Medicare and Medicaid Services, National Surgical Infection Prevention Project, review of medical records (Bratzler et al. 2005). *Prophylactic antibiotic given within one hour prior to surgical incision. **Among patients who were given prophylactic antibiotics. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 81 2: SAFETY 2: SAFETY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 82 PAT I E N T S A F E T Y • C H A R T 2 : 5 Potentially Inappropriate Prescribing for the Elderly Why is this important? Inappropriate use of medications Implications: These results imply that about 4.7 million in circumstances when risks outweigh benefits community-dwelling elderly individuals received at poses harm to patients and is wasteful of health care least one of 33 potentially inappropriate medications resources. Inappropriate medication use is a particular in 2000, and that about 840,000 received one of  concern among the elderly. The elderly are often more drugs that should always be avoided by elderly patients. physiologically vulnerable and tend to use a greater The downward trend in prescription of potentially number of medications (Kaufman et al. 2002), which puts inappropriate drugs suggests that physicians are heeding them at risk for potentially harmful drug-drug and drug- concerns for more careful prescribing to the elderly. As disease interactions (Zhan et al. 2005). important as minimizing medication overuse is for the To address this problem, experts have developed lists elderly, failing to prescribe recommended medications of medications that are inappropriate to use in older and to provide adequate patient education and medication adults because they may cause harm or have limited monitoring may be even more significant issues, according effectiveness. The most widely used list is the Beers to the Assessing Care for Vulnerable Elders (ACOVE) criteria (Beers 997; Fick et al. 2003). Some evidence Study (Higashi et al. 2004) (see Chart 2:6). suggests that use of these drugs by the elderly can lead to adverse health outcomes and increased use of health care resources (Chin et al. 999; Fick et al. 200; Fu et al. 2004). Another expert panel refined the Beers list to distinguish drugs that should always be avoided in the elderly (Zhan et al. 200). Findings: Applying expert criteria to a national sample of community-dwelling adults revealed that the proportion of elderly ages 65 and older who had used one or more potentially inappropriate drugs declined by more than one-third, from 2.3 percent in 996 to 3.5 percent in 2000. The proportion using drugs that should always be avoided changed little from 996 to 2000, ranging from 2 to 3 percent (AHRQ 2005b). PAT I E N T S A F E T Y • C H A R T 2 : 5 Potentially Inappropriate Prescribing for the Elderly Some prescription drugs should often be avoided in the elderly because they can cause harm or have questionable effectiveness for certain conditions. The proportion of elderly adults who were using one of these drugs declined by one-third from 1998 to 2000. Two to three percent of seniors were taking a drug that experts agree should never be used in the elderly. ������������������������������������������������������������������������������ �������������������������������������������������������������������������������� �� ���� ���� ���� �� ���� ���� �� ���� �� ��� ��� ��� � ����������������������������������������������� ��������������������������������������� ����������������������������� �������������������������������� Sources: 1997 Beers criteria and Zhan expert criteria applied to the Medical Expenditure Panel Survey (AHRQ 2004; 2005b). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 83 2: SAFETY 2: SAFETY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 84 PAT I E N T S A F E T Y • C H A R T 2 : 6 Preventable Adverse Drug Events in Ambulatory Care Why is this important? Research indicates that 90 percent shown) found that 38 percent (578) of the adverse events of all community-dwelling adults ages 65 and older were serious, life-threatening, or fatal; 42 percent of these take at least one medication per week, more than 40 were deemed preventable (Gurwitz et al. 2003). percent use five or more different medications per week, and 2 percent use 0 or more different medications Implications: These findings, although limited, indicate that per week (Kaufman et al. 2002). Given this extensive adverse drug events are a serious problem for seniors in use of medications, adverse drug events (harm from the ambulatory setting. If these results are generalizable medication) are a serious concern. Up to one-quarter to the entire Medicare population, then the authors’ of hospitalized patients experience an adverse drug calculations imply that about one-half million preventable event (Rozich et al. 2003) and about one-third of these adverse drug events occur annually among seniors in events are associated with preventable medication ambulatory care, of which 90,000 may be life-threatening. errors (Kanjanarat et al. 2003). Little is known about the Routine automated monitoring of adverse drug events incidence or preventability of adverse drug events among may become feasible as electronic health records come elderly patients in the ambulatory setting, where the into widespread use. majority of health care is provided. The authors suggest that several interventions might reduce the occurrence of adverse drug events: Findings: Analysis of a group of more than 30,000 Medicare computerized physician order entry with decision beneficiaries cared for in a large multispecialty group support, more systematic decision-making about practice during a 2-month period identified ,523 adverse prescribing and monitoring drugs with known drug events (a rate of 50 per ,000 person-years of potential for adverse events, closer collaboration enrollment). Of these, 28 percent (42) were considered between physicians and clinical pharmacists who preventable by researchers at either the provider or are knowledgeable about drug interactions, and the patient level. About 60 percent of the preventable enhanced patient education and involvement to adverse events were associated with prescription and improve medication adherence (Gurwitz et al. 2003). monitoring errors. More than 20 percent were related to Other research suggests that physicians can mitigate patient adherence such as taking the wrong dose, failing the consequences of adverse drug events by routinely to take prescribed medication, or failing to stop taking asking patients about medication reactions (Weingart medication when instructed. A separate analysis (not et al. 2005). PAT I E N T S A F E T Y • C H A R T 2 : 6 Preventable Adverse Drug Events in Ambulatory Care A year-long study at a large, HMO-affiliated multispecialty group practice identified 1,523 adverse drug events (harm from medication) among 30,000 Medicare beneficiaries served during 1999–2000, of which 421 events were considered preventable because of medication errors of various kinds by health care providers or lack of patient adherence to prescriptions. ��������������������������������������������������������������������������������������������� ����������������������������������������������������������������������������� � �� �� �� �� ��� ������������������ �� ����������������������������������� �� ���������� �� ���������������������������� �� ��������������������� �� ����������������� �� ���������������������������������������� �� ������������������������������������� �� � �� � � � � � � � � � � � � � � �� Source: Incident reports, hospital discharge summaries, medical records analyzed by Gurwitz et al. (2003). Categories do not add because an adverse drug event may have been associated with multiple errors. *Information relating to clinical findings or laboratory results. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 85 2: SAFETY SECTION 3 Access and Timeliness Access and timeliness mean “obtaining needed care and minimizing unnecessary delays in getting that care.” – Institute of Medicine 2001b 3: ACCESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 88 ACCESS AND TIMELINESS • CHART 3:1 Unmet Need and Delay in Seeking Care Why is this important? High-quality health care depends Implications: Individuals are more likely to have unmet on timely access to needed services in an appropriate care health care needs if they are uninsured, lack a usual setting. The Institute of Medicine has defined access to source of care, and have lower income (Shi and Stevens health care as “the timely use of personal health services 2004). Medicare provides near-universal coverage for the to achieve the best possible health outcomes” (IOM 993). elderly. In contrast, 3 percent of middle-age adults are uninsured (NCHS 2004b). The elderly are more likely Findings: In 2002, elderly adults (ages 65 and older) were than middle-age adults to have a usual source of care (see less likely than middle-age adults (ages 45–64) to report Chart 3:4). Access to care for the elderly appears to have that they did not get needed medical care or that they improved since 992, when 0 percent reported that they delayed seeking care because of cost. Specifically, 2.5 delayed seeking care because of cost (FIFARS 2004). percent to 3.6 percent of elderly adults versus 6. percent This survey question addressed general perceptions to 8.3 percent of middle-age adults reported unmet needs of unmet need for medical care. Asking about specific or delayed care-seeking (NCHS 2004b). services and problems may elicit a fuller understanding of unmet needs (see Charts 3:2 and 4:). ACCESS AND TIMELINESS • CHART 3:1 Unmet Need and Delay in Seeking Care Elderly adults (ages 65 and older) are less likely than middle-age adults (ages 45–64) to have unmet medical care needs or to delay seeking care because of cost. �������������������������������������������������������������������������������������������������� �� ���������� �������� �� �� �� ��� � ��� � � ��� ��� � � ������������������������������������������������� ������������������������������������������ �������������������������� ������������������������������� Source: National Health Interview Survey (NCHS 2004b). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 89 3: ACCESS 3: ACCESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 90 ACCESS AND TIMELINESS • CHART 3:2 Financial Barriers to Prescription Adherence Why is this important? The Medicare Modernization Act Implications: Prescription drug coverage alone does of 2003 establishes outpatient prescription drug benefits not protect against high out-of-pocket spending for for Medicare beneficiaries starting in 2006. Historically, medications, nor does it sufficiently defray medication beneficiaries have had to pay for prescription drugs out- costs so that all seniors adhere to their prescription drug of-pocket, or they relied on supplemental coverage to regimens. The situation is even direr for those without defray some of their out-of-pocket expenses. Sources any prescription drug coverage at all. of supplemental prescription drug coverage include Seniors with chronic medical conditions must employer-sponsored retiree health plans, Medicare strictly adhere to their medication regimens for disease managed care plans, and certain Medigap coverage; management. Yet this study found that one-quarter or however, the depth of drug coverage provided by these more of seniors with congestive heart failure, diabetes, programs varies considerably. Many states offer additional or hypertension who lacked prescription drug coverage assistance to beneficiaries to help cover the gaps in did not fill at least one prescription in the previous coverage in the form of state-sponsored Medicaid and year, and up to one-third skipped medication doses to state pharmacy-assistance programs, especially for low- make prescriptions last longer (Kitchman et al. 2002; income seniors and those lacking supplemental coverage. Safran et al. 2002). The national Medicare prescription drug discount Findings: A survey conducted in 200 of community- cards were recently implemented to help ameliorate this dwelling elderly Medicare beneficiaries (ages 65 and access problem. Research such as this study will need to older) residing in eight states found that 4 percent be repeated after the Medicare prescription drug benefit of seniors decided not to fill a prescription because of is implemented in 2006 to measure its effect on reducing cost, 6 percent skipped doses of medicine to make the prescription nonadherence. prescription last longer, and 22 percent reported either type of cost-related nonadherence. Seniors without prescription drug coverage were twice as likely as those with drug coverage to report restricting prescriptions: 35 percent of those without prescription drug coverage, versus 8 percent of those with drug coverage, either did not fill a prescription or skipped doses because of cost (Kitchman et al. 2002; Safran et al. 2002). ACCESS AND TIMELINESS • CHART 3:2 Financial Barriers to Prescription Adherence Lack of patient adherence to prescriptions can lead to adverse health outcomes. In a 2001 survey of elderly Medicare beneficiaries in eight states, those without prescription drug coverage were twice as likely as those with drug coverage to report that they had decided not to fill a prescription or that they had skipped medication doses for financial reasons. �������������������������������������������������������������������������� ������������������������������������������������������������������ ����� ���������������������������������������������� ������������������������������������������� �� �� �� �� �� �� �� �� �� �� �� �� � ��������������������� ����������������������� ��������������������������������� ������������������������������ ����������������������������� ��������������������������� ���������������������������� ������������������������ Source: Kaiser Family Foundation / Commonwealth Fund / Tufts-New England Medical Center, Survey of Seniors (Kitchman et al. 2002; Safran et al. 2002). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 91 3: ACCESS 3: ACCESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 92 ACCESS AND TIMELINESS • CHART 3:3 Financial Barriers to Access Why is this important? The Medicare fee-for-service Implications: Supplemental Medicare coverage promotes benefit package does not cover some important health access to and the use of necessary services, yet care services (e.g., some preventive care; long-term improvement is needed even for those with supplemental care; dental, hearing, and vision services) and has high coverage, especially for preventive care. Recent trends cost-sharing requirements, leaving coverage of these suggest that gaining supplemental insurance coverage expenses up to the beneficiary. To fill in this expense gap, may be more difficult as employer-provided retiree about 90 percent of beneficiaries obtain supplemental coverage declines (Fronstin 2005). This implies that an coverage such as an employer- or union-sponsored retiree increasing proportion of elders may be without access or benefit plan, an individually purchased Medigap plan, or unable to afford to pay for necessary services in the future Medicaid assistance. Alternatively, some beneficiaries join (MedPAC 2002a). Identifying ways to improve access Medicare managed plans that typically provide broader to needed medical care for all beneficiaries and creating benefits than fee-for-service Medicare (MedPAC 2002a). a more efficient system for delivering high-quality care remain challenges for the Medicare program. Findings: An analysis of claims data from 996 to 999 * Only services for which there was a difference are shown on the chart; annual for elderly Medicare fee-for-service beneficiaries (ages physician visits are not shown. Managed care enrollees are not included. 65 and older) found that those without supplemental coverage were less likely than those with supplemental coverage to receive  of 7 measured services that an expert panel had deemed essential for high-quality care.* Services with the largest gaps included mammography every two years for females (27% vs. 62%) and assessment of visual impairment every two years for all elderly individuals (3% vs. 56%). Differences between those with supplemental coverage and those without a supplement were smaller for more highly used services, such as biannual physician visits for patients with congestive heart failure or diabetes (MedPAC 2002a). ACCESS AND TIMELINESS • CHART 3:3 Financial Barriers to Access Elderly Medicare beneficiaries without supplemental insurance are less likely than those who have at least some supplemental coverage to use services that an expert panel ranked necessary for high-quality care. �������������������������������������������������������������������������������������������������������������������� � �� �� �� �� ��� �� ������������������������������������������������������������������� �� ���������������� �������� �� ���������������������������������������������� �� ������������������ �������� �� ��������������������������������������������� �� �� ����������������������������������� �� �� ����������������������������������������������������������������� �� �� �������������������������������������������������������������������� �� �� �������������������������������������������������������������������� �� �� �������������������������������������������� �� �� ����������������������������������������������� �� �� ��������������������������������������������������������������� �� Source: RAND Access to Care for the Elderly Project indicators applied to Medicare Current Beneficiary Survey Cost and Use Files (MedPAC 2002a). Results shown are those for which there was significant difference in receipt by type of coverage, except that annual physician visit is not shown. *COPD = chronic obstructive pulmonary disease. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 93 3: ACCESS 3: ACCESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 94 ACCESS AND TIMELINESS • CHART 3:4 No Usual Source of Health Care Why is this important? The most important benefit of Implications: These findings do not describe whether insurance is to facilitate having a regular source of care patients have established a personal and continuing (Starfield 998). Whether an individual has a regular relationship with a particular physician or clinician at place to go for health care—such as a physician’s office or their usual place of care. Nevertheless, the improvements clinic—is an even more powerful predictor of receiving seen here were likely to have had positive effects on preventive care than whether he or she has health the provision of preventive care and potentially other insurance coverage (Breen et al. 200). Among elderly important services as well. Medicare beneficiaries surveyed in 998, for example, 65 percent of those with a usual source of care received a mammogram versus only 23 percent of those without a usual source of care. Findings: The elderly are more likely than middle-age adults to have a usual source of care. In 2002, only 4.5 percent of those ages 65 to 74 and 2.7 percent of those ages 75 and older had no usual source of care; by comparison, 8.7 percent of those ages 45 to 64 had no usual source of care. The proportion without a usual source of care declined by 2 to 3 percentage points from 993 to 2002 among both elderly and middle-age adults. Among the elderly, the proportion without a usual source of care declined by 8 percentage points among those with Medicare coverage only, from 2 percent in 993 to 4.2 percent in 2002. As a result, the disparity by type of coverage was greatly reduced (NCHS 997a, 997b, 2002, 2004c). ACCESS AND TIMELINESS • CHART 3:4 No Usual Source of Health Care Adults with a usual source of health care are more likely to get recommended preventive care, such as cancer screening, than those without a usual source of care. The elderly are more likely than middle-age adults to have a usual source of care. Between 1993 and 2002, the proportion without a usual source of care declined for both middle-age and elderly age groups. This improvement was especially pronounced among elderly Medicare beneficiaries with Medicare coverage only. ����������������������������������������������� ����������������������������������������������������� ������������������������������������������� �������������������������������������������������������� �� �� ���������� ���������������������� ���������� ���������������������� �������� ���� ���� �������������������������������� �� �� ���� ��� � � ��� ��� ��� ��� ��� ��� ��� ��� � � ��� ��� ��� ��� ��� ��� � � ���� ���� ���� ���� ���� ���� Source: National Health Interview Survey (NCHS 1997a, 1997b, 2002, 2004c). *For 1993, Medicare and Medicaid category includes those with other state -sponsored health plans, including medical assistance programs. Other coverage and uninsured categories are omitted from type of coverage. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 95 3: ACCESS 3: ACCESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 96 ACCESS AND TIMELINESS • CHART 3:5 Waiting Time for Physician Visits for a Specific Illness Why is this important? The Medicare physician payment Implications: The parallel trends in waiting time increases rate reduction of 5.4 percent in 2002 and additional for both elderly and near-elderly patients suggest “that annual rate reductions on the horizon have raised health system developments were much more important concern that access to needed health care for the elderly influences on beneficiary access than any effects of will decline if a growing proportion of physicians are Medicare’s 2002 physician payment rate reduction” unwilling to serve new Medicare beneficiaries. From 997 (Trude and Ginsburg 2005). Whether these increases in to 200, the proportion of physicians who were willing waiting times correlate with any changes in the clinical to accept all new Medicare patients into their practice quality of health care for patients is not known. The decreased from 75 percent to 7 percent (Trude and authors noted that although waiting times have increased, Ginsburg 2005). patient complaints about delayed care did not increase proportionally. “Presumably, patients now expect longer Findings: In a nationally representative survey, both elderly waits for appointments and no longer consider these patients (ages 65 and older) and near-elderly patients longer waits as delaying care,” the authors write (Trude (ages 55 to 64) waited almost nine days on average in 2003 and Ginsburg 2005). This kind of data deserves continued to see a physician for a specific illness. Waits were longer monitoring to determine whether these trends continue to see a specialist than to see a primary care doctor. and what effect they may have on patient experience and Compared to the near-elderly, the elderly waited one clinical quality of care. day longer to see a primary care physician (6.2 days vs. 5 days), but they waited two days fewer to see specialty physicians (2.5 days vs. 4.5 days). Both groups waited longer in 2003 than they did in 997; this overall increase was attributable mainly to increases of more than two days in waiting times to see specialist physicians (Trude and Ginsburg 2005). ACCESS AND TIMELINESS • CHART 3:5 Waiting Time for Physician Visits for a Specific Illness From 1997 to 2003, the average waiting time to see a physician for a specific illness increased for both elderly adults (ages 65 and older) and near-elderly adults (ages 55 to 64), primarily because of an increase in time to see specialists. In 2003, elderly and near-elderly adults waited the same amount of time overall. The elderly waited about one day longer than near-elderly adults to see a primary care physician, but near-elderly adults waited two days longer than elderly adults to see a specialist. ������������������������������������������������������������������� ����������������������������������������������������� �� ���� ���� �� ���� ���� ���� �� ���� ��� ��� � ��� ��� ��� ��� ��� ��� � � ������������ ������� ������������ ������� ������������ ������� �������������������� ����������������������������� ��������������������������� Source: Center for Studying Health System Change, Community Tracking Study ( Trude and Ginsburg 2005). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 97 3: ACCESS 3: ACCESS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 98 A C C E S S A N D T I M E L I N E S S • C A R E AT E N D O F L I F E • C H A R T 3 : 6 Use of Hospice at End of Life Why is this important? Hospice is a patient- and family- Implications: Increasing use of hospice among Medicare centered concept of health care for the terminally ill that beneficiaries may reflect better understanding of the goals aims to maintain the comfort of the dying person, rather of hospice. Hospice is used by about 60 percent of those than seek a cure for the illness. Hospice is not a place but who die of cancer, but hospice use increased the most an approach to care that frequently allows the terminally among those with other life-threatening chronic illnesses ill to be cared for at home, where most people say they such as heart disease and Parkinson’s disease (MedPAC would prefer to die (Tang 2003). A multidisciplinary 2002c). The Medicare Modernization Act of 2003 includes hospice care team provides home visits, on-call provisions that may increase the use of hospice, including professional health care, teaching and emotional support coverage for a one-time consultation session to evaluate for the family, pain management, and spiritual care for a patient’s eligibility and need for hospice care (HCFO the patient. Since 983, Medicare has covered hospice care 2004). for beneficiaries whose doctors certify that they have a life expectancy of six months or less. Understanding trends in hospice care will become more important with the growing elderly population (MedPAC 2004b). Findings: Hospice use among Medicare beneficiaries increased by 9 percentage points from 998 to 2002. The increase was greatest among the oldest. Those in managed care plans were more likely to use hospice services at the end of life than those in traditional, fee-for-service Medicare. Among Medicare fee-for-service beneficiaries in 998, hospice use declined with increasing age, but by 2002, hospice use was similar across all age groups (MedPAC 2004b). A C C E S S A N D T I M E L I N E S S • C A R E AT E N D O F L I F E • C H A R T 3 : 6 Use of Hospice at End of Life Hospice use increased among Medicare beneficiaries from 1998 to 2002 but especially among the oldest beneficiaries. Those in managed care plans were more likely to use hospice services at the end of life than those in traditional Medicare. ��������������������������������������� ����������������������������������������������� ������������������������������������������ ������������������������������������������������������� ���� ���� ���� ���� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � � �������� �������� ���������� ���������� ���������� �������� ��������������� ������������ Source: Medicare Payment Advisory Commission (2004b) analysis of Medicare administrative data. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 99 3: ACCESS SECTION 4 Patient and Family Centeredness Patient and family centeredness refers to “health care that establishes a partnership among practitioners, patients, and their families (when appropriate) to ensure that decisions respect patients’ wants, needs, and preferences and that patients have the education and support they need to make decisions and participate in their own care. – Institute of Medicine 2001b 4 : PAT I E N T / FA M I LY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 102 PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C H A R T 4 : 1 Experiences with Insurance and Care Why is this important? Medicare is the United States’ only Implications: These results reflect elderly Medicare national social health insurance program, covering 4 beneficiaries’ overall experiences with insurance million Americans including 35 million elderly. When and health care, including coverage provided by the the Medicare program was created in 965, its structure Medicare program and any supplemental insurance was modeled on the dominant approach to private they may have had. Whatever the relative contribution, insurance and fee-for-service health care at that time. the combination appeared to result in a more positive Approaches to health care delivery and private insurance experience for Medicare beneficiaries than that reported have since diversified, raising questions about whether by privately insured nonelderly adults. The elderly Medicare offers good value and has adequately evolved had more positive experiences despite their higher to be effective in service delivery and responsive to prevalence of poor health and low income compared to public expectations. the privately insured nonelderly. The study authors speculated that differences Findings: In a national survey of adults ages 9 and older in perceptions might relate to factors such as plan conducted in 200, elderly Medicare beneficiaries (ages administration, choices, and benefit structures under 65 and older) were more likely than privately insured Medicare and private insurance (Davis et al. 2002). nonelderly adults (ages 9 to 64) to rate their health Understanding what aspects of the Medicare program are insurance coverage as very good or excellent and to report working well for beneficiaries is important to preserving that they were very satisfied with the care they received. the best of Medicare for the future. In contrast, privately insured nonelderly adults were more likely to report coverage problems with their insurance and that they did not seek or receive medical care in the past year because of costs (Davis et al. 2002). PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C H A R T 4 : 1 Experiences with Insurance and Care Compared to privately insured nonelderly adults, elderly Medicare beneficiaries were more likely to rate their insurance highly and to be satisfied with their care, and were less likely to report problems with coverage and access to care. Elderly respondents’ ratings of insurance reflect their experiences with the Medicare program and any supplemental coverage. ������������������������������������������������������������������������������������������������������ ��� ������������������������������ ��������������������������� �� �� �� �� �� �� �� �� �� �� �� �� � ���������������������� ���������������������������� ������������������������� ������������������������� ���������������������� ����������������������� ������������������������������� ���������������� ��������������� Source: Commonwealth Fund 2001 Survey of Health Insurance (Davis et al. 2002). *Any of the following responses: plan did not pay anything for care that respondent thought was covered; plan covered only a part of service; reached limit on what plan paid for specific illness or injury. **Any of the following responses: did not fill prescription; did not get needed specialist care; skipped recommended test or follow-up; had a medical problem but did not visit doctor or clinic. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 103 4 : PAT I E N T / FA M I LY 4 : PAT I E N T / FA M I LY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 104 A C C E S S A N D T I M E L I N E S S / PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C H A R T 4 : 2 Consumer Assessment of Health Plans Why is this important? “Since many older Americans Implications: The comparative data shown in the chart are suffer from one or more chronic health problems, it is consistent with the findings of prior studies. For example, especially important for them to understand their health an analysis of 2000 and 200 CAHPS data found that care options and make informed choices about health elderly fee-for-service Medicare beneficiaries “generally insurance coverage” (AHRQ 2002c). When selecting rated their care and physicians higher and reported fewer among health care coverage arrangements, consumers are problems obtaining needed care than did [Medicare often most interested in learning about the experiences managed care] enrollees. In contrast [Medicare managed that other people like themselves have had with these care] enrollees reported receiving recommended options (KFF/AHRQ 996). In response, the federal preventive services...more frequently and reported fewer government sponsored development of the Consumer problems related to paperwork and information” (Landon Assessment Health Plans Study (CAHPS) survey “to et al. 2004). The same study found that results varied help consumers identify the best health care plans and geographically and that variation among competing services for their needs” (AHRQ 998). The federal managed care plans was as great as the overall difference Centers for Medicare and Medicaid Services uses CAHPS between managed care plans and fee-for-service to report comparative information on the experiences Medicare. This means that it is important for Medicare of beneficiaries in the original Medicare fee-for-service beneficiaries to examine the performance of particular program and Medicare managed care plans, at both the Medicare health plans available in their local market. national and local levels. Findings: As of 2003, Medicare fee-for-service beneficiaries were somewhat more likely than those in managed care plans to give high ratings to their plan, doctor, and care. In contrast, Medicare managed care plan members were somewhat more likely to report that they had received a recent flu shot. Differences between average ratings for Medicare fee-for-service and for Medicare managed care were small, ranging from 2 to 5 percentage points among six publicly reported CAHPS measures (CMS 2005c). A C C E S S A N D T I M E L I N E S S / PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C H A R T 4 : 2 Consumer Assessment of Health Plans Medicare fee-for-service beneficiaries were somewhat more likely than those in managed care plans to give high ratings to their plan, doctor, and care in 2003. In contrast, Medicare managed care plan members were somewhat more likely to report that they had received a flu shot. ����������������������������������������������������������������������������������������� � �� �� �� �� ��� �� ���������������������������������������������� �� �� ����������������������������������� �� �� ����������������������������������������� ����������������� �� ��������������� �� ���������������� ���������������������������������� ���������� �� �� �������������������������������������������������������� �� �� ���������������������������������������������� �� Source: Center for Medicare and Medicaid Services, Medicare Health Plan Compare Web site, Consumer Assessment of Health Plans Survey (CMS 2005c). Rates are case -mix adjusted to control for differences in respondents’ age, education, and self-reported health status, and in whether respondents had assistance answering the survey. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 105 4 : PAT I E N T / FA M I LY 4 : PAT I E N T / FA M I LY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 106 PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C H A R T 4 : 3 Interpersonal Quality of Care Why is this important? The quality of communication Implications: The fact that seniors gave higher ratings between patients and their health professionals may to interpersonal quality of care is encouraging given affect patients’ receptivity to receiving advice, their that they often have more complex needs than younger adherence to treatment regimens, and their satisfaction adults. Interpersonal deficits in care might account with and outcomes of care (Stewart 995; Stewart et for some of the perception of inadequate time spent al. 2000). The amount of time that patients spend with the physician (Gross et al. 998). Several types of with clinicians may affect their ability to fully explain interventions directed at both physicians and patients their personal needs and to raise questions about their might be effective in improving physician-patient diagnosis and treatment. interactions and patient outcomes, such as: • education and incentives for health professionals Findings: Among seniors who visited a doctor’s office during and their staff to help improve patient-centered 200, two-thirds reported that the doctor or other health communication skills (Lewin et al. 200); professional always listened carefully and showed respect. • culturally relevant questionnaires, written and audiovisual More than half reported that health professionals always materials, and coaching or skills training to help prepare explained things carefully and spent enough time with patients (and their family members) for effective health them. Seniors gave somewhat better ratings than middle- care encounters (Cegala et al. 200; Post et al. 2002); age adults on three of four measures of interpersonal • interpreter services and teams of professionals that aspects of care (AHRQ 2005b). include at least one bilingual professional to overcome language barriers (Brach and Fraser 2000); • use of mid-level practitioners (physician assistants and nurse practitioners) to increase time spent with patients during intake and follow-up care (Berry et al. 2003); and • follow-up services such as telephone calls to determine how the patient is doing post-care (Car and Sheikh 2003). PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C H A R T 4 : 3 Interpersonal Quality of Care About two-thirds of seniors reported that their health professionals always listened carefully and showed respect and more than one-half reported that health professionals always explained things well and spent enough time with them. Seniors gave somewhat better ratings than middle-age adults on three of these four measures of interpersonal aspects of care. �������������������������������������������������������������������������������������������������������������������� ������������������ ������� ������ ��� � � � � � � � �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � ���������� �������� ���������� �������� ���������� �������� ���������� �������� ����������������� ���������������� ����������������� ����������������� ������������������ ������������������������ �������������� ����������������� Source: Medical Expenditure Panel Survey (AHRQ 2005b). Numbers may not add to 100 because of rounding. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 107 4 : PAT I E N T / FA M I LY 4 : PAT I E N T / FA M I LY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 108 PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C H A R T 4 : 4 Beneficiary Knowledge of Medicare and Accuracy of Medicare Information Why is this important? The Balanced Budget Act of 997 A separate audit conducted by the Government required that the federal Centers for Medicare and Accountability Office in 2004 found that only six of every Medicaid Services (CMS) provide Medicare beneficiaries 0 calls to the -800-MEDICARE beneficiary help line with educational materials to help them understand were answered accurately. The auditors concluded that the Medicare program and their coverage options. In customer service representatives “provided inaccurate response, CMS designed a National Medicare Education information largely because they did not always Program that uses multiple communication channels understand enough about the Medicare program to including printed materials, a toll-free telephone access a script that answered the question or could not information line, Internet sites, and training and support clearly explain the material in the script that they were for information intermediaries such as state health using” (GAO 2004a). insurance assistance programs. A reference handbook called Medicare & You was mailed to all Medicare Implications: Although beneficiary education improved beneficiary households in 999 (following a five-state between 998 and 2002, less than half of Medicare pilot conducted in 998) and continues to be mailed to all beneficiaries felt that they had all the information they newly enrolled beneficiaries monthly (Goldstein 200). needed. CMS reports that it is intensifying its educational efforts to prepare beneficiaries for changes brought Findings: The proportion of elderly Medicare beneficiaries about by the Medicare Modernization Act of 2003. The who have all the information about Medicare they say agency hired more customer service representatives and they need has increased since Medicare enhanced its conducted additional training to improve call accuracy educational efforts, from 35 percent in 998 to 46 percent and active listening. It is partnering with Medicare in 2002. Likewise, the proportion who say they have consumer organizations and nonprofit community little or none of what they needed to know declined organizations that provide advice and counseling, from 36 percent in 998 to 24 percent in 2002. Minority focusing especially on low-income beneficiaries and beneficiaries and those with lower incomes and less their caregivers. A Regional Education About Choices education were less likely to report that they knew all or in Health (REACH) campaign is providing culturally most of the Medicare information they needed (data not and linguistically appropriate information to those who shown) (CMS 2000a, 2004a). may not receive information through traditional media channels (CMS 2005e). PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C H A R T 4 : 4 Beneficiary Knowledge of Medicare and Accuracy of Medicare Information The proportion of elderly Medicare beneficiaries who have the information about Medicare they say they need has increased somewhat since Medicare enhanced its educational efforts, although more than one-half do not yet have the information they need. Only six of every 10 calls to the Medicare information line were answered accurately in a 2004 government audit. ������������������������������������ ������������������������������������������������� ���������������������������������������������� ����������������������������������������������� ������������������������������������� ������������������������������������ ��� ���������� ������������� �� ��������������� �� �� ������������ �������� �� �� ������������� ������������� ������������ �� ������� �� ������������ �������� �� ��������������� �� ������������ �� ������� � ���� ���� Sources: Authors’ calculations using Medicare Current Beneficiary Survey, Access to Care File results published by the Centers for Medicare and Medicaid Services (CMS 2000a, 2004a); Government Accountability Office audit (GAO 2004a). *Responses represent all community-dwelling (noninstitutionalized) Medicare beneficiaries. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 109 4 : PAT I E N T / FA M I LY 4 : PAT I E N T / FA M I LY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 110 PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C A R E AT E N D O F L I F E • C H A R T S 4 : 5 T O 4 : 7 Quality of Care at End of Life Why is this important? As medical care has changed over at home with hospice care were more likely to rate the the last century, deaths today are more likely to occur in overall quality of care as excellent (7% vs. 42% to 47%; health care institutions than in individuals’ homes. In not shown) (Teno et al. 2004). response to concerns about the quality of care at the end of life, a 997 Institute of Medicine report recommended Implications: Family perceptions of the quality of end-of-life that health care stakeholders should collaborate to care raise concerns about how well the health care system strengthen methods for measuring the quality of care for is meeting expectations of patients and their families for dying patients and their families (IOM 997). A synthesis “death with dignity.” Measuring these facets of end-of-life of research suggests that patient- and family-centered care on an ongoing basis at the national and individual end-of-life care involves providing dying individuals provider levels would help to identify and monitor with desired physical comfort and emotional support, progress at efforts for improvement. Data are needed to supporting shared decision-making, treating the dying determine whether and how these types of perceptions person with respect, providing emotional support to might vary among racial and ethnic groups. The study family members, and coordinating care across settings authors concluded that these “results call for a public (Teno et al. 200). health approach that uses sustained and multifaceted interventions to improve end-of-life care in the United Findings: In a nationally representative study, 5 percent to States” (Teno et al. 2004). Increasing access to hospice 50 percent of family members expressed concerns about care and other palliative care programs at end of life are some aspects of the care delivered at the end of life to a two possible approaches (see Charts 3:6 and 6:8). relative who died in 2000 from chronic illness (average age 74). Inadequate emotional support for the patient and family were the most often-cited concerns across all settings and types of care. Family members of patients who died at home with hospice care were less likely to report concerns than were family members of patients who died with other care arrangements or in other settings. Moreover, family members of patients who died PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C A R E AT E N D O F L I F E • C H A R T 4 : 5 Quality of Care at End of Life: Part I Up to one-half of family members expressed concerns about some aspects of the care delivered to a deceased relative at the end of life. The issue eliciting the greatest concern was emotional support for the patient and family. ������������������������������������������������������������������������� �������������������������������������������������������������������������� � �� �� �� �� ��� ������������������������������������������������������� �� ��������������������������������������� �� ���������������������������������������������� �� ��������������������������������������� �� ��������������������������������������������� �� ���������������������������������������������������� �� ���������������������������������������������������������� �� ������������������������������������������� �� �������������������������������������������������������� �� Nationally representative mortality follow-back telephone survey (N = 1,380 decedents who used health care services at end of life; average age 74) ( Teno et al. 2004). Last place of care is the place where the decedent spent more than 48 hours prior to death. *Information regarding what to expect while patient was dying. **Among those who had contact with a physician. ***Dyspnea is difficulty breathing. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 111 4 : PAT I E N T / FA M I LY 4 : PAT I E N T / FA M I LY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 112 PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C A R E AT E N D O F L I F E • C H A R T 4 : 6 Quality of Care at End of Life: Part II Family members generally had fewest concerns for patients who died at home with hospice care. ������������������������������������������������������������������������� ��� �������������������������������������������������������������������������� ������������������� ���������������������� ������������������� ������������ �������� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � � ���������������� ���������������� ��������������������������� ���������������������� ���������������������� ���������������������� ������������������������ �������������������� ����������������� ������������������ �������������������������������������� ��������������� Nationally representative mortality follow-back telephone survey (N = 1,380 decedents who used health care services at end of life; average age 74) ( Teno et al. 2004). Last place of care is the place where the decedent spent more than 48 hours prior to death. *Dyspnea is difficulty breathing. PAT I E N T A N D FA M I LY C E N T E R E D N E S S • C A R E AT E N D O F L I F E • C H A R T 4 : 7 Family Ratings of Quality of Care at End of Life: Part III Family members generally had fewest concerns for patients who died at home with hospice care. ������������������������������������������������������������������������� ��� �������������������������������������������������������������������������� ������������������� ���������������������� ������������������� ������������ �������� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � � � �������������� �������������� �������������� �������������� ����������������� ���������������� ����������� ����������������� �������������� ���������������������� ������� ������������ �������������� ���������������� ��������������� ������������ ���������������������� ����������������� Nationally representative mortality follow-back telephone survey (N = 1,380 decedents who used health care services at end of life; average age 74) ( Teno et al. 2004). Last place of care is the place where the decedent spent more than 48 hours prior to death. *Information about what to expect while patient was dying. **Among those who had contact with a physician. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 113 4 : PAT I E N T / FA M I LY SECTION 5 Equity Equity means “providing care that does not vary in quality because of personal characteristics such as gender, ethnicity, geographic location, and socioeconomic status.” – Institute of Medicine 2001a 5: EQUITY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 116 E Q U I T Y • PAT I E N T S A F E T Y • C H A R T 5 : 1 Racial and Ethnic Disparities in Adverse Events and Complications of Care Why is this important? A substantial body of research has Implications: Disparities in patient safety are disturbing documented that racial and ethnic minority Americans to reasonable expectations that the health care system are worse off than white Americans on a wide variety should provide a basic level of safety for all. Additional of indicators of health care access and quality that research is warranted to determine how much the racial determine health outcomes (IOM 2003). Understanding and ethnic variation documented here results from the incidence and nature of adverse events in different differing care within the same institution as opposed racial and ethnic groups might encourage health care to differences between institutions that may primarily organizations to investigate their own performance and serve blacks, Asians, and Hispanic Americans. Chart develop strategies for improvement. This chart focuses 5:0 documents that black patients are predominantly on three adverse events or complications of care in the seen by a small number of physicians who report hospital (as measured using Patient Safety Indicators relatively higher levels of constraints in their ability to applied to hospital billing records) that can often be deliver high-quality care. Similar constraints might act prevented with good medical and nursing care (see Chart as barriers to high-quality hospital care for black and 2:3 for more detailed discussion). possibly Hispanic patients. Findings: Among elderly patients (ages 65 and older) hospitalized in 200: • Black patients were 72 percent more likely than white patients to develop infections related to intravenous lines and catheters, 7 percent more likely to suffer blood clots in their legs or lungs following surgery, and 2.3 times more likely to develop pressure sores. • Hispanic patients were 72 percent more likely than white patients to develop infections related to intravenous lines and catheters and 36 percent more likely to develop pressure sores. • Asian/Pacific Islander patients were 44 percent more likely than white patients to develop infections related to intravenous lines and catheters but less likely to suffer blood clots and pressure sores (AHRQ 2005a). E Q U I T Y • PAT I E N T S A F E T Y • C H A R T 5 : 1 Racial and Ethnic Disparities in Adverse Events and Complications of Care Some adverse events or complications of care can often be prevented with good medical and nursing care. Compared to white elderly patients, minority elders were more likely to acquire infections in the hospital. Black patients were more likely than white patients to suffer blood clots in their legs or lungs following surgery. Black and Hispanic patients were more likely than whites to develop pressure sores. Asian/Pacific Islander patients were less likely to suffer blood clots and pressure sores. ������������������������������������������������������������������ �������������������������������������������������������������������������������� ������������������ ������������������ ���������������������������������� �������� ��� ��� ��� ��� ��� ��� ��� ��� ��� ��� �� �� ��� �� �� �� �� �� � ������������������������������� �������������������������������� ������������������������������������ ������������������������� Source: Patient Safety Indicators applied to Health Care Utilization Project State Inpatient Database (AHRQ 2005a). Rates exclude complications present on admission and are adjusted for age, gender, age -gender interactions, comorbidities, and diagnosis-related group clusters. *Infections primarily related to intravenous lines and catheters. **Among surgical patients. ***Among patients with hospital stays of five days or longer. See the Technical Appendix for specific exclusions. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 117 5: EQUITY 5: EQUITY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 118 E Q U I T Y • S TAY I N G H E A LT H Y • C H A R T S 5 : 2 T O 5 : 4 Disparities in Preventive Care Why is this important? The provision of effective preventive Implications: These results were not adjusted for services is a fundamental aspect of high-quality health confounding and must be interpreted with caution. For care. Disparities in the receipt of preventive care may example, those without supplemental coverage are likely perpetuate disparities in both short- and long-term to have lower income and vice versa. Socioeconomic health outcomes. Understanding the factors giving rise to factors may have a larger influence on disparities in the disparities in health care can help to identify root causes receipt of preventive care than race or ethnicity alone that are amenable to change by health care professionals (Fiscella et al. 2000). Interventions can be targeted or that require changes in wider public policies affecting to address access barriers, but cost barriers are more health care. difficult to address without financial assistance of some kind. Almost two of five elderly Americans (38%) live in Findings: Among community-dwelling elderly adults poverty or near-poverty. The poorest beneficiaries are surveyed during 998, 2000, and 200 (AHRQ 2005a): dually eligible for Medicare and Medicaid. Other low- • Chart 5:2 – By Race and Ethnicity: minorities were less income Medicare beneficiaries may qualify for assistance likely than whites to receive some preventive services through the Medicare Savings Program. However, only such as immunizations and colorectal cancer screening, about 60 percent of Medicare beneficiaries who are but rates of care were similar for other services such as eligible for one of these programs are actually enrolled mammography and blood pressure testing. (Williams 2004). • Chart 5:3 – By Family Income: those with higher family income were generally more likely to receive preventive services than those with lower family income. This income disparity in quality was greatest for cancer screening tests, intermediate for vaccination, and least for tests for cardiovascular disease risk factors. • Chart 5:4 – By Type of Coverage: those with private supplemental coverage were generally more likely to receive preventive services than those with Medicaid or no supplemental coverage. There was no clear pattern except that disparity was smallest for blood pressure reading, which is routinely done during physician visits. E Q U I T Y • S TAY I N G H E A LT H Y • C H A R T 5 : 2 Disparities in Preventive Care by Race and Ethnicity Among elderly adults, minorities were often but not always less likely than whites to receive preventive care. Disparities varied by type of service and were generally smaller for services with the highest overall rates of use. ������������������������������������������������������������������������������������������������������� ������������������ ������������������ ����� �������� ��� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � �������� ������������ ������������ ���������������� ����������� �������������� �������������� ���������������� ������������ ���������������� ���������������� ���������������� ���������������� ������ ������������ ������ �������������� ��������������� Source: National Health Interview Survey (AHRQ 2005a). Numbers were too small to report mammograms for Asians or any measure for Native Americans. *Blood pressure checked and respondent can state whether it is normal or high. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 119 5: EQUITY 5: EQUITY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 120 E Q U I T Y • S TAY I N G H E A LT H Y • C H A R T 5 : 3 Disparities in Preventive Care by Family Income Level Elderly adults with higher family income were generally more likely to receive recommended preventive services than those with lower family income. The disparity was greatest for cancer screening tests and was least for tests for heart and circulatory disease risk factors. ������������������������������������������������������������������������������������������������������� ���������������� ������������������������� ���������������������� ����������������� ��� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � �������� ������������ ������������ ���������������� ����������� �������������� �������������� ���������������� ������������ ���������������� ���������������� ���������������� ���������������� ������ ������������ ������ �������������� ��������������� Source: National Health Interview Survey (AHRQ 2005a). *Blood pressure checked and respondent can state whether it is normal or high. FPL = federal poverty level. E Q U I T Y • S TAY I N G H E A LT H Y • C H A R T 5 : 4 Disparities in Preventive Care by Type of Coverage Elderly Medicare beneficiaries with private supplemental coverage were generally more likely to receive preventive services than those with public supplemental coverage (such as Medicaid) or no supplemental coverage. ������������������������������������������������������������������������������������������������������� ����������������������������� ���������������������������������� ���������������������� ��� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � �������� ������������ ������������ ���������������� ����������� �������������� �������������� ���������������� ������������ ���������������� ���������������� ���������������� ���������������� ������ ������������ ������ �������������� ��������������� Source: National Health Interview Survey (AHRQ 2005a). *Blood pressure checked and respondent can state whether it is normal or high. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 121 5: EQUITY 5: EQUITY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 122 E Q U I T Y • S TAY I N G H E A LT H Y • C H A R T 5 : 5 Impact of Medicare Coverage on Receipt of Preventive Care Why is this important? Thirteen percent of near-elderly at risk of heart disease, revealed a 29 percent reduction adults (ages 55 to 64) did not have health insurance in in the gap in cholesterol screening rates between the 2003 (U.S. Census Bureau 2005). Uninsured adults are previously uninsured and insured groups once they less likely than those with health insurance to obtain became eligible for Medicare coverage, as compared to an recommended preventive care (Ayanian et al. 2000). 8 percent reduction in the gap among those with neither Medicare coverage helps improve elderly adults’ access condition (McWilliams et al. 2003). to health care (see Chart 4:). Over time, Medicare has covered more preventive care services. For example, Implications: Medicare coverage reduced the gaps in Medicare began covering screening mammography preventive services between previously uninsured and in 99, subject to patient cost-sharing (GAO 2002a). insured adults. Rates of non-recommended screening Cholesterol testing became a Medicare-covered benefit (prostate exam) increased along with recommended for all beneficiaries in 2005 (CMS 2004b); previously, this screening. The study authors speculated that test was covered for cardiovascular risk assessment among socioeconomic factors might contribute to remaining those with hypertension and diabetes (McWilliams et al. disparities (McWilliams et al. 2003). Adults with 2003). Although Medicare has covered prostate cancer diabetes and hypertension, who are generally in need screening (PSA test or digital rectal exam) in men since of cardiovascular risk reduction, particularly benefited 2000, these tests are not currently recommended by the from health insurance coverage for cholesterol testing. U.S. Preventive Services Task Force because of insufficient Starting in 2005, Medicare began covering a “Welcome to evidence to determine their effectiveness (USPSTF 2002e). Medicare” physical exam for newly eligible beneficiaries, which may further enhance the benefit of gaining Findings: A national survey of adults before and after Medicare coverage for uninsured individuals. Although they became eligible for Medicare at age 65 showed that insurance coverage promotes access to preventive previously uninsured near-elderly adults had significantly screening, coverage alone is not enough to assure high- lower rates of preventive screening than insured near- quality care. elderly adults; once the uninsured became eligible for Medicare coverage, the disparities in rates of screening were reduced by half or more. A subanalysis (not shown) of individuals with diabetes or hypertension, who are E Q U I T Y • S TAY I N G H E A LT H Y • C H A R T 5 : 5 Impact of Medicare Coverage on Receipt of Preventive Care After older adults became eligible for Medicare at age 65, existing disparities in screening between those who were previously insured and those who were previously uninsured were greatly reduced, but not eliminated. Screening increased for tests that are recommended based on evidence for their effectiveness, such as cholesterol testing and mammography, and for services that have not been proven effective at improving health outcomes, such as prostate exams. ����������������������������������������������������������������������������� ����������������������������������������������������������������������������� ��� ������������������������������������������������ ����������������������������������������������� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � ����������� ����������� ������������� ����������� ����������� ������������� ������� ������� ����� ������� ������� ����� �������������������������� ������������������������ Source: Health and Retirement Study (McWilliams et al. 2003). *Results are shown only for individuals who were continuously uninsured in both 1994 and 1996 or continuously insured in both 1994 and 1996. Results are not shown for those who were intermittently uninsured (uninsured in 1994 or 1996 but not both). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 123 5: EQUITY 5: EQUITY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 124 EQUIT Y • LIVING WITH ILLNESS • CHART 5:6 Racial and Ethnic Disparities in Chronic Care Management Why is this important? The Report of the Secretary’s Task Implications: The magnitudes and patterns of chronic Force on Black and Minority Health (Nickens 986) first and mental health care vary between different racial drew attention to disparities in access to medical care for and ethnic groups. Continued measurement of such minority Americans. Numerous studies emerging since disparities is important to help guide interventions to that time continue to document minority disparities in ensure equity in access, use, and outcomes across all racial health care, primarily differences between blacks and and ethnic groups. Insight as to why these disparities exist whites, and the phenomenon is best documented in the in the first place is also sorely needed. One study found Medicare fee-for-service program in which Americans that ethnic disparities in care were explained largely ages 65 and older receive basic health care coverage by differences in English fluency, but racial disparities (Gornick 2000). Less research has focused on access and in care were not explained by commonly used access quality for racial and ethnic populations other than blacks factors (Fiscella et al. 2002). Because care is suboptimal and whites, the quality of mental health care received for all groups, quality improvement efforts provide the by racial and ethnic minorities, or the quality of care opportunity to achieve the twin goals of equity and received by Medicare beneficiaries enrolled in managed effectiveness for all populations. care plans. Findings: Among Medicare beneficiaries ages 65 and older enrolled in managed care plans during 999, blacks consistently received poorer quality of care than whites across all quality measures studied. All minorities received worse quality of mental health care than white patients. For other measures, Asians received equal or better care. Hispanics and Native Americans (when their numbers were large enough to report) were less likely than whites to receive some care but were equally or more likely to receive other types of care (Virnig et al. 2002; Virnig et al. 2004). (Only a subset of measures are shown on the chart but other measures showed similar patterns.) EQUIT Y • LIVING WITH ILLNESS • CHART 5:6 Racial and Ethnic Disparities in Chronic Care Management Among Medicare beneficiaries enrolled in managed care plans, blacks were less likely than whites to receive recommended chronic care services and achieve good outcomes. Hispanics, Asian Americans, and Native Americans were less likely than whites to receive some services but equally or more likely to receive other services or achieve good outcomes. ��������������������������������������������������������������������������������������������������� ����� ����� ����� �������� ��������������� ��� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� � ������������� ������������������� ������������������� �������������������� �������������������� ������������������ ����������� ���������� �������������� ��������������������� ���������� ������������������ Source: Analysis of HEDIS by Virnig et al. (2002; 2004). Numbers for Native Americans were too small to report for some measures. “Other ” race omitted for clarity. *Those newly diagnosed with depression, prescribed an antidepressant, and who continued using an antidepressant during the 12-week acute -treatment phase. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 125 5: EQUITY 5: EQUITY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 126 E Q U I T Y • C A R E AT E N D O F L I F E • C H A R T 5 : 7 Racial and Ethnic Disparities in Use of Hospice at End of Life Why is this important? Hospice is a multidisciplinary Implications: Cultural issues related to death and dying approach to health care for the terminally ill that aims present unique challenges to offering hospice care to to maintain comfort of dying patients while they are diverse populations (Lorenz et al. 2004a; Lyke and Colon cared for at home (see Chart 3:6). Hospice care generally 2004). Low-income, urban African Americans and provides symptom management, pain control, spiritual Latinos report barriers to hospice care related to lack of care, and family support. In the U.S. health care system, awareness, language differences (for Latinos), general disparities related to race, ethnicity, and socioeconomic mistrust of the health care system, and the overall cost status are pervasive (IOM 2003). Disparities in hospice of health care (Born et al. 2004). On the other hand, enrollment may mimic those general disparities because they are receptive to assistance with end-of-life care that of cultural differences or practical obstacles such as provides relief for patients and caregivers and emphasizes language differences (Lorenz et al. 2004a). spirituality and family. Addressing the disparities in hospice care among Findings: Hospice use at end of life increased among all minority Medicare beneficiaries is challenging, but it Medicare fee-for-service beneficiaries between 998 and presents an urgent need for improving end-of-life care 2002, but it increased most among white beneficiaries. for diverse patients. Tailoring hospice services to reduce Specifically, hospice use increased 9 percentage points barriers may increase use and improve satisfaction. among whites but only 4 to 6 percentage points among Expert recommendations include hiring a diverse staff, black, Asian, and Hispanic Americans during this time. providing cross-cultural training programs, offering As a result, minorities remain less likely than whites to translation services and diverse spiritual care, and using use hospice (MedPAC 2004b). linguistically and culturally specific outreach materials (Lorenz et al. 2004a; Lyke and Colon 2004). Physicians also have an important role in discussing hospice as an option for terminally ill patients to consider. E Q U I T Y • C A R E AT E N D O F L I F E • C H A R T 5 : 7 Racial and Ethnic Disparities in Use of Hospice at End of Life Hospice use at end of life increased among all Medicare beneficiaries over the past decade, but increased most among white beneficiaries. As a result, minorities remain less likely than whites to use hospice. ����������������������������������������������������������������������������������� �� ����� ����� ����� �������� �� �� �� �� �� �� �� �� �� �� �� �� � ���� ���� Source: Medicare Payment Advisory Commission (2004b) analysis of Medicare administrative data. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 127 5: EQUITY 5: EQUITY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 128 E Q U I T Y A N D E F F I C I E N C Y • C A R E AT E N D O F L I F E • C H A R T 5 : 8 Unexplained Variation in Care at End of Life Why is this important? Medicare per capita spending diseases at a given hospital. These findings suggest that among regions is closely correlated with use of hospitals, the hospital where patients are treated—rather than the intensive care units, and physician services in managing nature of their illness—dictates the amount of care they chronic illness such as congestive heart failure (CHF), receive (Wennberg et al. 2004b). chronic obstructive pulmonary disease (COPD), and solid tumor cancers (Wennberg et al. 2004a). To preserve Implications: Medicare claims can be used to measure Medicare’s ability to provide universal access for seniors, population-based, provider-specific use of resources delivery of effective care in an efficient manner is for patients enrolled in traditional fee-for-service paramount. To that end, health care institutions require Medicare. Strikingly wide differences between hospitals information on their performance over time to assess exist with regard to the amount of care provided to their efficiency and to identify areas where intervention chronically ill patient populations, and more care does is needed. Numerous quality indicators exist that can not necessarily equate to higher-quality care (see Chart identify the underuse of effective care, but indicators that 5:9). Medicare is testing a pay-for-performance initiative measure the possible overuse of care in managing chronic for physician groups, called the Physician Group Practice illness are much less well developed. Demonstration, that will reward physicians for improving the quality and efficiency of health care services delivered Findings: An analysis of chronically ill Medicare to Medicare fee-for-service beneficiaries, especially beneficiaries receiving most of their care at 77 of patients with chronic illness who account for a significant America’s best-ranked hospitals during 999–2000 proportion of Medicare expenditures (CMS 2005d). revealed extensive variation in the amount of care * Rates of use were case-mix adjusted to control for differences in patients’ age, sex, provided to terminally ill patients in their last six months race, and disease comorbidities. of life.* Rates of service use for cancer patients varied more than fourteen-fold, for COPD patients more than seven-fold, and for CHF patients more than six-fold. Notably, the frequency of services used by patients with one chronic disease were closely correlated with the frequency of services used by patients with other chronic E Q U I T Y A N D E F F I C I E N C Y • C A R E AT E N D O F L I F E • C H A R T 5 : 8 Unexplained Variation in Care at End of Life Among chronically ill Medicare beneficiaries who received the majority of their care during 1999–2000 at 77 hospitals ranked as the best in America, there was striking variation in use of resources in the last six months of life, suggesting that where one receives care— more than the nature of one’s illness—determines the amount of care that is provided. ������������������������������������������������������������������������������������������������������������������� �������������������������������������������������������������������������������������������������������� ����������������������������� ����������������������������� ������������������������������ �� �� ��� ���� ���� ���� ���� ���� ���� �� �� �� ���� ���� �� � �� ��� ���� ���� �� ���� � �� ���� ���� ���� ��� ��� ��� ��� ���� � � �� ��� ���� ���� ���� ��� ��� ��� � � � ������ ���� ��� ������ ���� ��� ������ ���� ��� �������������������������� ��������������������� ����������������������������� Source: Medicare administrative data ( Wennberg et al. 2004b). Rates were case -mix adjusted to control for differences in patients’ age, sex, race, and disease comorbidities. ICU = intensive care unit. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 129 5: EQUITY 5: EQUITY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 130 E Q U I T Y A N D E F F I C I E N C Y • M U LT I P L E C O N D I T I O N S • C H A R T 5 : 9 Relationship Between Quality of Care and Medicare Spending Why is this important? The quality of care delivered to Implications: The authors state that this study “clearly Medicare beneficiaries varies widely by state (see Chart does not suggest that we mandate lower spending, :22). The Medicare Quality Improvement Organization because it is probably not spending per se that reduces program currently measures the effectiveness of care quality” (Baicker and Chandra 2004). Moreover, these 22 for several indicators such as the administration of beta indicators do not capture all aspects of high-quality care. blockers following a heart attack, mammograms for The composition of the medical workforce accounted older women, influenza vaccines, and eye exams for for almost half of the state-level variation in Medicare diabetics. These evidence-based practices are beneficial, spending per beneficiary. States with a higher proportion relatively inexpensive, and (with some exceptions) rarely of primary care practitioners (vs. specialists) exhibited contraindicated. Differences in the provision of effective better performance on these quality indicators and lower care likely depend on multiple factors, which may include costs per beneficiary. Hence, specialists may be clustered Medicare spending levels that differ across the states in areas where costly care “crowds out” the kinds of and the composition of the care provider workforce (i.e., effective care measured by this study. The authors suggest primary care clinicians, physician specialists, registered that possible interventions in these areas could focus on nurses, and others). promoting greater access to primary care clinicians and/ or involving specialists in the provision of more effective Findings: A national study found that states with higher care. Expanding this type of analysis to include a broader Medicare fee-for-service spending tend to deliver lower array of representative quality measures would provide a quality care to Medicare fee-for-service beneficiaries, fuller understanding of the relationship between the costs as assessed based on states’ overall ranking across 22 and quality of health care. quality indicators (see Appendix Table a for a list of the * The Medicare Quality Improvement Organization program included 24 quality indicators included in this study*). Furthermore, for every indicators but two indicators measuring time to reperfusion were excluded from the ,000 increase in Medicare spending per beneficiary, a state rankings described in this chart. state’s quality ranking dropped 0 positions (this inverse relationship between spending and quality is represented by the solid line on the graph). Higher spending was associated with greater use of hospital resources but was not associated with higher patient satisfaction (not shown) (Baicker and Chandra 2004). E Q U I T Y A N D E F F I C I E N C Y • M U LT I P L E C O N D I T I O N S • C H A R T 5 : 9 Relationship Between Quality of Care and Medicare Spending States with higher spending per Medicare beneficiary tended to rank lower on 22 quality of care indicators. This inverse relationship might reflect medical practice patterns that favor intensive, costly care rather than the effective care measured by these indicators. ������������������������������������������������������������������������������������������������������ ����������������������� ����������� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� �� ����������� �� �� ����� ����� ����� ����� ����� ����� ������������������������������������������������� Source: Medicare administrative claims data and Medicare Quality Improvement Organization program data, as analyzed by Baicker and Chandra (2004). The solid line shows that for every $1,000 increase in Medicare spending per beneficiary, a state’s quality ranking dropped by 10 positions. Adapted and republished with permission of Health Affairs from Baicker and Chandra, “Medicare spending, the physician workforce, and beneficiaries’ quality of care” ( Web Exclusive), 2004. Permission conveyed through the Copyright Clearance Center, Inc. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 131 5: EQUITY 5: EQUITY Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 132 E Q U I T Y • M U LT I P L E C O N D I T I O N S • C H A R T 5 : 1 0 Physician Perceptions of Quality of Care for White and Black Patients Why is this important? Racial disparities exist in the quality Implications: Visits by black patients were highly of care, and black patients generally receive lower-quality concentrated among a small subgroup of primary care care than white patients (see Chart 5:6). The Medicare physicians who more frequently reported difficulties program has the potential to help reduce disparities in gaining access to high-quality services for their in the quality of health care through the influence of patients than those physicians treating white patients. its purchasing and regulatory powers (Eichner and The differences in access to resources between these two Vladeck 2005), assuming that the causes underlying groups of physicians could translate into differences in these disparities can be accurately determined. One the quality of care delivered to patients. One of the two contributing factor may be that individuals of different overarching goals of the U.S. Department of Health and racial groups obtain their care—whether by choice or Human Services’ Healthy People 200 initiative is to because of availability—from doctors who differ in their eliminate health disparities, including differences that clinical qualifications and/or access to clinical resources. occur by race or ethnicity, by providing access to high- quality health care to all individuals (DHHS 2002a). The Findings: A study using a nationally representative sample of findings from this study suggest that these efforts must primary care physicians treating black and white Medicare address structural factors that influence physicians’ ability beneficiaries ages 65 and older in 2000 and 200 found to deliver high-quality care. that 80 percent of black patients received their care from only 22 percent of physicians. In a comparison of visits by white patients and black patients, the physicians visited predominantly by black patients were less likely than those visited predominantly by white patients to agree that they could provide high-quality care to their patients. They were also less likely to report that they could obtain access to high-quality specialists, high-quality diagnostic imaging, nonemergency hospital admissions, and high- quality ancillary services (Bach et al. 2004). E Q U I T Y • M U LT I P L E C O N D I T I O N S • C H A R T 5 : 1 0 Physician Perceptions of Quality of Care for White and Black Patients About 20 percent of physicians deliver care to 80 percent of black patients. As compared with physicians visited predominantly by white Medicare patients, physicians visited predominantly by black Medicare patients were less likely to report that they can deliver and obtain access to high-quality care for their patients. ������������������������������������������������������������������������������� ������������������������������������������������������������ � �� �� �� �� ��� �� ������������������������������������������� �� �� ����������������������������������������� �� �� ������������������������������������������������ �� �� ������������������ ������������������������������������������������������� ������������������ �� ������������������ ������������������ �� ������������������������������������������������ �� Source: Community Tracking Study Physician Survey linked to Medicare administrative data (Bach et al. 2004). *Results are weighted by patient visits and to be nationally representative. **Responses tabulated for this question were agree or somewhat agree; responses tabulated for all other questions were always or almost always. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 133 5: EQUITY SECTION 6 Capacity to Improve This section illustrates the promise of systematic improvements for achieving one or more of the Institute of Medicine’s six aims for the health care system: effective, safe, timely, patient-centered, equitable, and efficient health care. 6: C A PAC I T Y Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 136 C A PA C I T Y T O I M P R O V E • S TAY I N G H E A LT H Y • C H A R T 6 : 1 Improving Pneumococcal Vaccination Screening in the Hospital Why is this important? Little more than half of the elderly Findings: In the year following the intervention, the number of have received the pneumococcal vaccine despite the inpatients with pneumonia who were screened to determine recommendations of experts and the fact that Medicare their vaccination status while hospitalized at the Medical will pay for vaccination (see Chart :). To help improve Center of Louisiana increased by 60 percentage points among vaccination rates, the American Thoracic Society and the all patients and by 72 percentage points among elderly patients. Infectious Diseases Society of America support vaccinating The proportion who received the vaccination increased by 3 patients at risk for community-acquired pneumonia when they percentage points among all patients and by 34 percentage are hospitalized (Niederman et al. 200; Mandell et al. 2003). points among the elderly (Kruspe et al. 2003). Many elderly patients hospitalized with pneumonia have been admitted to the hospital before, which suggests that a hospital Implications: This educational intervention provides one stay provides a good opportunity to identify those who should model for increasing pneumococcal vaccination rates among be vaccinated to help prevent future hospitalizations for hospital patients. The Medicare program also permits the use pneumococcal infections (Fedson et al. 990). In 2002, only 26 of “standing orders” authorizing vaccination by nurses and percent of Medicare patients were screened for or received a pharmacists without the need for a physician’s examination pneumococcal vaccination while hospitalized with pneumonia and direct order (CDC 2003a). Medicare has eliminated (AHRQ 2005b). financial barriers by reimbursing hospitals for pneumococcal vaccination of Medicare beneficiaries in addition to regular Intervention: This study evaluated the impact of a year- payment for patients’ care under the prospective payment long pneumococcal vaccine educational intervention for system (CDC 997). Educational initiatives such as this one, in Louisiana State University (LSU) internal medicine primary combination with standing orders, may offer the most time- caregivers (i.e., house staff). The intervention was based efficient and effective solution for improving pneumococcal on recommendations from the Advisory Committee on vaccination rates among high-risk patients. Immunization Practices (CDC 997) and included: • lectures reviewing the benefits and indications of pneumococcal and other commonly accepted vaccinations, • reinforcement of these concepts at monthly physician orientation meetings, and • posted reminders for pneumococcal vaccination. C A PA C I T Y T O I M P R O V E • S TAY I N G H E A LT H Y • C H A R T 6 : 1 Improving Pneumococcal Vaccination Screening in the Hospital An intensive educational intervention for internal medicine physicians at a teaching hospital significantly improved the proportion of pneumonia patients who were screened to determine whether they had received a pneumococcal vaccination and the proportion who were given the vaccine when needed to prevent severe pneumococcal disease. ������������������������������������������������������������������������ ������������������������������������������������������������������������� ��� ������������������������������� ������������������������������ �� �� �� �� �� �� �� �� �� �� � � � �������� ���������� �������� ���������� ������������ ��������������������������� Source: Kruspe et al. 2003. Results based on review of patient records (N = 240 pre - and 194 post-intervention). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 137 6: C A PAC I T Y 6: C A PAC I T Y Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 138 C A PA C I T Y T O I M P R O V E • G E T T I N G B E T T E R • C H A R T 6 : 2 Improving Hospital Treatment for Heart Attack Why is this important? The American College of Cardiology GAP tools were used, five of six quality indicators showed and the American Heart Association first published significant improvement compared to control hospitals: evidence-based guidelines for the management of heart aspirin administration within 24 hours of admission and attack patients in 996. Many patients still do not receive smoking cessation counseling increased by 6 and 58 all the therapies recommended in the guidelines (see Chart percentage points, respectively, and prescription of aspirin, :8). Better adherence to evidence-based therapy could help beta blockers, and ACE inhibitors at discharge increased by prevent many of the 300,000 recurrent heart attacks that 6, 3, and 0 percentage points, respectively, from before occur annually (AHA 2005b). to after the intervention. Among all Medicare beneficiaries, GAP-participating hospitals showed a significant improvement Intervention: Ten acute-care hospitals in southeast Michigan only in prescribing aspirin at discharge compared with control implemented the American College of Cardiology’s Guidelines group hospitals (data not shown). Overall, the intervention Applied in Practice (GAP) quality-improvement project, effect tended to be greatest among older patients (those ages which is designed to incorporate national heart attack 75 and older). Furthermore, the intervention showed signs treatment guidelines into care practice. The one-year GAP of equalizing treatment among white and nonwhite patients intervention consisted of customizing guideline-based (Mehta et al. 2002). tools, assigning local physician and nurse opinion leaders, performing grand rounds site visits, and measuring quality Implications: The authors attributed the success of the GAP indicators among random samples of patients who were intervention to the development of tools that reinforce ideal candidates for therapy. The control group consisted the key goals of heart attack therapy, the identification of of  Michigan hospitals that volunteered for but were not implementation barriers, the flexibility of the intervention, selected to participate in the intervention, although they and the advantage of established relationships from prior were encouraged to undertake improvements in heart attack quality-improvement initiatives. Because tool use correlated treatment (Mehta et al. 2002). with the greatest improvements in quality measures, future initiatives might emphasize a longer implementation period Findings: Three GAP-promoted tools (standard admission with insistence on routine tool use (Mehta et al. 2002). orders, clinical pathways, and standard discharge forms) were documented for about one-quarter of patients in GAP- participating hospitals. Among Medicare patients for whom C A PA C I T Y T O I M P R O V E • G E T T I N G B E T T E R • C H A R T 6 : 2 Improving Hospital Treatment for Heart Attack Medicare heart attack patients at 10 southeast Michigan hospitals were more likely to receive evidence-based treatment after the hospitals engaged in a structured intervention that included customized, guideline-oriented tools, local physician and nurse opinion leaders, grand rounds site visits, and measurement of quality indicators. ������������������������������������������������������������������������������������������������������ ��������������������������������������������������������������������������������� ����������������������������������������� ���������������������������������������� �� ��� ��� �� �� �� �� �� �� �� �� �� �� �� �� � ����������������������� ����������������� ������� ������������ ������������� ���������� ������������������������������ �������������������������������� Source: American College of Cardiology Guidelines Applied in Practice (GAP) Initiative (Mehta et al. 2002). Results are based on random samples of medical records for patients who were ideal candidates for therapy during baseline (N = 515) and intervention (N = 663). ACE = angiotensin converting enzyme. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 139 6: C A PAC I T Y 6: C A PAC I T Y Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 140 C A PA C I T Y T O I M P R O V E • L I V I N G W I T H I L L N E S S • C H A R T 6 : 3 Reducing Rehospitalization for Congestive Heart Failure Why is this important? Older adults with multiple chronic Findings: At one year after hospital discharge, patients who health conditions and complex medication regimens are at received the transitional care intervention were less likely to high risk for poor outcomes following discharge from the have been readmitted to the hospital or to have died; also, they hospital, which often leads to hospital readmission (Naylor had 36 percent fewer hospital readmissions than patients who 2002). Elders suffering from congestive heart failure—a life- received usual care. The total cost of care for the intervention threatening condition in which the heart cannot pump enough group was 4,845 (39%) lower per patient than for the usual blood to meet the body’s oxygen needs—have the highest rate care group, after accounting for the cost of the intervention of rehospitalization among adult patients (AHA 2005b). (Naylor et al. 2004). Intervention: Elderly patients ages 65 and older who were Implications: The authors attributed the success of this hospitalized with heart failure at one of six Philadelphia area intervention to increased continuity of care and the hospitals during 997 to 200 were randomly assigned to individualized, holistic approach that APNs took to address receive either a transitional care intervention delivered by patients’ complex care needs. A meta-analysis of 8 other specially trained advanced practice nurses (APNs) or usual randomized controlled trials found that comprehensive care. Guided by a flexible, evidence-based protocol, the APNs discharge planning plus post-discharge support (of varying collaborated with physicians to provide individualized needs intensity) for patients with heart failure reduced hospital assessment, care planning, patient education, and therapeutic readmissions by 25 percent on average (Phillips et al. 2004). support during the patient’s hospitalization and in a series Implementing such a program nationally for all Medicare of home visits for three months after discharge. APNs were beneficiaries could prevent up to 84,000 hospital readmissions available seven days a week for telephonic patient support each year. (Naylor et al. 2004). (This care management approach is known as the Quality-Cost Model of Advanced Practice Nursing Transitional Care.) C A PA C I T Y T O I M P R O V E • L I V I N G W I T H I L L N E S S • C H A R T 6 : 3 Reducing Rehospitalization for Congestive Heart Failure Elderly patients hospitalized for heart failure were less likely to be readmitted to the hospital or to die and had lower health care costs overall when they received transitional care from an advanced practice nurse who provided needs assessment, care planning, patient education, and therapeutic support through discharge planning and home follow-up visits. ���������������������������������������������������������������������������������������������������������� ���������������������������������������������������������������������������������������������������������� ���������������� ������������������ ��� ��� ������� ��� �� ������� ��� ������� �� �� ��� �� ������ ��� ������ �� �� �� ������ � � � ������������������������������� ������������������������������� �������������������� �������������������������� Source: Medical records and patient interviews (N = 239) (Naylor et al. 2004). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 141 6: C A PAC I T Y 6: C A PAC I T Y Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 142 C A PA C I T Y T O I M P R O V E • L I V I N G W I T H I L L N E S S • C H A R T 6 : 4 Improving Depression Treatment and Outcomes Why is this important? Depressed older adults report poorer Findings: One year after the study began, 2 percent more quality of life, are at increased risk of death because of intervention patients were using antidepressant medication medical illness or suicide, and use more medical services or psychotherapy, 29 percent more were satisfied with their than nondepressed older adults (see Chart :9). Efforts to depression care, and 26 percent more demonstrated at least a improve the treatment of the depressed elderly through 50 percent improvement in depressive symptoms than those patient screening and practitioner education have fallen receiving usual care. The self-reported functional impairment short of expectations, suggesting that a more comprehensive score was 2 percent lower and the quality of life score was 9 intervention strategy is needed (Unutzer et al. 2002). percent higher for the intervention group compared to usual care (Unutzer et al. 2002). Intervention: Patients ages 60 and older who were treated at one of 8 primary care clinics affiliated with eight diverse Implications: This model offers a promising approach organizations received either usual care or a care intervention to improving depression care among elderly patients. delivered by specially trained nurses or psychologists Improvements were seen across all participating organizations, (depression clinical specialists) in collaboration with suggesting that this approach is feasible in diverse primary the patient’s primary care physician. Working under the care settings. Treatment of late-life depression is challenging, supervision of a psychiatrist and primary care expert reflected by the fact that less than 50 percent of patients and guided by evidence-based protocols, depression care receiving this intervention reported at least a 50 percent specialists conducted initial visits, devised treatment plans, decrease in depressive symptoms. The investigators predict and maintained weekly or biweekly contact (in person or by that the 2-month intervention cost of 553 per patient will telephone) with patients for up to 2 months. Care included likely offset health care costs otherwise incurred by this initiation of antidepressant medication and/or psychotherapy population, which are up to 50 percent higher than for older followed by regularly scheduled assessments to maintain or adults without depression (Unutzer et al. 2002). amend treatment (Unutzer et al. 200). C A PA C I T Y T O I M P R O V E • L I V I N G W I T H I L L N E S S • C H A R T 6 : 4 Improving Depression Treatment and Outcomes Older adults with depression were more likely to receive treatment and to be satisfied with care and achieved better outcomes when assigned to a trained nurse or psychologist who collaborated with the patient and primary care physician to support medication management and/or provide brief psychotherapy, under supervision of a psychiatrist and primary care expert. ��������������������������������������������������������������������������������������������������� �������������������������������������������������������������������������������������������������������� ���������� ������������ ��� ���������� �� ������������ �� �� �� � ���� �� ���� �� � �� �� ���� �� � ���� �� �� � � � ���������� ����������������� ������������ ������������������ ����������������������� �������������� ��������������� �������������� ���������������� ������������������� ������������� ������������������� ���������� ����������������� ������������������ ������������� ���������� �������� Source: Improving Mood: Promoting Access to Collaborative Treatment (IMPAC T ) program (Unutzer et al. 2002). Results based on patient interviews (N = 1,801). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 143 6: C A PAC I T Y 6: C A PAC I T Y Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 144 C A PA C I T Y T O I M P R O V E • L I V I N G W I T H I L L N E S S • C H A R T 6 : 5 Supporting Caregivers of Patients with Alzheimer’s Disease Why is this important? Family caregivers looking after Findings: During the first year of the study, caregivers receiving relatives with Alzheimer’s disease often experience a great enhanced services demonstrated a gradual decrease in deal of psychological distress, which may ultimately lead to symptoms of depression, whereas those receiving usual depression and compromise their caregiving ability. A recent services showed an increase in depressive symptoms. Although nationwide study found that 32 percent of family caregivers the difference in depression scores decreased in magnitude of patients with moderate to severe dementia reported six or with increasing time, caregivers in the intervention group had more symptoms of depression or were classified as depressed significantly lower depression scores than those in the usual (Covinsky et al. 2003). care group more than three years (6 weeks) after enrollment. The median time before Alzheimer’s patients were placed in Intervention: Spouses (average age 7 years) caring for patients a nursing home was almost  months (329 days) longer for with Alzheimer’s disease at home were randomly assigned those being cared for by caregivers receiving the intervention to receive either enhanced counseling and support or usual than by those receiving usual services (Mittelman et al. 996; services from the New York University Alzheimer’s Disease Mittelman et al. 2004). Center. Each caregiver in the intervention group was assigned a counselor who provided six individual and family counseling Implications: A short course of intensive counseling and ongoing sessions and ongoing ad hoc counseling customized to the support can have long-lasting effects in reducing symptoms of needs of each caregiver (e.g., techniques for managing difficult depression among family caregivers of Alzheimer’s patients, patient behavior and facilitating family communication). permitting Alzheimer’s patients to be cared for at home Caregivers in the intervention group attended weekly support almost one year longer before needing institutional care. The group meetings for continuous emotional support and sustained effects of the intervention might be attributable to education. Usual services consisted of information and advice its flexibility and ability to help caregivers develop long-term services and access to ad hoc counseling and support groups coping skills and resources (Mittelman et al. 2004). Wider on request (Mittelman et al. 2004). availability of interventions such as this might improve quality of life for the 25 million families caring for Alzheimer’s patients and potentially reduce the family and societal costs of care for Alzheimer’s disease. C A PA C I T Y T O I M P R O V E • L I V I N G W I T H I L L N E S S • C H A R T 6 : 5 Supporting Caregivers of Patients with Alzheimer’s Disease Family members who care for patients with Alzheimer’s disease often experience psychological distress. Providing spouse-caregivers with enhanced counseling and ongoing support reduced their burden of depression compared to those who received usual supportive services; this positive effect was sustained over three years on average. Alzheimer’s patients whose spouses received enhanced services were cared for at home nearly a year longer before being institutionalized. ���������������������������������������������������������������������� ������������������� ������������������������������������������������������������������������� ���������������������� �� ���������������������������� �� ��������������������������������� ���� ����������������������� ���� �� �� ���� � ��� � ��� � � ��� �������������� �������������� � � ��� ����������������� ����������������� � � � � �� �� �� �� ��� ��� ��� ��� ����� �������� �������� �������� ���������������������������������������������������� Source: New York University Spouse -Caregiver Intervention Study (Mittelman et al. 1996, 2004). Results based on caregiver interview (N = 406) using the Geriatric Depression Scale. *Depression scores were covariate adjusted to equalize rates at baseline. Depression score charts reprinted with permission from the American Journal of Psychiatry, Copyright 2004, American Psychiatric Association. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 145 6: C A PAC I T Y 6: C A PAC I T Y Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 146 C A PA C I T Y T O I M P R O V E • C H A N G I N G N E E D S • C H A R T 6 : 6 Improving Home Health Care Outcomes Why is this important? The goals of home health care and a protocol for nurses to contact the patient’s physician include “assisting patients to become or remain sufficiently within 24 hours to schedule a follow-up call or visit (Richard independent to stay in their home environment, avoiding et al. 2000). institutional long-term care or acute care” to the degree possible (Shaughnessy et al. 2002a). Although hospital Findings: Through participation in a national OBQI admissions are sometimes planned or necessary to provide demonstration program, which included training and technical optimal care and assure patients’ health, some hospitalizations assistance to implement OBQI, 54 HHAs in 27 states reduced represent adverse outcomes resulting from preventable events, risk-adjusted hospitalization rates by 22 percent over three years. such as falls or acute exacerbations of chronic conditions. The 9 HHAs participating in a New York State demonstration Home health agencies (HHAs) might be able to reduce such achieved a similar reduction of 26 percent over four years. adverse outcomes through proactive needs assessment and In contrast, hospitalization rates changed only very little for coordination with the patient’s physician to provide timely a random sample of non-OBQI Medicare patients in the care interventions in the home. same 27 states during the three-year national demonstration. Other targeted patient outcomes improved 5 to 7 percent Intervention: The federal government, in collaboration with per year versus an improvement rate of about  percent for New York State and the Robert Wood Johnson Foundation, nontargeted outcomes in participating HHAs (data not shown) developed the Outcome-Based Quality Improvement (Shaughnessy et al. 2002a). (OBQI) system to support continuous quality improvement in HHAs (Shaughnessy et al. 2002b). Using the Outcome Implications: The authors noted that physician involvement and Assessment Information Set (see Chart :2), patient was an important factor in agency-level improvement and assessments are centrally collected and analyzed to produce that most agencies needed to improve communication with annual reports comparing an agency’s performance with a physicians to achieve this effect. Although use of OBQI is national reference for 4 risk-adjusted outcomes measures. voluntary, the federal government has collaborated with HHAs use these reports to target outcomes for improvement, state agencies to offer training on the OBQI system to HHAs investigate care processes to determine problems, identify nationwide. Medicare Quality Improvement Organizations best practices to improve care, plan and implement actions (QIOs) also are helping HHAs implement OBQI. The Centers to achieve those practices, and monitor effectiveness. for Medicare and Medicaid Services has proposed that QIOs For example, one agency’s plan to reduce unplanned work more closely with HHAs to improve selected outcomes, hospitalizations included criteria to identify patients with with a focus on reducing hospitalizations (CMS 2004c). unstable conditions or otherwise in need of follow-up care C A PA C I T Y T O I M P R O V E • C H A N G I N G N E E D S • C H A R T 6 : 6 Improving Home Health Care Outcomes Home health care agencies participating in a national demonstration program used regular reports on their patients’ outcomes to plan and make improvements in care. The hospitalization rate fell by 22 percent over three years among agencies nationally and by 26 percent over four years among New York State agencies. �������������������������������������������������������������������������������������� �������������������������������������������������������������������������������������� ������������������������������������������������������ ������������������������������������������������ �� �� ���� �� ���� ���� ���� ���� ���� ���� ���� ���� ���� �� �� � ������������� ������������� ������������� ������������� ������������� ��������������������������������������� ����������������������������� ��������������������������������������� ������������������������������������������ Source: Reprinted with permission of Blackwell Publishing from Shaughnessy PW, et al. Improving patient outcomes of home health care. Journal of the American Geriatrics Society 2002; 50(8): 1354-64. Results are based on Outcome and Assessment Information Set (OASIS) patient assessments (N = 157,548 national and 105,917 New York State). Rates differ between comparison periods because of risk-adjustment and the number of participating agencies in each comparison period. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 147 6: C A PAC I T Y 6: C A PAC I T Y Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 148 C A PA C I T Y T O I M P R O V E • C H A N G I N G N E E D S • C H A R T 6 : 7 Program of All-Inclusive Care for the Elderly (PACE) Why is this important? The Medicaid program pays for the percent fewer days in the hospital and 6 percent fewer days costs of long-term care for impoverished elderly adults. Many in a nursing home. Although self-reported health status was states find it financially difficult to reimburse providers of similar between groups, PACE participants reported better traditional nursing home care while concurrently developing quality of life, higher satisfaction with care overall, and home and community-based alternatives to institutional care, more social contact than the comparison group. Notably, 24 which many elders prefer. Furthermore, the failure to integrate percent fewer PACE participants died during the 2-month Medicare and Medicaid funding fragments health care and observation period (Chatterji et al. 998). drives up costs (NPA 2002). Implications: A multivariate survival analysis indicated that Intervention: PACE provides comprehensive medical and social PACE participants had a median life expectancy of 5.2 years services to frail and impaired Medicare beneficiaries ages 55 versus 3.9 years for those in the comparison group (Chatterji and older who would otherwise need nursing home care but et al. 998). Moreover, black patients enrolled in PACE for one are able to live in the community. Enrollees must be eligible for year had lower mortality rates and less decline in activities Medicaid or self-pay the portion of costs that Medicaid would of daily living than white patients (Tan et al. 2003). Medicare pay. Service delivery is centered around attendance at an adult costs were 38 percent less during the first six months of day health center an average of three days per week, although enrollment in PACE and 6 percent less during the second the program pays for services in any setting under capitated six months than if individuals had continued to receive funding. An interdisciplinary care team of clinical and social fee-for-service Medicare (White 998). These outcomes led services professionals assesses participant needs, develops care Congress to make PACE a permanent Medicare program in plans, and delivers all services, thus creating an integrated, 997. Currently, more than 0,500 individuals are enrolled comprehensive care plan (Chatterji et al. 998; NPA 2002). in 73 PACE centers nationwide (NPA 2002). Despite early success, PACE expansion has lagged behind the congressional Findings: Frail elderly who participated in a PACE demonstration authorization for up to 90 PACE programs to be operating by project for one year reported a shift in the types of services 2004, indicating a need to overcome barriers to its widespread they received in the past six months and the settings in which implementation (Gross et al. 2004). they received them. Relative to a comparison group, PACE participants had 80 percent more ambulatory care visits but 60 percent fewer nurse visits to their home; they spent 67 C A PA C I T Y T O I M P R O V E • C H A N G I N G N E E D S • C H A R T 6 : 7 Program of All-Inclusive Care for the Elderly (PACE) PACE serves frail elders eligible for Medicare and Medicaid who are at risk of nursing home placement. Health care and supportive services are provided by an interdisciplinary team focused around attendance at an adult day care center. Participants enrolled in a PACE demonstration during 1995 to 1997 in 11 cities spent fewer days in a hospital or nursing home, had equal or better outcomes, and were less likely to die during the demonstration than those in a comparison group. ������������������������������������������������������ ������������������������������������������������������� � �� �� �� �� � �� �� �� �� ��� ��������������� ��� ����������������� �� ������������� ��� ������������� �� ���������������� ������������ ����������������� �������������� ���������� ��� ������������� �� ������������� ��� ���������� �� ��������������� ������������� ��� �� ��������������� ���������������� ��� ��������� �� ������������ ���� ������������� �� ������������ �������� ���� ����������������� �� ���������� ����������� ��� �� �������������� ������������������ ������������� ��� �� Source: Patient interviews conducted for the PACE demonstration evaluation (N = 783) (Chatterji et al. 1998). All differences between PACE and comparison, except in health status, were statistically significant in regression analyses controlling for baseline characteristics. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 149 6: C A PAC I T Y 6: C A PAC I T Y Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 150 C A PA C I T Y T O I M P R O V E • C A R E AT E N D O F L I F E • C H A R T 6 : 8 Expanding Palliative Care Options at the End of Life Why is this important? The leading causes of death among Findings: Patients enrolled in the palliative care program with persons ages 65 and older include congestive heart failure, congestive heart failure, chronic obstructive pulmonary cancer, stroke, and chronic obstructive pulmonary disease disease, or cancer were more satisfied with the care they (CDC 999). The health care system has failed to provide received (measured 60 days after enrollment) and more an adequate continuum of care for these individuals, who likely to die at home compared to those who received usual often cycle between bouts of acute hospital care followed by Medicare home health care before dying. Those receiving home health care (IOM 997). Many patients find that acute palliative care received 2.6 times more home health care visits care at the end of life causes pain, discomfort, and distress to and reported half as many visits to the physician, 75 percent themselves and their families (Baker et al. 2000). Moreover, fewer hospital days, and 80 percent fewer nursing home days although most people say they would prefer to die at home, as those in usual care. The average cost of personal health care one-half of Americans die in the hospital and almost one- services (not including facility charges) in the intervention quarter die in nursing homes (BMS 2004). Many patients group was 6,580 (45%) lower per patient than for the usual who desire and would benefit from palliative care may not be care group (not shown). Patients died an average of 02 admitted to hospice if they desire to continue receiving some days after enrolling in the palliative care program (Brumley ongoing complex services (Lorenz et al. 2004b). et al. 2003a, 2003b; personal communication with Susan Enguidanos 2005). Intervention: The Kaiser Permanente Palliative Care Project is an interdisciplinary home-based system of health care designed Implications: The authors suggest that by introducing palliative to provide patients suffering from life-threatening chronic care to chronically ill patients before the onset of dramatic conditions with the option of continuing to receive curative functional declines, patients nearing the end of life can better care while gradually transitioning to receive more palliative manage their care to their own satisfaction in their homes. care at the end of life. Palliative care enhances comfort and The Robert Wood Johnson Foundation’s Promoting Excellence improves patients’ quality of life through the provision of in End-of-Life Care program at the University of Montana is symptom control and pain relief, emotional and spiritual working with several organizations to create similar flexible, support, and patient education. The central care team consists innovative palliative care options in diverse care settings of the patient and family plus a physician, nurse, and social (Promoting Excellence 200). worker (Brumley et al. 2003a; 2003b). C A PA C I T Y T O I M P R O V E • C A R E AT E N D O F L I F E • C H A R T 6 : 8 Expanding Palliative Care Options at the End of Life Kaiser Permanente designed an innovative palliative care program for patients with life-threatening chronic illnesses who don’t yet qualify for hospice and wish to maintain their options for receiving curative care while gradually obtaining more supportive care at home as their condition worsens. This program allowed more participants to receive services and die at home rather than in an institution—an outcome most people say they desire—with increased satisfaction and at lower cost. ���������������������������������������������������������������������������������������������������� �������������������������������������������������������������������������������������������������������� ������������������������ ���������������������������� �� ��� �� �� �� �� �� �� �� �� �� �� ���� ���� �� ��� �� ��� ��� ��� ��� ��� ��� � � ���� ��������� ��������� �������� ������� ���������� �������������� ������������� ����������� ���� ������� ������������� ������������ ������������� ���������� Source: Adapted and reprinted from The Permanente Journal <www.kp.org/permanentejournal> 7(2), Brumley RD, Enguidanos S, Hillary K, The palliative care program, 7-12, Copyright 2003, by permission of the publisher, The Permanente Medical Groups. *Service use based on administrative records and adjusted for days enrolled, congestive heart failure diagnosis, and severity of illness (N = 300). **Satisfaction measured by patient interview 60 days after enrollment. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 151 6: C A PAC I T Y Appendices Tables 1a and 1b. Medicare Quality Improvement Organization Program Results for Medicare Fee-for-Service Beneficiaries Table 2. HEDIS Quality of Care Results for Managed Care Plans Technical Appendix Appendix Tables TABLE 1a. Medicare Quality Improvement M E D I A N S TAT E R AT E S W E I G H T E D N AT I O N A L A V E R A G E S Organization Program Results for Absolute Relative Absolute Relative Medicare Fee-for-Service Beneficiaries 1998 –1999 2000 –2001 1998– 1999** 2000– 2001 Change Change* Change Change* H E A R T AT TA C K * * * Aspirin given within 24 hours of admission 84 85 3 15 82 84 2 10 Aspirin prescribed at discharge 85 86 2 14 83 84 1 6 Beta-blocker given within 24 hours of admission 64 69 6 17 62 68 6 17 Beta-blocker prescribed at discharge 72 79 7 28 71 78 7 23 ACE Inhibitor prescribed at discharge (when indicated) 71 74 4 10 71 71 0 1 Smoking cessation counseling during hospitalization 40 43 3 5 38 38 0 0 H E A R T FA I L U R E * * * Evaluation of ejection fraction (LVEF) 65 70 5 14 63 71 8 22 ACE Inhibitor prescribed at discharge (when indicated) 69 68 -4 -10 68 66 -2 -6 STROKE*** Warfarin prescribed at discharge for atrial fibrillation 55 57 3 7 53 57 4 8 Antithrombotic prescribed at discharge for stroke or TIA 83 84 2 12 80 83 3 13 Avoidance of sublingual nifedipine for acute stroke 95 99 4 77 94 99 5 78 PNEUMONIA Antibiotic given within 8 hours of hospital arrival 85 87 2 10 83 85 2 12 Antibiotic selection consistent with current guidelines 79 85 7 32 76 84 8 34 Blood culture drawn (if done) before antibiotic 82 82 -2 -9 83 81 -2 -9 Influenza vaccination screening 14 27 9 10 15 24 9 11 Pneumococcal vaccination screening 11 24 11 12 11 23 12 13 I M M U N I Z AT I O N ( A G E S 6 5 + ) Influenza vaccination in past year 67 72 5 16 66 71 5 14 Pneumococcal vaccination ever 55 65 10 22 54 64 10 22 BREAST CANCER (WOMEN AGES 50-69) Mammogram in past 2 years 55 60 5 11 56 60 4 10 DIABETES (AGES 18-75) Hemoglobin A1c test in past year 70 78 8 29 61 70 9 28 Eye exam in past 2 years 68 70 1 4 73 74 1 3 Lipid profile in past 2 years 60 74 16 38 59 76 17 40 L O W E S T R AT E 11 24 -4 -10 11 23 -17 -9 M E D I A N S TAT E 70 73 4 13 H I G H E S T R AT E 95 99 16 77 94 99 17 78 Source: Centers for Medicare and Medicaid Services, Quality Improvement Organization program **Approximate weighted average rates for 1998–1999 were calculated by chartbook authors (Jencks et al. 2003). Adapted and used with permission from: Journal of the American Medical by subtracting absolute change from 2000–2001 average rates; actual rates may differ slightly Association, Jan. 15, 2003, 289: 310. Copyrighted © 2003, American Medical Association. All because of rounding. ***Excludes patients with documented contraindications to the medications. Rights reserved. Some numbers may not add because of rounding. TIA = transient ischemic ACE inhibitor measured for those with left ventricular systolic dysfunction attack.*Relative change = absolute change / (100 - baseline). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 155 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 156 TABLE 1b. Medicare Quality Improvement Organization Program U N W E I G H T E D N AT I O N A L A V E R A G E S Results for Medicare Fee-for-Service Beneficiaries 2000– 2001 2002 Absolute Change Relative Change* PNEUMONIA Blood culture drawn (if done) before antibiotic 81 81 -0 -1 Antibiotic given within 4 hours of hospital arrival NA 63 NA NA Antibiotic selection consistent with current guidelines** 85 68 -17 -106 Influenza vaccination screening 26 28 1 2 Pneumococcal vaccination screening 25 26 1 2 H E A R T AT TA C K * * * Aspirin given within 24 hours of admission 85 85 0 1 Aspirin prescribed at discharge 86 87 2 11 Beta-blocker given within 24 hours of admission 69 76 7 23 Beta-blocker prescribed at discharge 79 82 3 14 ACE Inhibitor prescribed at discharge (when indicated) 74 67 -7 -27 Smoking cessation counseling during hospitalization 43 50 7 12 H E A R T FA I L U R E * * * Evaluation of ejection fraction (LVEF) 69 76 7 22 ACE Inhibitor prescribed at discharge (when indicated) 66 65 -1 -3 Source: Centers for Medicare and Medicaid Services, Medicare Quality Improvement Organization program (AHRQ 2005b). These data are not comparable to data in Table 1a because of differences in the way that averages were calculated. The chartbook authors calculated absolute and relative change using reported rates. Some numbers may not add because of rounding. *Relative change = absolute change / (100- baseline). **Guidelines were updated in 2002. ***Excludes patients with documented contraindications to the medications. ACE inhibitor measured for those with left ventricular systolic dysfunction. Note: The decrease in rate of ACE inhibitor use from 2000–2001 to 2002 might reflect substitution of angiotensin receptor blockers (ARBs). TABLE 2. HEDIS Quality of Care Results for Managed Care Plans* MEDICARE MANAGED EMPLOYER-SPONSORED CARE PLANS MANAGED CARE PLANS TOPIC MEASURE 2000 2003 2000 2003 Immunization (ages 50-64 or 65+) Flu shot in past year NA 74.5 NA 48.0 Breast cancer screening (women ages 52-69) Mammogram in past 2 years 73.9 74.0 74.5 75.3 Colorectal cancer screening (ages 52-80) Colorectal cancer screening test in appropriate interval NA 49.5 NA 47.4 Smoking cessation counseling (ages 18+) Medical assistance to quit smoking (current smokers) 59.7 63.3 66.3 68.6 Osteoporosis management (women ages 67+) Osteoporosis screening or treatment following a fracture NA 18.0 NA NA Heart attack treatment (ages 35+) Beta-blocker treatment after a heart attack 89.3 92.9 89.4 94.3 Cholesterol screening 70.6 81.0 74.2 80.3 Cholesterol management after heart attack Cholesterol control (LDL<130) 52.9 66.7 53.4 65.1 (ages 18-75) Cholesterol control (LDL<100) NA 49.6 NA 47.6 Controlling high blood pressure (ages 46-85) Adequate blood pressure control (<=140/90 mmHg) 46.7 61.4 51.5 62.2 Eye exam in past year 62.8 64.9 48.1 48.8 Hemoglobin A1c (HbA1c) test in past year 82.5 87.9 78.4 84.6 Poor blood sugar control (HbA1c >9)** 33.4 23.4 42.5 32.0 Comprehensive diabetes care (ages 18-75) Lipid profile in past year 80.5 91.1 76.5 88.4 Cholesterol control (LDL<130) 50.9 67.7 44.3 60.4 Cholesterol control (LDL<100) NA 41.9 NA 34.7 Screening for kidney disease in past year 45.0 53.6 41.4 48.2 Effective acute phase treatment 51.3 53.3 56.9 60.7 Antidepressant medication management*** Effective continuation phase treatment 36.8 39.2 40.1 44.1 (ages 18+) Optimal practitioner contacts 11.9 10.5 19.8 20.3 Follow-up within 7 days 37.5 38.8 48.2 54.4 Hospitalization for mental illness (ages 6+) Follow-up within 30 days 59.3 60.3 71.2 74.4 Rated health plan highly (8-10 on 10-point scale) 78.8 72.0 59.3 61.8 Patient experience (adults) No problem with customer service 80.3 79.9 66.6 70.8 No problem getting needed care 85.0 84.1 75.4 78.4 Source: National Committee for Quality Assurance, The State of Health Care Quality: 2004. Adapted with permission. *These data may not be comparable to the data in Table 1 because of differences in measures and data sources. **A lower rate represents better performance for this measure. ***The first year shown for antidepressant medication management is 2001, not 2000. LDL = low-density lipoprotein. See the Technical Appendix for other notes on HEDIS. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 157 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 158 Technical Appendix This appendix provides more information on conditions. For more information, see: http://jama. “highlight potential quality concerns, identify areas data sources and study methodologies. Quality ama-assn.org/cgi/content/full/284/18/2325. that need further study and investigation, and track measurement sets and national data sources are changes over time” (AHRQ 2003a). Researchers described first, followed by notes on specific charts The Assessing Care of Vulnerable Elders at the University of California San Francisco and studies. Study populations (denominators) (ACOVE) quality measurement system focuses on and Stanford University refined the indicators are defined for each chart. Numerators are 22 health conditions, selected through an expert through a review of validity, reliability, and described for selected charts to clarify information ranking process based on prevalence, impact, usefulness based on a literature review, empirical displayed in the chart. The chartbook presents effectiveness, feasibility, quality gap, and geriatric testing, and an expert clinician panel review. data from many different kinds of sources and focus (Sloss et al. 2000), that together represent Inpatient Quality Indicators (IQIs) included studies conducted by different researchers using “the most important conditions vulnerable in the chartbook represent conditions and potentially different methodologies. Therefore, elders encounter in all care venues” (Wenger et procedures “for which mortality has been data may not be comparable between charts. al. 2003). They fall into three broad categories: shown to vary substantially across institutions Each type of data source has strengths and General medical: depression, diabetes, hearing and for which evidence suggests that high limitations for quality measurement. Differences impairment, heart failure, hypertension, ischemic mortality may be associated with deficiencies described as statistically significant reflect heart disease, osteoarthritis, osteoporosis, in the quality of care” (AHRQ 2002a). a 95 percent confidence level or greater. pneumonia, stroke, and vision impairment. Prevention Quality Indicators (PQIs) “identify Geriatric: dementia, end-of-life care, falls or hospital admissions that evidence suggests could Quality Measurement Sets and National mobility disorders, malnutrition, pressure have been avoided, at least in part, through Data Sources Used in the Chartbook ulcers, and urinary incontinence. high-quality outpatient care” (AHRQ 2002b). Cross-cutting: continuity of care, hospital Patient Safety Indicators (PSIs) identify The Access to Care for the Elderly Project (ACE- care, medication use, pain management, potentially preventable complications of care PRO) indicators measure underuse of necessary and screening and prevention. and adverse events in the hospital. In empirical care that is “likely to be associated with avoidable Using systematic evidence reviews and expert testing against medical records, PSIs were more poor outcomes” (Asch et al. 2000). Researchers at judgment, researchers at RAND developed likely to identify process of care failures than a RAND, a nonprofit research institute, reviewed potential indicators to represent quality of random sample of control cases (AHRQ 2003b). evidence and expert opinion to develop proposed care for these conditions, of which 236 were For more information, see: www. indicators of necessary care for which: “(1) the accepted as valid by two expert panels and the qualityindicators.ahrq.gov. benefits of the care outweigh the risks..., (2) the American College of Physicians Task Force on benefits to the patient are likely and substantial, and Aging. For more information, see: www.rand. The Behavioral Risk Factor Surveillance System (3) physicians have judged that not recommending org/health/tools/vulnerable.elderly.html. (BRFSS) is a continuous, cross-sectional telephone the care would be improper.” A multispecialty survey of the civilian, noninstitutionalized adult expert physician panel accepted 40 indicators The Agency for Healthcare Research and Quality population conducted in cooperation between representing 15 common acute and chronic medical (AHRQ) Quality Indicators are designed to use the Centers for Disease Control and Prevention hospital administrative data (billing records) to (CDC) and health departments in the 50 states, the District of Columbia, Puerto Rico, Guam, noninstitutionalized adults ages 19 and states, approximating a 20 percent stratified and the Virgin Islands. Rates are weighted to be older. Results are weighted to be nationally sample of U.S. community hospitals. Results representative of the adult population in each representative. The content of the survey are weighted to give national estimates. state or territory. The 2002 BRFSS included focuses on topics of current interest. The 2001 The 2001 HCUP Statewide Inpatient Database 247,964 respondents (response rate 58 percent), survey (the source for Chart 4:1) included 2,829 includes all discharges from hospitals in 33 of whom 51,082 were ages 65 and older (CDC adults ages 19–64 and 628 adults ages 65 and participating states, representing approximately 2003b). The 2003 BRFSS included 266,346 older. The overall response rate was 54 percent 85 percent of all U.S. hospital discharges, totaling respondents (response rate 53 percent), among (Davis et al. 2002). For more information, over 28 million inpatient discharge abstracts. whom 56,547 were ages 65 and older (CDC 2004a). see: www.cmwf.org/surveys/surveys.htm. For more information, see: www. For more information, see: www.cdc.gov/brfss. ahrq.gov/data/hcup. The Community Tracking Study (CTS) CAHPS (Consumer Assessment of Health Plans Household Survey, conducted by the nonprofit The Health Plan Employer Data and Information and Providers Study) is “a comprehensive and Center for Studying Health System Change, Set (HEDIS) is “a set of standardized performance evolving family of surveys that ask consumers and is a periodic, nationally representative, cross- measures designed to ensure that purchasers and patients to evaluate...those aspects of care for which sectional telephone survey of the civilian, consumers have the information they need to consumers and patients are the best and/or only noninstitutionalized population. In-person reliably compare the performance of managed source of information” (SUN 2004). CAHPS was interviews are conducted with households without health care plans” (NCQA 2005). HEDIS was originally developed by researchers at Harvard, telephones to ensure representation. The survey developed by the National Committee for Quality RAND, and the Research Triangle Institute for sample consisted of 47,000 to 60,000 individuals Assurance (NCQA), a nonprofit accreditation and the Agency for Healthcare Research and Quality. depending on the year, with response rates of quality-monitoring organization. NCQA collects The Centers for Medicare and Medicaid Services 57 to 65 percent. The CTS Physician Survey Medicare HEDIS data on behalf of the Centers administers a Medicare version of CAHPS annually is a biannual telephone survey of physicians for Medicare and Medicaid Services (CMS) (in English and Spanish) to those who have been in 60 randomly selected metropolitan areas. from all Medicare managed care plans that CMS enrolled in Medicare for at least six months. A self- The survey includes physicians who report requires to report HEDIS data. HEDIS data on administered survey is sent by mail with follow-up providing at least 20 hours of direct patient care employer-sponsored health plans represent 262 of nonrespondents by telephone or special delivery. in an office- or hospital-based practice. The commercial organizations that submitted results to Medicare managed care members are randomly response rate among physicians in the 2000–2001 NCQA. HEDIS uses data from member surveys, sampled at the plan level (N=128,000 in 2003; survey was 59 percent. For more information, administrative claims, and medical records. Results response rate 81 percent). Medicare fee-for-service see: www.hschange.org/index.cgi?data=12. are audited according to NCQA’s standards (NCQA beneficiaries are randomly sampled at the county 2004). For more information, see: www.ncqa.org. level (N=122,000 in 2003; response rate 70 percent) The Healthcare Cost and Utilization Project (Goldstein et al. 2001; RTI 2001; Landon et al. 2004; (HCUP) is a collaboration between state and The Health and Retirement Study (HRS) is a personal communication with Elizabeth Goldstein private data organizations, hospital associations, nationally representative, longitudinal survey 2005). For more information, see: www.cms.hhs. and the federal Agency for Healthcare Research and of community-dwelling adults conducted gov/researchers/projects/consumers/cahps.asp. Quality “to create a national information resource by the University of Michigan, Ann Arbor, of discharge-level health care data” (AHRQ 2005b). for the National Institute on Aging. The first The Commonwealth Fund (CMWF) Health The 2001 HCUP Nationwide Inpatient cohort represents individuals born between Insurance Survey is a biennial, cross- Sample contains over seven million inpatient 1931 and 1941, and their spouses regardless sectional telephone survey of U.S. civilian, discharges from 986 hospitals located in 33 of age. In 1992, baseline home interviews Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 159 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 160 were conducted in English and Spanish with access to care, information needs, satisfaction by state) and from Medicare administrative data 9,825 individuals in 7,702 households in the with care, and income. The sample represents for post-discharge surveillance (such as hospital continental United States (response rate 82 beneficiaries who were enrolled in the Medicare readmissions). Medical record abstraction percent). Follow-up interviews with this cohort program for the entire year. (Some reports using emphasizes transparency and reliability using have been conducted biennially. For more this data have been adjusted to represent Medicare explicit clinical criteria. For more information, see: information, see: http://hrsonline.isr.umich.edu. beneficiaries enrolled at any time during the year.) www.qualidigm.org/what_con_patientSafety.shtml. The MCBS Cost and Use File includes complete The Medical Expenditure Panel Survey (MEPS), expenditure and source of payment data on The Medicare Quality Improvement sponsored by the federal Agency for Healthcare all health care services received by Medicare Organization (QIO) program uses quality Research and Quality (AHRQ), “produces beneficiaries, including those not covered by measures derived from professionally developed nationally representative estimates of health care the Medicare program. This data is derived practice guidelines to assess the quality of care use, expenditures, sources of payment, insurance from linked survey and Medicare claims data. received by Medicare fee-for-service beneficiaries. coverage, and quality of care for the U.S. civilian, The file also includes most items from the During 1998–2001, measurement focused on six non-institutionalized population” (NCHS 2004a). Access to Care file. The sample represents all common conditions “for which there is strong The core survey, called the Household Component beneficiaries who were enrolled in the Medicare scientific evidence and professional consensus (HC), consists of a series of interviews with a program at any time during the year. that the process of care either directly improves subsample of participants in the National Health For more information, see: www.cms. outcomes or is a necessary step in a chain of care Interview Survey. The 1996, 1998, and 2000 hhs.gov/mcbs/Overview.asp. that does so” (Jencks et al. 2003). Inpatient data samples included 10,000 families and the 2001 were abstracted (at two central abstraction centers) sample included 13,500 families, with response Medicare administrative data, maintained from medical records for systematic random rates of about 66 percent for full-year participation. by the Centers for Medicare and Medicaid samples of hospital discharges identified from For more information, see: www.meps.ahrq.gov. Services, include enrollment data for all Medicare Medicare hospital claims. Sample sizes ranged beneficiaries and claims data on covered services from 600 to 900 records for each condition (acute The Medicare Current Beneficiary Survey paid for by the traditional Medicare program myocardial infarction, heart failure, pneumonia, (MCBS) is a continuous, longitudinal survey of (NCHS 2004a). The Medicare Provider Analysis and stroke/atrial fibrillation) in each state. Median a representative national sample of the Medicare and Review (MedPAR) files contain information interrater reliability for medical record abstraction population, conducted by the federal Centers for on hospital inpatient stays by Medicare fee-for- was 90 percent. Diabetes care and mammography Medicare and Medicaid Services. Each survey service beneficiaries. For more information, rates were calculated using outpatient Medicare participant is interviewed three times per year see: www.cms.hhs.gov/data/default.asp. claims for services. Immunization rates were for four years. About 15 percent of community- derived from the BRFSS or a special survey dwelling respondents designate a proxy to answer The Medicare Patient Safety Monitoring System designed to emulate the BRFSS, representing all for them. Personal interviews are conducted with (MPSMS) is “a nationwide surveillance project community-dwelling elderly. See Appendix Table 1a 15,000 to 19,000 respondents in each round, aimed at identifying the rates of specific adverse for a list of the quality indicators used during 1998– with response rates “in the mid to high 80s” for events within the Medicare population” (Hunt et 2001. For more information, see: www.medqic.org. the initial interview and about 95 percent in al. 2004). The MPSMS was created by the Centers subsequent rounds (NCHS 2004a; CMS 2004a). for Medicare and Medicaid Services in consultation The National Ambulatory Medical Care Survey The MCBS Access to Care File combines survey with other federal agencies. Data are drawn from (NAMCS), conducted by the National Center for data with Medicare administrative data to represent a national random sample of medical records Health Statistics, is a nationally representative insurance coverage, health status and functioning, for all Medicare hospital discharges (stratified survey of nonfederal, office-based physicians who are primarily engaged in direct patient care. participant during the home interview and 2005a). Research on interrater reliability suggests The specialties of anesthesiology, pathology, physical examination. NHANES III (1988–1994) substantial to excellent agreement on standardized and radiology are excluded. Participating selected 39,695 persons of whom 78 percent patient assessments by different clinicians (Sangl physicians complete an encounter form for participated in the medical examination. NHANES et al. 2005). The federal government requires that each patient visit during a randomly selected 1999–2000 selected 12,160 persons of whom 76 all Medicare-certified home health agencies collect week, listing new or ongoing diagnoses and percent participated in the medical examination and report OASIS data for adult, nonmaternity prescribed medications. In recent years, about (NCHS 2004a). For more information, see: patients whose skilled care is paid for by 1,000 to 1,500 physicians have participated, http://www.cdc.gov/nchs/nhanes.htm. Medicare or Medicaid. For more information, representing a response rate of 63 to 71 percent see: www.cms.hhs.gov/oasis/hhoview.asp. (NCHS 2004a). For more information, see: www. The National Hospital Ambulatory Medical Care cdc.gov/nchs/about/major/ahcd/ahcd1.htm. Survey (NHAMCS), conducted by the National Notes on specific charts Center for Health Statistics, is a nationally Acronyms in bold refer to the quality measurement The National Health Interview Survey (NHIS) representative survey of visits to emergency sets and national data sources described is a continuous, cross-sectional, nationally departments (EDs) and outpatient departments above. Terms in italics refer to chart labels. representative household interview survey of (OPDs) of nonfederal, acute-care hospitals in the the civilian, noninstitutionalized population of United States. Hospital staff complete encounter Chart 1:1—National data are from the NHIS the United States, conducted by the National forms for a systematic random sample of patient (questions about vaccination were not asked in Center for Health Statistics (NCHS). In recent visits during a randomly selected four-week certain years) (NCHS 2004a, fig. 10). The reference years, about 30,000 adults have participated in period. About 500 hospitals participate each population is U.S. civilian, noninstitutionalized the core survey, which has achieved household year, of which about 80 percent have EDs and adults ages 65 and older. National rates were age- response rates ranging from 90 to 98 percent. about 50 percent have OPDs. Response rates adjusted to the 2000 U.S. standard population using Supplements are conducted on selected topics, ranged from 93 to 97 percent for EDs and 86 to two age groups: ages 65–74 and ages 75 and older. such as cancer screening, in selected years. 95 percent for OPDs in recent years. Data are State data are from the 2003 BRFSS (CDC Response rates for survey supplements have ranged weighted to represent national estimates (NCHS 2004a, table 1). The reference population is from 70 to 80 percent (NCHS 2004a). For more 2004a). For more information, see: www.cdc. U.S. civilian, noninstitutionalized adults ages information, see: www.cdc.gov/nchs/nhis.htm. gov/nchs/about/major/ahcd/ahcd1.htm. 65 and older. Those with unknown vaccination status were excluded from state rates. Adult The National Health and Nutrition The Outcome and Assessment Information Set vaccination rates measured by the BRFSS tend Examination Survey (NHANES) is a nationally (OASIS) is “a group of data elements that represent to be somewhat higher than those measured by representative, cross-sectional survey of civilian, core items of a comprehensive assessment for an the NHIS, probably because of differences in noninstitutionalized Americans. Trained adult home care patient; these core items and a question wording (Nelson et al. 2003). In state interviewers survey participants at home and comprehensive assessment serve as the basis for quartile rankings for pneumococcal vaccination, participants attend a mobile examination center the development of the care plan and ongoing Florida and Connecticut were both placed in the (MEC) to undergo medical examination and management of the patient; and form the basis second quartile because they have the same rate; provide blood, urine, and other tissue samples for measuring patient outcomes for purposes of one of the two states would have been placed for laboratory analysis following standard outcome-based quality improvement...Skilled home in the third quartile based on ordinal ranking. protocols. Those who cannot attend the MEC health staff gather the information by observing the Hawaii’s pneumococcal vaccination rate was 69.4 are examined at home. An average of three blood patient and the patient’s home and situation, and percent in 2003, not 44.5 percent as reported in pressure readings are taken for each survey by talking with the patient and caregivers” (CMS Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 161 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 162 the source (personal communication with Judy is U.S. civilian, noninstitutionalized adults assessment and improvement studies (personal Strait-Jones, Hawaii Department of Health, 2005). ages 65 and older. For state quartile rankings, communication with Paul Shekelle 2005). Arizona, California, and Vermont were all Chart 1:2—National data are from supplements placed in the second quartile because they all Chart 1:6—Data are from the NAMCS and to the NHIS in certain years (NCHS 2004a, table have the same rate; one of the three states would NHAMCS for 1997 and 1998 (combined) and 2000 81). The reference population is U.S. civilian, have been placed in the first quartile based on and 2001 (combined) (AHRQ 2005b, table 1.91). noninstitutionalized female adults of specified ordinal ranking. The BRFSS national median The measure is derived from Healthy People 2010, ages. The question regarding mammography rate for ever receiving a sigmoidoscopy or indicator 14-19 (DHHS 2002b). The reference varied slightly across NHIS survey years. “Data colonoscopy was 58 percent among those ages population is U.S. civilian, noninstitutionalized starting in 1997 are not strictly comparable 65 and older in 2002, as measured across 54 adults of specified ages. The numerator is the with data for earlier years due to the 1997 BRFSS reporting units (www.cdc.gov/brfss). number of antibiotic courses ordered, supplied, questionnaire redesign” (NCHS 2004a). administered, or continued at a specific visit for State data are from the 2002 BRFSS (CDC/ Chart 1:4—Data are from the 2000 MCBS Access persons diagnosed with the common cold. MIAH 2004, table 3). The reference population is to Care File (Adler and Shatto 2002, fig. 1, 2). U.S. civilian, noninstitutionalized female adults The reference population is community-dwelling, Chart 1:7—Data are from the Medicare QIO ages 65 and older. In 2002, the BRFSS median female Medicare beneficiaries ages 65 and older. program for 2002 (AHRQ 2005b, tables 1.85a, mammography rate for women ages 65 and 1.86a, 1.87a). The reference population is Medicare older was 77 percent across 54 reporting units. Chart 1:5—The ACOVE-2 study included 644 fee-for-service beneficiaries discharged from the The same rate was reported on the 2000 BRFSS community-dwelling patients ages 75 and older hospital with a principal diagnosis of pneumonia. across 52 reporting units (www.cdc.gov/brfss). who were being treated at one of two California Blood cultures collected before antibiotics given refers In contrast, a rate of 68 percent was reported on medical groups. Patients were included in this study to pneumonia patients for whom blood cultures the 2000 NHIS for women ages 65 and older. if they screened positive for falls or fear of falling, were ordered, among whom the blood culture urinary incontinence, or memory impairment. was collected before the date and time that the Chart 1:3—National data are from the 2000 One practice focused on serving managed care initial antibiotic dose was administered. Antibiotic NHIS cancer control module (Seeff et al. patients while the other served a mix of managed given within 4 hours of hospital arrival refers to 2004, table 1). The reference population is care and fee-for-service patients. Baseline data pneumonia patients who received any antibiotic U.S. civilian, noninstitutionalized adults of shown in the chart were collected from medical within four hours of hospital presentation. specified ages. For national rates, those with records from Sept. 2000 through Sept. 2001 at Antibiotic was consistent with guidelines refers a history of colorectal cancer were excluded one site and from Dec. 2000 through Dec. 2001 to immunocompetent pneumonia patients who from the analysis. Respondents who received at the second site, before the intervention phase received an initial antibiotic regimen consistent a home fecal occult blood test (FOBT) as of the study (Reuben et al. 2003b; Wenger et al. with current professional guidelines (for intensive part of a routine physical exam/screening or 2005). The quality of care represented in the care or non-intensive care patients) during the first because of a family history of cancer were ACOVE study might be better than average care 24 hours of their hospitalization (Jencks et al. 2000). counted as having received FOBT for screening in the United States, given that the participating purposes. Respondents who had ever received practices were large groups, participate in managed Chart 1:8—Data are from the Medicare QIO sigmoidoscopy, colonoscopy, or proctoscopy were care and are therefore subject to ambulatory program for 2002 (AHRQ 2005b, tables 1.36a, asked about the timing of the most recent test. care quality audits, and have a history of 1.37a, 1.38a, 1.39a, 1.40a). The reference population State data are from the 2002 BRFSS (CDC/ participating with academic researchers in quality is Medicare fee-for-service beneficiaries discharged MIAH 2004; table 3). The reference population from the hospital with a principal diagnosis of acute myocardial infarction and no documented Chart 1:10—Rates were calculated by the Medicare of Medicare hospital admissions for these contraindications to the particular treatment or Payment Advisory Commission, applying the conditions as reported by MedPAC (2004c). other documented reason for not prescribing AHRQ Inpatient Quality Indicators (AHRQ the drug. The study authors noted that, “we 2002a) to all hospital claims in the MedPAR file Chart 1:12—Data for blood pressure awareness know from...field experience with the measures for specified years (MedPAC 2004c, table 2-2). The are from the 1998 NHIS (AHRQ 2005b, table that valid, unmeasured contraindications are reference population is Medicare fee-for-service 1.33). The reference population is U.S. civilian, not frequent” (Jencks et al. 2000). The ACE beneficiaries discharged from the hospital with a noninstitutionalized adults of specified ages. Data inhibitor measure is limited to those with principal diagnosis for the selected condition or for high blood pressure control are from NHANES documented left ventricular systolic dysfunction with a procedure code for the selected procedure. III and the 1999–2000 NHANES (AHRQ 2005b, (left ventricular ejection fraction less than 40 The 30-day mortality rate was measured from table 1.46). The reference population is U.S. percent or narrative description of left ventricular hospital admission. Mortality rates were adjusted civilian, noninstitutionalized adults of specified function indicating moderate or severe systolic for age, gender, and severity of illness using ages with elevated blood pressure (average dysfunction). This measure did not account for the all-patient refined diagnosis-related groups systolic pressure of at least 140 mmHg or average the substitution of angiotensin receptor blockers (APR-DRGs) to control for changes in these diastolic pressure of at least 90 mmHg) or who (ARBs), which may add up to 10 percentage characteristics of the patient population. were taking antihypertension medication. The points to the rate (Masoudi et al. 2004) and will be numerator represents those in the denominator counted for compliance in future years (personal Chart 1:11—Rates were calculated by the Medicare whose average systolic blood pressure was lower communication with Edwin Huff 2005). Payment Advisory Commission, applying the than 140 mmHg and whose average diastolic AHRQ Prevention Quality Indicators (AHRQ blood pressure was lower than 90 mmHg. Chart 1:9—Data are from the Medicare QIO 2002b) to all hospital claims in the MedPAR program for 2000–2001 (Jencks et al. 2003, table file for specified years (MedPAC 2004c, table Chart 1:13—Data are from NHANES 1999–2000 2). The reference population is Medicare fee-for- 2-5). The reference population is Medicare (Ford et al. 2003, table 3). The reference population service beneficiaries discharged from the hospital fee-for-service beneficiaries. Admissions were is U.S. civilian, noninstitutionalized adults of with a principal diagnosis of acute myocardial identified based on principal diagnosis codes for specified ages with total cholesterol concentration infarction, with documentation of ST-segment the selected conditions, except that admissions of 5.2 mmol/L (200 mg/dL) or greater, or who elevation myocardial infarction or left bundle for lower extremity amputation were identified were taking cholesterol-lowering medication. branch block on the interpretation of the 12- from a relevant procedure code in any field with a Survey participants were considered to have lead ECG done closest to hospital arrival, and diagnosis of diabetes in any field (AHRQ 2002b). their cholesterol controlled if their cholesterol who received reperfusion. The PTCA measure Hospitalization rates were adjusted for age and concentration was less than 5.2 mmol/L (200 excludes those who received thrombolysis during sex to control for changes in these characteristics mg/dL), as determined from a blood test. the hospital stay (CMS 2003; AHRQ 2005b). of the patient population. The analysis excluded The study authors noted that results for specific beneficiaries admitted to the hospital from other Chart 1:14—Data are from HEDIS for the 2000 states must be interpreted with caution because hospitals or long-term-care facilities; however, and 2003 measurement years (NCQA 2004, of small sample sizes in some states; hence, rates MedPAC notes that “the reliability of admission 26, 31, 33). Beta-blocker prescribed after heart are not identified for specific states on the chart. source is somewhat questionable.” The cost- attack refers to adults ages 35 and older who However, “the effect of small denominators is to savings calculation described in the narrative received an outpatient prescription for a beta- increase the variation among the states, not to was adapted from an example constructed by blocker within seven days after being discharged bias the median downward” (Jencks et al. 2000). Kruzikas et al. (2004), substituting the number alive from the hospital with a diagnosis of acute myocardial infarction (NQMC 2003d). Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 163 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 164 The specification changed in 2003 to include Chart 1:15—Data are from the NAMCS for Chart 1:17—This chart reports follow-up data certain patients with complicating conditions 1991 and 1992 (combined) and 1999 and collected from medical records and patient who were excluded in previous years. 2000 (combined) (Fang et al. 2004, table 3). interviews for the ACOVE-2 study (see Chart Cholesterol management after heart attack The reference population is U.S. civilian, 1:5 for description of the study population). refers to adults ages 18 to 75 who were discharged noninstitutionalized adults with atrial fibrillation The interviews were conducted at the end of alive in the prior year for an acute coronary (AF) who visited community physicians. the intervention phase of the study (Apr. 2002 event (hospitalization for acute myocardial Encounter records identified new or continued through Apr. 2003 at one site and July 2002 infarction or coronary artery bypass graft, or oral anticoagulant medication (warfarin sodium, through July 2003 at the second site). Osteoarthritis performance of percutaneous transluminal dicumarol, anisindione, and phenprocoumon) or was not a target condition for the intervention. coronary angioplasty in any setting) and who were aspirin that was ordered, supplied, or administered The indicators relating to non-steroidal anti- enrolled continuously in the health plan (with no at a patient visit. The analysis excluded patient visits inflammatory drugs (NSAIDs) were applied to all more than one gap) for one year after discharge. with diagnoses that might contraindicate the use of ACOVE-2 patients treated with NSAIDs, not only Cholesterol screening means that a low-density anticoagulant medication. The increasing trend in those with osteoarthritis (Wenger et al. 2005). lipoprotein cholesterol (LDL-C) screening was anticoagulant use was statistically significant only performed between 60 and 365 days after the among patients ages 80 and older and those at high Chart 1:18—This chart reports baseline discharge. Cholesterol control means the patients risk for stroke. AF patients were considered at high data collected from medical records for the had LDL-C levels less than 130 mg/dl (acceptable) risk for stroke if they were older than age 75 or ACOVE-2 study (see Chart 1:5 for description or less than 100 mg/dl (optimal) measured had a diagnosis of prior transient ischemic attack of the study population). To meet the “targeted between 60 and 365 days after the discharge. or stroke (excluding intracranial hemorrhages), history” indicator, the medical record had to (Measurement is required 60 days or more after valvular heart disease, hypertension, or congestive document at least two of the following elements: discharge because LDL-C decreases temporarily heart failure. The estimated number of physician (1) characteristics of voiding, (2) ability to get following reperfusion and revascularization.) visits for AF increased during the study period, to the toilet, (3) prior treatment of urinary (NQMC 2003e). The terms “acceptable” and from 2.9 million in 1991 to 4.5 million in 2000. incontinence, and (4) importance of the “optimal” cholesterol control were chosen by the problem to the patient. To meet the “targeted chartbook authors to simplify the chart labels; Chart 1:16—Data are from the 2001 MEPS physical exam” indicator, the medical record these terms were not derived from HEDIS. Diabetes Care Survey, a self-administered paper had to document a rectal exam for men or a High blood pressure controlled refers to adults survey given to all MEPS participants identified pelvic exam for women (Wenger et al. 2005). ages 46 to 85 who had a diagnosis of hypertension as ever having had diabetes (AHRQ 2005b, or documentation of high blood pressure in their tables 1.15a, 1.16a, 1.17a, 1.18a). The reference Chart 1:19—This chart reports baseline data from medical record, were enrolled continuously in population is U.S. civilian, noninstitutionalized the IMPACT (Improving Mood: Promoting Access the health plan (with no more than one gap) adults with diabetes. Measures were derived to Collaborative Treatment) study conducted during the measurement year, and had systolic from the National Alliance for Diabetes Quality in 18 clinics affiliated with eight health care blood pressure lower than 140 mmHg and Improvement. Those who did not respond and organizations in five states (Unutzer et al. 2003, diastolic blood pressure lower than 90 mmHg those who answered “don’t know” were excluded table 3). The organizations included two staff- on the most recently recorded blood pressure from the analysis. For more information on model HMOs, two regions of a large group-model measurement. Patients with end-stage renal measure specifications and survey question HMO, the Veterans Health Administration, two disease (ESRD) are excluded (NQMC 2003f). wording, see: www.qualitytools.ahrq.gov/ university-affiliated primary care systems, and one qualityreport/browse/browse.aspx?id=5116. private practice physician group. Potential study participants were identified through referral from primary care practitioners or clinic staff, self- with a practitioner during the 12 weeks after two measures, improvement in bathing and in referral, and screening at primary care facilities. diagnosis. At least two of the three contacts ambulation/locomotion, which showed differences Based on responses to a structured interview, must have been face-to-face visits and at least between observed and risk-adjusted rates slightly patients were included if they were ages 60 and one of these visits must have been with a greater than 1 percent). Comparing changes older, intended to use one of the study clinics prescribing practitioner (NQMC 2003c). between 2002 and 2004, observed rates reflect less as their usual source of care in the coming year, Follow-up after hospitalization for mental health improvement than do risk-adjusted rates for all but and met diagnostic criteria for current major refers to discharges for health plan members one measure (improvement in toileting) (personal depression or dysthymia (chronic depressed ages six and older who were hospitalized for communication with the Center for Health mood). Those with current drinking problems treatment of selected mental health disorders Services Research, University of Colorado, 2005). or a history of bipolar disorder or psychosis, (depression, schizophrenia, attention deficit or who were in psychiatric treatment, severely disorder, and personality disorders), who were Chart 1:22—Data are from the Medicare QIO cognitively impaired, or at acute risk for suicide, enrolled continuously (without gaps) during program for 1998–1999 and 2000–2001 (Jencks were excluded. Participants were interviewed the seven- or 30-day follow-up period, and who et al. 2003, fig. 1, 2). The reference population by trained lay interviewers to collect baseline were seen on an ambulatory basis or were in day/ is Medicare fee-for-service beneficiaries. See information. Potentially effective recent depression night treatment with a mental health provider Appendix Table 1a for conditions and measures treatment was defined as taking antidepressants during the seven- or 30-day follow-up period that were included in state rankings. The Medicare for two or more months or receiving four or after hospital discharge (NQMC 2003g, 2003h). QIO program included 24 quality indicators but more psychotherapy or counseling sessions for two indicators measuring time to reperfusion depression within the past three months. Chart 1:21—OASIS measures shown in the chart were excluded from the state rankings described are those for which results were reported in 2002 in this chart. Relative improvement was defined as Chart 1:20—Data are from HEDIS for the 2000 or (AHRQ 2005b, tables 1.111 to 1.118, 1.121) and a reduction in the failure rate or quality gap. For 2001 and 2003 measurement years (NCQA 2004, 2004 (CMS 2005a). The reference population is each state, the study authors calculated “a median 23, 37). Antidepressant medication management adult, nonmaternity patients (ages 18 and older) amount of absolute and relative improvement refers to adults ages 18 and older who were of Medicare-certified home health agencies across the set of indicators in the state. [T]he diagnosed with a new episode of depression, whose episode of care was paid for by Medicare median absolute and relative national improvement treated with antidepressant medication, and or Medicaid. The denominator is episodes of [is] the median of these state medians” (Jencks enrolled continuously in the health plan (with one care for these patients that began and ended et al. 2003). State rankings were determined by allowable gap) with pharmacy and mental health in the survey year. Measures of improvement calculating each state’s rank on each of the 22 benefits during the 12 months encompassing exclude episodes of care for patients already at quality indicators and then averaging each state’s the new episode of medication therapy. the highest assessment level, since their outcomes rankings across the 22 quality indicators. Effective acute phase treatment means the cannot improve. The stabilization in bathing patient remained on an antidepressant during measure excludes episodes of care for patients Chart 1:23—Participants in the ACOVE-1 study the 12 weeks after diagnosis (NQMC 2003a). already at the lowest assessment level, since their were randomly selected from community-dwelling Effective continuation phase treatment means outcomes cannot worsen. The rates shown in patients ages 65 and older who were members of the patient remained on an antidepressant the chart are observed rates. A 2005 analysis of two managed care organizations in the United continuously during the six months (180 2002–2004 rates, conducted for the Centers for States (one in the Northeast and one in the days) after diagnosis (NQMC 2003b). Medicare and Medicaid Services, showed that the Southwest) from July 1, 1998, to July 31, 1999, and Optimal practitioner contacts means at least difference between observed and risk-adjusted who had four times the risk for functional decline three follow-up contacts for mental health rates is small (less than 1 percent for all but or death over the next two years (compared to Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 165 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 166 other elders) based on self-report or proxy-reported Postoperative pneumonia and ventilator- admissions, and patients with diseases of responses to a brief 13-item telephone screening associated pneumonia events exclude the skin, subcutaneous tissue, and breast. survey. Patients who could not speak English, who patients admitted with tracheostomies. (2) Postoperative pulmonary embolism or deep were not enrolled continuously in the managed care CVC-associated mechanical adverse events vein thrombosis per 10,000 surgical discharges, organization for at least 13 months, and who were include allergic reaction, tamponade, excluding patients admitted for deep vein receiving out-of-plan care or active treatment for perforation, pneumothorax, hematoma, thrombosis, obstetrics, and secondary malignancy (excluding non-melanoma skin cancer) shearing off of the catheter, air embolism, procedure of plication of vena cava before were excluded from the study. The final study misplaced catheter, thrombosis/embolism, or after surgery. Also excludes admissions sample consisted of 372 vulnerable elders for whom knotting of the pulmonary artery catheter, specifically for such thromboembuli, such medical records were available for abstraction. arrhythmia requiring treatment during insertion, as cases from earlier admissions, from Telephone interviews were conducted with bleeding, equipment malfunction, and pain. other hospitals, or from other settings. patients between Aug. and Oct. 2000 to determine Postoperative venous thromboembolic events include processes of care that participants had received pulmonary embolism or deep vein thrombosis (3) Postoperative sepsis per 10,000 elective- (for services not captured in medical records) occurring postoperatively during the hospital stay surgery discharges of longer than three days, and to collect demographic and functional status and readmission of surgical patients within 30 days excluding patients admitted for infection, information (Wenger et al. 2003, tables 3, 4, and 5). for pulmonary embolism or deep vein thrombosis. patients with cancer or immunocompromised For more information on measure states, and obstetric conditions. Chart 2:1—Data are from the MPSMS for 2002 specifications, see: www.qualitytools.ahrq.gov/ (4) Postoperative respiratory failure per 10,000 (AHRQ 2005b, tables 2.26 to 2.36). The reference qualityreport/browse/browse.aspx?id=5106. elective-surgery discharges, excluding population is hospitalized Medicare fee-for-service patients with respiratory disease, circulatory beneficiaries at risk of certain adverse events, Chart 2:2—Rates were calculated by the disease, and obstetric conditions. including complications after joint replacement, Medicare Payment Advisory Commission, (5) Deaths per 10,000 admissions for low- urinary tract infection or pneumonia after surgical applying AHRQ Patient Safety Indicators mortality diagnosis-related groups (DRGs) procedures, pneumonia associated with use of a (AHRQ 2003b) to all hospital claims in the with less than 0.5 percent mortality mechanical ventilator, infections associated with MedPAR file for specified years (MedPAC 2004c, (among all hospital patients, not only central venous catheters (CVCs), and bloodstream table 2-4). Rates were adjusted by age, gender, Medicare beneficiaries), excluding trauma, infections among all hospital patients. age-gender interactions, comorbidities, and immunocompromised, and cancer patients. Complications of joint replacement include diagnosis-related group (DRG) clusters. The postoperative infections, postoperative pneumonia, numerators are based on secondary diagnoses (6) Postoperative wound dehiscence (reclosure postoperative urinary tract infection, postoperative only, to exclude complications that were present of postoperative disruption of abdominal deep vein thrombosis or pulmonary embolus, on admission. The denominators are limited to wall) per 10,000 abdominopelvic surgery dislocation, wound complications other than Medicare fee-for-service beneficiaries most likely discharges, excluding obstetric conditions. infection, nerve injury, postoperative bleeding to be at risk for the complication, as described Also excludes admissions specifically for requiring four or more blood transfusions, below for the 10 indicators shown in the chart: such wound dehiscence, such as cases from cardiovascular complications, same side revision earlier admissions or from other hospitals. (1) Decubitus ulcer per 10,000 discharges of during the index hospital stay, return to the length five or more days, excluding paralysis (7) Accidental puncture or laceration during operating room for reasons other than same side patients, patients admitted from long- procedures per 10,000 discharges, excluding revision during the index hospital stay, and death. term care facilities, neonates, obstetrical obstetric admissions. Also excludes admissions specifically for such problems, such as cases numerators are based on secondary diagnoses weighted to be nationally representative after from earlier admissions or from other hospitals. only, to exclude complications that were present adjusting for the state-specific sampling scheme. (8) Infections due to medical care (primarily on admission. The denominators are limited to related to intravenous lines and catheters) hospital inpatients most likely to be at risk for the Chart 2:5—Rates were calculated by the Agency per 10,000 discharges, excluding complication, as described for Chart 2:2, above. for Healthcare Research and Quality, applying the immunocompromised patients, cancer Rates were adjusted by gender, comorbidities, 1997 Beers criteria (Beers 1997) and the 2001 Zhan patients, and neonates. Also excludes and DRG clusters. The rates per 1,000 reported expert panel criteria (Zhan et al. 2001) to the 1996, admissions specifically for such infections, in the source were converted to rates per 10,000 1998, and 2000 MEPS (AHRQ 2004, table 2.22a; such as cases from earlier admissions, from to be consistent with data reported in Chart 2:2. AHRQ 2005b, tables 2.37a, 2.37b). The MEPS other hospitals, or from other settings. Prescribed Medicines Database combines data from Chart 2:4—This baseline data from the National the household interview and a follow-back survey (9) Postoperative hip fracture per 10,000 surgical Surgical Infection Prevention Project evaluated of pharmacy providers to confirm medications discharges, excluding obstetrical patients and the medical records of 34,133 Medicare inpatients dispensed to survey participants. Beers criteria patients susceptible to falling (i.e., patients with undergoing cardiac, vascular, hip/knee, colon, or were limited to 33 drugs that should always be musculoskeletal disease; patients admitted hysterectomy surgery at one of 2,965 acute care avoided regardless of dosage, frequency, or duration for seizures, syncope, stroke, coma, cardiac hospitals nationwide from Jan. 1 through Nov. of treatment. The Zhan expert panel identified a arrest, poisoning, trauma, delirium, psychoses, 30, 2001 (Bratzler et al. 2005, tables 2, 3, 5). These subset of 11 drugs that should always be avoided in and anoxic brain injury; and patients with surgical procedures were chosen based on their the elderly. The remaining 22 drugs were classified metastatic cancer, lymphoid malignancy, frequency in the Medicare population, the rates as rarely appropriate (8 drugs) or often misused (14 bone malignancy, and self-inflicted injury). of surgical site infection, and consensus regarding drugs). The reference population is all U.S. civilian, (10) Postoperative physiologic and metabolic antibiotic prophylaxis. Performance measures were noninstitutionalized elderly adults. The number derangements per 10,000 elective-surgery developed by an expert panel based on a review of individuals affected, described in the narrative, patients, excluding obstetric admissions of the literature (Bratzler and Houck 2004). The was calculated by the chartbook authors based and some serious disease (i.e., patients analysis excluded patient records documenting on the U.S. Census count of 35 million resident with diabetic coma and patients with renal preoperative infection, antibiotic use prior to elderly Americans in 2000 (www.census.gov). failure who also were diagnosed with acute hospital admission, or more than 24 hours of myocardial infarction, cardiac arrhythmia, preoperative antibiotic prophylaxis. Measures of Chart 2:6—This study included 30,397 elderly cardiac arrest, shock, hemorrhage, antibiotic timing excluded cases that were missing Medicare beneficiaries ages 65 and older who or gastrointestinal hemorrhage). documentation of relevant dates and times. received ambulatory health care at a large Among those lacking documentation of surgical multispecialty group practice in the New England These definitions are derived from those incision time, results were similar when surgical area from July 1, 1999, to June 30, 2000 (Gurwitz et reported by the Agency for Healthcare start time was used as a proxy for incision time. al. 2003). Approximately 90 percent were enrolled Research and Quality (AHRQ 2005b). The measure of appropriate antibiotic excluded in Medicare managed care plans. Drug-related cases in which no antibiotic was given before incidents were primarily detected using (a) reports Chart 2:3—Rates were calculated by the Agency surgery, intraoperatively, or within 24 hours of from health care providers, (b) review of hospital for Healthcare Research and Quality, applying the end of surgery. This measure also excluded discharge summaries, (c) review of emergency AHRQ Patient Safety Indicators (AHRQ 2003b) colon surgery and hysterectomy patients with a department notes, and (d) review of administrative to the HCUP Nationwide Inpatient Sample for ß-lactam allergy because there are no guidelines incident reports concerning medication errors. 2001 (AHRQ 2005b, tables 2.9, 2.11a, 2.16). The for appropriate antibiotic selection. All rates were An adverse drug event was defined as an injury Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 167 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 168 resulting from the use of a drug. Adverse drug percent after exclusions. Data in the chart refer Chart 3:4—Data are unadjusted rates from the events were considered preventable if they were to 10,416 respondents for whom information was 1993, 1997, and 2002 NHIS (NCHS 1997a, table caused by an error (i.e., prescription errors, available on drug coverage. Probability sampling 1; 1997b, table 1; 2002, table 33; 2004c, table dispensing errors, patient adherence errors, and weights were applied to the data to correct for XIX). The reference population is U.S. civilian, monitoring errors). Monitoring errors included unequal sampling across states and populations. All noninstitutionalized adults of specified ages. Data inadequate laboratory monitoring of drug therapies differences in the restriction of drug use between for 1993 were based on responses to the question: or a delayed or failed response to signs, symptoms, Medicare beneficiaries without prescription drug “Is there a particular person or place that [the or laboratory evidence of drug toxicity. Some coverage and those with prescription drug coverage survey subject] usually goes to when [subject] is adverse events were associated with multiple errors shown in the chart were significant (Kitchman sick or needs advice about [his/her] health?” Data at more than one stage of pharmaceutical care. et al. 2002, fig. 13; Safran et al. 2002, exhibit 4). for 1997 and 2002 were based on the question: “Is there a place that you usually go to when you are Chart 3:1—Data are unadjusted rates from the Chart 3:3—Rates were calculated by the Medicare sick or need advice about your health?” Unknown 2002 NHIS (NCHS 2004b, table 15). The reference Payment Advisory Commission, applying the responses were excluded from the denominators. population is U.S. civilian, noninstitutionalized ACE-PRO indicators to the MCBS Cost and Use adults of specified ages. The survey questions File for 1996 through 1999 (MedPAC 2002a, table Chart 3:5—Data are from the 1996–1997 and asked: “During the past 12 months, was there 2-3). The reference population is elderly Medicare 2003 CTS Household Survey (Trude and Ginsburg any time when [person] needed medical care fee-for-service beneficiaries ages 65 and older. 2005, table 5). The reference population is U.S. but did not get it because [person] could not “Some supplemental coverage” refers to individuals civilian, noninstitutionalized adults of specified afford it?” and “During the past 12 months has with at least six months of additional coverage in a ages. Appointment waiting times of more than [person] delayed seeking medical care because year (i.e., employer-sponsored insurance, Medigap 90 days were excluded to remove the effects of worry about the cost?” Both questions insurance, Medicaid). The analysis included of potential standing appointments (personal excluded dental care. Unknown responses 22 indicators that were applicable to at least 20 communication with Sally Trude 2005). The were excluded from the denominators. survey participants covered only by Medicare. change in waiting times from 1997 to 2003 for all The chart shows 10 of 11 indicators for which physician visits and for specialist physician visits Chart 3:2—The 2001 Survey of Seniors was there was a statistically significant difference in was statistically significant for both near-elderly administered by mail and telephone to community- the use of services between Medicare beneficiaries and elderly patients. There was no statistically dwelling Medicare beneficiaries (ages 65 and without supplemental coverage and those with significant change in waiting times for primary older) living in eight U.S. states (Ill., Mich., some supplemental coverage (rates of annual care physician visits for a specific illness. Data on N.Y., Pa., Calif., Colo., Ohio, Tex.). The elderly physician visits also showed a difference but are waiting times for checkup visits are not shown. population of these states represents 42 percent of not shown). The Medicare Payment Advisory all U.S. adults ages 65 and older. Four of the eight Commission notes that “other factors, such as Chart 3:6—Rates were calculated by the Medicare states had established, state-funded pharmacy- education or income, may be correlated with Payment Advisory Commission from a 5 percent assistance programs. The Centers for Medicare both the necessary care indicators and insurance sample of the Medicare enrollee database and Medicaid Services provided a 10 percent status, and may therefore confound our results. (MedPAC 2004a, fig. 6-1, 6-2). The reference probability sample of community-dwelling Multivariate analysis might show a smaller impact populations are all Medicare beneficiaries (left Medicare beneficiaries ages 65 and older in each from having additional coverage, but would not be chart) and elderly Medicare fee-for-service state, which included information about Medicaid likely to eliminate the effect” (MedPAC 2002a). beneficiaries ages 65 and older (right chart). coverage. The survey was administered in either English or Spanish and had a response rate of 55 Chart 4:1—Data are from the 2001 CMWF Had no problems getting needed care is a published by CMS from the MCBS Access to Health Insurance Survey, conducted from composite of four questions asking how often, Care File (CMS 2000a, table 7-1; 2004a, table 7-1). Apr. 27 through July 29, 2001 (Davis et al. 2002, in the last six months, respondents had any The reference population is community-dwelling, exhibit 2). “In the analyses, persons with more problems with: 1) finding a personal doctor or elderly Medicare beneficiaries ages 65 and older. than one source of coverage were assigned nurse, 2) getting a referral to a specialist that they The right chart presents results of a Government hierarchically to the Medicare, Medicaid, wanted to see, 3) getting the care they and their Accountability Office audit based on 420 calls employer, and individual insurance categories, doctor believed necessary, and 4) getting care to the 1-800-MEDICARE beneficiary help line so that Medicare beneficiaries with supplemental approved by the health plan without delays. operated by two contractors for the Centers for coverage such as Medicaid, retiree coverage, Rated their health plan the best possible is based Medicare and Medicaid Services (CMS). Calls or Medigap coverage are categorized as being on a question asking respondents to rate all their were randomly placed during July 2004 to match Medicare enrollees.” Most (91%) of those with experiences with their health plan, using a scale typical calling patterns for the help line. For each private coverage were in employer-sponsored from 0 (worst possible plan) to 10 (best possible call, auditors asked one of six questions about the plans. In regression analysis controlling for health plan). Rated their care the best possible is based on Medicare program that were preselected from status, poverty, and other factors (not shown), a question asking respondents to rate the care they among the 100 questions most frequently addressed elderly Medicare beneficiaries were significantly received in the last six months from all doctors and by the help line. Answers were considered more likely to rate health insurance excellent providers in their health plan, using a scale from inaccurate if they did not provide “sufficient and and to report being very satisfied with overall 0 (worst possible care) to 10 (best possible care). complete” information to meet criteria developed quality of care, and were significantly less likely to Said that doctors in their plan always from the Medicare Web site’s frequently asked report negative plan experiences and any access communicated well is a composite of four questions section. In the six months previous problems due to cost, as compared to nonelderly questions asking how often, in the last six months, to the audit, the call volume to the Medicare adults with employer-sponsored insurance. respondents’ doctors or other health providers: 1) help line had more than tripled in response to listened carefully, 2) explained things in a way they changes brought about by passage of the Medicare Chart 4:2—Data are from the Medicare CAHPS could understand, 3) showed respect for what they Modernization Act of 2003 (GAO 2004a). survey for 2003 (CMS 2005c). The reference had to say, and 4) spent enough time with them. populations are Medicare fee-for-service Charts 4:5 to 4:7—Data are from a mortality beneficiaries and Medicare managed care plan Chart 4:3—Data are from the 2001 MEPS Self- follow-back study of 1,578 adults who died of members. Rates were case-mix adjusted. Administered Questionnaire (SAQ), which collects chronic illness in 2000 (Teno et al. 2004, table 3). Always got care when needed without long respondents’ perceptions of health care quality Interviews were conducted with informants listed waits is a composite of four questions asking how using questions from the CAHPS survey (AHRQ on death certificates, who were typically family often, in the last six months, respondents: 1) got 2005b, tables 4.1a, 4.3a, 4.5a, 4.7a). The reference members, or someone else whom the informant help or advice they needed when they called the population is U.S. civilian, noninstitutionalized recommended as knowledgeable about the person’s doctor’s office during regular office hours, 2) adults of specified ages who visited a doctor’s death (response rate 65 percent). Most interviews got treatment as soon as they wanted when they office in the past year (78 percent of those ages were conducted between nine and 15 months needed to be seen right away for an illness or injury, 45–64 and 88 percent of those ages 65 and older). after the patient died. Respondents were asked 3) got an appointment as soon as they wanted Those who did not respond or who answered about the quality of health care at the last place the for regular or routine health care, and 4) waited “don’t know” were excluded from the analysis. patient spent at least 48 hours. Those who died only 15 minutes or less past their appointment at home without any formal care (13 percent of time to see the person they went to see. Chart 4:4—The chartbook authors calculated rates study subjects) were excluded from the analysis. reported in the left chart using summary statistics Data were weighted to be nationally representative. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 169 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 170 After controlling for potentially confounding adjusted using two age groups: ages 65–74 and Medicaid Services (CMS). Individuals were factors (decedent’s age, years of education, sex, race, ages 75 and older (DHHS 2002b). For the other excluded if their HEDIS and CMS data could not underlying cause of death, and the respondent’s measures, rates by race/ethnicity and by income be merged, if their race was classified as unknown perceptions of whether death was unexpected were drawn from unadjusted age-specific categories or other, if there was no evidence of managed and the decedent’s functional status), ratings for in the source; rates by type of insurance were not care enrollment, or if they were younger than 65 pain control, emotional support, and treating age-adjusted for sigmoidoscopy/colonoscopy, years of age in 1999. All measures were adjusted patients with respect were significantly better for according to the source. Numbers were too for age and sex using direct standardization those who died at home with hospice care than small to report mammography rates for Asian methods. Persons of Native American race were for those who received all other types of care. Americans or any measure for Native Americans. not included in the analysis for some measures Comparing home care, hospital care, and nursing because their numbers were too small to report. home care to hospice care at home, ratings were Chart 5:5—Data are from the HRS 1931–1941 Compared to rates of treatment among white significantly worse for four, six, and seven of the birth cohort (McWilliams et al. 2003, table patients, the following rates were significantly nine measures shown in the chart, respectively. 2). Analysis was limited to a subset of 2,203 different: beta-blocker treatment among black HRS participants who were ages 60 to 64 when patients; hemoglobin A1c testing among black Chart 5:1—Rates were calculated by the Agency interviewed in 1996 and who were interviewed and Asian patients; high blood pressure control for Healthcare Research and Quality, applying again in 2000 when they were age-eligible for among black patients; effective acute-phase AHRQ Patient Safety Indicators (AHRQ 2003b) Medicare. Based on self-reported insurance status, antidepressant treatment among black and Asian to the HCUP State Inpatient Database for 2001 167 participants were classified as continuously patients; 30-day follow-up after hospitalization (AHRQ 2005a, tables 101a, 104a, 108a). Rates were uninsured before age 65 if they were uninsured in for mental illness among black patients. adjusted by age, gender, age-gender interactions, both 1994 and 1996. Likewise, 1,820 participants comorbidities, and diagnosis-related group (DRG) were classified as continuously insured before Chart 5:7—Rates were calculated by the Medicare clusters. The numerators are based on secondary age 65 if they were insured in both 1994 and Payment Advisory Commission from a 5 percent diagnoses only, to exclude complications that 1996. Differences in rates of screening between sample of the Medicare enrollee database (MedPAC were present on admission. The denominators are the continuously uninsured and continuously 2004a, fig. 6-3). The reference population is limited to hospital inpatients most likely to be at insured groups were significantly reduced after Medicare fee-for-service beneficiaries. risk for the complication, as described for Chart Medicare eligibility. There was no significant 2:2, above. The rates per 1,000 reported in the change in the difference in screening among Chart 5:8—This study used data from the source were converted to rates per 10,000 to be those who were intermittently uninsured in 1994 Medicare Denominator file and the MedPAR file to consistent with the data reported in Chart 2:2. or in 1996, but not in both years (not shown). identify individuals who were enrolled in Medicare Part A and Part B for at least two years before their Charts 5:2 to 5:4—Data are from the 1998, Chart 5:6—This chart displays data from two death and who had at least one nonsurgical hospital 2000, and 2001 NHIS (AHRQ 2005a, tables 1a, studies that analyzed HEDIS data applicable to admission for one of 11 chronic conditions in the 1b, 1c, 3a, 3b, 3c, 24a, 24b, 24c, 25a, 25b, 25c, 7,498,496 Medicare beneficiaries enrolled in 301 last two years of life. Each patient was assigned to 68a, 68b, 68c, 72a, 72b, 72c). The population Medicare managed care plans for the reporting year the hospital that the patient most frequently used categories are shown as reported in the source. 2000 (based on 1999 experience) (Virnig et al. 2002, in the last two years of life; ties were decided in Immunization rates were age-adjusted to the exhibit 2; Virnig et al. 2004, tables 3, 4). Individual- favor of the hospital discharge closest to the date of 2000 U.S. standard population following methods level HEDIS records were linked with information death. The final analysis included 90,616 patients used to track Healthy People 2010 goals, which on age, race, sex, and state and county of residence who died in 1999–2000 and most frequently used specify that NHIS data for the elderly cohort be obtained from the Centers for Medicare and one of 77 hospitals listed in U.S. News and World Report’s 2001 rankings of the best U.S. hospitals for Resource File summed across all counties in availability of resources in the county in which the geriatric care and for care of heart and pulmonary each state. Determinants of state spending and practice was located (Bach et al. 2004, table 2). diseases. Utilization rates were adjusted to control quality were examined using generalized least for differences in patients’ age, sex, race, and squares regression weighted by the size of the Chart 6:1—This before-and-after study evaluated clinical comorbidities. To control for differences Medicare population in each state. Increased the medical charts of patients hospitalized for in the severity of illness, the analysis was restricted state spending was associated with statistically pneumonia in the Louisiana State University to care delivered in the last six months of life. significant reductions in rates for 15 of the 24 QIO Internal Medicine ward between July 2000 and Results focused on patient cohorts with solid tumor indicators; there was no significant effect for the June 2001 (pre-intervention) and July 2001 and cancer, chronic obstructive pulmonary disease, or other nine indicators. In a separate analysis, there June 2002 (post-intervention). A total of 435 congestive heart failure. Following the principles was no correlation between changes in state-level hospitalizations were evaluated to determine the of population-based epidemiology, utilization Medicare spending per beneficiary and changes percentage of patients who were screened for or rates were based on “the total experience of the in rates of four quality measures from 1995 to received pneumococcal vaccination. Measures cohort, not just on services provided by the index 1999 (Baicker and Chandra 2004, exhibit 1). were the same as those used in the Medicare QIO hospital and associated providers. However, since program. Patients who received nonacute care, the percentage of total hospital care provided by Chart 5:10—This study involved a cross-sectional transferred from another acute care facility, did the index hospital is high, the variations in illness- analysis of 150,391 patient visits for evaluation and not have pneumonia, or died while in the hospital adjusted use of care primarily reflect clinical management by 43,032 black and white Medicare were excluded from the analysis. There were no choices made by physicians associated with that beneficiaries ages 65 and older who were treated significant differences in patient demographics hospital” (Wennberg et al. 2004b, exhibit 2). by 4,355 primary care physicians (family or general (age and sex) or length of stay between the pre- practice, general internal medicine, or geriatrics). and post-intervention patients. All differences Chart 5:9—This study used states’ overall Data on patient visits from the Medicare “5 percent between the pre- and post-intervention groups rankings on 22 indicators of the quality of care, carrier file” were linked with survey data for 77 shown in the chart were statistically significant. as measured by the Medicare QIO program for percent of the physicians who participated in The educational intervention was implemented Medicare fee-for-service beneficiaries during the 2000–2001 CTS Physician Survey. Results from July 2001 through June 2002. A questionnaire 2000–2001. (See Appendix Table 1a for a list of the were weighted to be nationally representative. was administered to Internal Medicine house staff measures. The Medicare QIO program included The unit of analysis was the patient visit; some prior to and at the conclusion of the intervention 24 quality indicators but two indicators measuring patients saw more than one physician in the survey. to assess their understanding of the benefits time to reperfusion were excluded from the state Physicians who saw both black and white patients and indications of pneumococcal vaccination. rankings described in this chart.) Detailed risk are represented in both counts of visits by black Statistically significant improvements in adjustment has not been found critical when using patients and visits by white patients. The majority questionnaire scores were observed for all house such process-of-care measures for population- of visits by both white and black patients were with staff combined (Kruspe et al. 2003, tables 2, 4, 5). based analyses. Medicare fee-for-service claims white physicians, although black patients were data were used to calculate Medicare spending more likely to visit black physicians. Differences in Chart 6:2—The Guidelines Applied in Practice per beneficiary at the state level. Spending was measures for physicians visited by black patients (GAP) initiative of the American College of adjusted for inflation, differences in state price and physicians visited by white patients shown Cardiology was a comparative before-and-after levels, and the age, sex, and race of each state’s in the chart were statistically significant in both study that included Medicare and non-Medicare Medicare population. The numbers of specialists, unadjusted and adjusted analyses. The adjusted patients treated for acute myocardial infarction primary care physicians, and registered nurses analysis included measures of payer mix, median (AMI) at 10 acute-care hospitals in southeastern were determined using data from the 2003 Area income within the ZIP code of the practice, and Michigan. The chart is based on a subgroup Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 171 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 172 analysis of results for Medicare beneficiaries weeks after discharge to collect information on 12-month follow-up. The intervention effect identified from Medicare claims data (N = 515 resource use, functional status, quality of life, and increased over time but varied by organization at baseline; N = 663 at follow-up). Baseline data patient satisfaction. Resource costs were estimated (Unutzer et al. 2002, tables 3, 4). For more were collected from a random sample of medical using standardized Medicare reimbursement information, see: www.impact.ucla.edu. records for patients discharged from July 1998 rates; intervention costs were calculated based on to June 1999. Follow-up data were collected clinicians’ billable time devoted to intervention- Chart 6:5—The New York University Spouse- from medical records for all Medicare patients related efforts. All differences between the Caregiver Intervention Study was a randomized discharged from Sept. to Dec. 2000. Medical intervention and control groups shown in the controlled trial that recruited subjects through the record abstraction was performed by a central chart were statistically significant. In multivariate New York University Alzheimer’s Disease Center, center. The denominator for a quality indicator Cox proportional hazards analysis, the time to local chapters of the Alzheimer’s Association, included only eligible patients who did not have first rehospitalization or death was significantly media announcement, and physician referrals. a documented contraindication for the specific longer for patients in the intervention group. The patient or caregiver had to have at least one treatment. Follow-up rates shown in the chart, Efficacy did not vary by hospital although the other relative living in the nearby area. A total of which represent a subset of patients for whom intervention effect decreased over time. Only short- 406 spouse-caregivers (average age 71 years) were there was chart documentation of GAP-promoted term improvements were seen in quality of life randomly assigned to receive either enhanced tool use, were significantly greater than rates at and patient satisfaction (Naylor et al. 2004, tables counseling and support treatment or usual care. follow-up for a control group of Medicare patients 3, 4). For more information, see: www.nursing. Spouse-caregivers in the intervention group were at 11 nonparticipating hospitals. Data for the upenn.edu/centers/hcgne/TransitionalCare.htm. more often wives (66% vs. 55%) and had lower control group were collected for a public profiling depression scores. Following a comprehensive project among southeast Michigan hospitals Chart 6:4—The IMPACT (Improving Mood: battery of questionnaires answered at baseline, from Jan. to Dec. 1998 at baseline (N = 513) and Promoting Access to Collaborative Treatment) caregivers were interviewed in person or by from Mar. to Aug. 2001 at follow-up (N = 388) study was a randomized controlled trial that telephone every four months during the first (Mehta et al. 2002, table 4). For more information, enrolled 1,801 depressed individuals ages 60 and year and every six months thereafter to assess see: www.acc.org/gap/mi/ami_gap.htm. older who were treated at one of 18 primary care caregiver depression status and patient dementia facilities in five states (see the entry for Chart status. Caregiver depression was assessed using Chart 6:3—This randomized controlled trial 1:19 for information on study recruitment). the Geriatric Depression Scale. Follow-up included 239 eligible patients ages 65 and older Research assistants blinded to the study conditions interviews were conducted until two years after who were admitted from their homes to one of six conducted an in-person baseline interview and the death of the patient or until caregivers refused Philadelphia academic and community hospitals follow-up telephone interviews with patients at or were no longer able to participate. The graphs between Feb. 1997 and Jan. 2001 with a diagnosis three, six, and 12 months to collect information represent predicted values of the depression scores, of heart failure. To be included, patients had to regarding the severity of depressive symptoms, controlling for covariates, using random effects speak English, be alert and oriented, be reachable health-related functional impairment, overall growth curve modeling. Caregivers’ mean-centered by telephone after discharge, and reside within 60 quality of life in the past month, satisfaction baseline depression scores and gender were miles of the hospital. Patients with end-stage renal with depression care, and use of antidepressant included as covariates in the model to equalize rates disease were excluded because of their unique medications, counseling, or psychotherapy at baseline. The best-fitting longitudinal change service needs. Research assistants blinded to study within the past three months. All measured pattern was selected for the first year (logarithmic assignment interviewed patients in the hospital differences between the intervention and model) and follow-up years (linear model) to obtain baseline information and conducted control groups were statistically significant in (Mittelman et al. 2004, fig. 1, 2). The median time telephone interviews at two, six, 12, 26, and 52 adjusted regression analysis at three-, six-, and before nursing home placement is a weighted average of Kaplan-Meier survival estimates for in the hospitalization rate of 0.4 percent from Year pulmonary disease, or cancer was performed; men and women (Mittelman et al. 1996). For 1 to Year 2 and a 0.3 percent decrease from Year 2 161 patients were enrolled in the KP Palliative more information, see: http://aging.med.nyu. to Year 3 for the comparison patients (Shaughnessy Care Project intervention. The comparison group edu/programs/clinicalresearch/adrc/psychosocial. et al. 2002a, fig. 3). For more information, see: included 139 patients who received usual Medicare https://www2.uchsc.edu/chsr/center/meqa.shtml. home care. All patients had a life expectancy of Chart 6:6—This comparative before-and- less than 24 months. Palliative care patients could after study included 157,548 patients admitted Chart 6:7—The Program of All-Inclusive Care for maintain their primary care physician while being over three years to 54 Outcome-Based Quality the Elderly (PACE) was a federal demonstration treated at home by the palliative care physician. Improvement (OBQI) agencies participating in project at the time of the analysis. Data for this Pain was assessed at each home visit, and 24-hour the national demonstration trial in 27 states and comparison study came from interviews with telephone support was provided to palliative care 105,917 patients admitted over four years to 19 PACE applicants conducted at the patients’ patients. Data on resource use were obtained OBQI agencies participating in the New York State homes between Jan. 1995 and Aug. 1997. The from the KP service utilization database. Research demonstration trial. The trials occurred from 1995 treatment group included those who decided assistants blinded to the group assignments to 2000. OASIS data on 41 outcome measures to enroll in PACE and were accepted into one conducted telephone interviews with patients were collected for each patient within the OBQI of 11 PACE demonstration sites prior to the seven days after their enrollment in the study and program at the start of care and every 60 days until follow-up interview. Those in the comparison every 60 days thereafter to obtain information discharge, when final data were collected. For each group qualified for PACE but decided not to regarding demographics, the severity of illness, pre/post comparison shown in the chart, the first enroll in the program. Participants were excluded and satisfaction with services. Satisfaction was year is risk-adjusted (using logistic regression) if they refused to participate in the baseline measured using the Reid-Gundlach Satisfaction and the second year is an observed rate. Risk survey or had missing information, had missing with Services instrument. Scores at 60 days models were revised for each annual reporting information for the enrollment decision, or had post-enrollment were 43.55 and 40.97 for the period. Rates differ between comparison periods missing information on the outcome of interest. intervention and usual care groups, respectively, as because of risk-adjustment and differences in PACE outcomes were measured during follow-up compared to 41.13 and 40.19 at baseline. The mean sample sizes (a few agencies were excluded in some interviews at six months (N = 1,098), 12 months satisfaction score was significantly higher at 60 days years because of data reporting issues). The net (N = 783), 18 months (N = 529), and 24 months post-enrollment only for the intervention group. decreases in rates between each year in the chart (N = 296) following the baseline survey. Only For purposes of the chart, satisfaction scores were were statistically significant. The 22 percent relative the 12-month results are shown in the chart. converted into a percentage by dividing them by the decrease in hospitalization rate for the national All differences shown in the chart between the total possible score of 48. Costs of care (not shown) demonstration (described in the narrative) reflects PACE group and the comparison group, except were calculated based on 1999 staff salary rates; a risk-adjusted net decrease of 7.2 percentage in health status, were statistically significant medication, facility, and administrative costs were points when Year 3 is compared to Year 1. The 26 in regression analyses controlling for baseline not included. Rates of service use were adjusted to percent relative decrease in the New York State characteristics (Chatterji et al. 1998, exhibit 14). control for days enrolled, congestive heart failure demonstration reflects a risk-adjusted net decrease For more information, see: www.npaonline.org. diagnosis, and severity of illness. All between- of 7.9 percentage points when Year 4 is compared group differences in service use shown in the chart to Year 1. A comparative analysis used Medicare Chart 6:8—This comparison study was conducted were statistically significant (Brumley et al. 2003b, claims data for a 5 percent random sample of home at the Kaiser Permanente (KP) TriCentral Service fig. 1, 2, 3; personal communication with Susan health patients in the same 27 states who were Area located in southern California. A subgroup Enguidanos 2005). For more information on this cared for by agencies that were not participating in analysis of 300 patients who died during the study intervention, see: www.growthhouse.org/palliative. the demonstration. This analysis found a decrease from congestive heart failure, chronic obstructive Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 173 References AAOS (American Academy of Orthopaedic Surgeons). 2003. AAOS ——— 2002c. Elderly people find new Medicare information Baicker, K. and A. Chandra. 2004. Medicare spending, the clinical guideline on osteoarthritis of the knee—Phase . materials useful in navigating the maze of health physician workforce, and beneficiaries’ quality of care. Chicago, Ill.: American Academy of Orthopaedic Surgeons. plan choices. Research Activities, April. Health Aff (Millwood) Suppl Web Exclusive: W84-97. ACR (American College of Rheumatology). 2000. Recommendations ——— 2002d. Managing osteoarthritis: Helping the elderly Baker, R., et al. 2000. Family satisfaction with end-of-life care in seriously for the medical management of osteoarthritis of the hip maintain function and mobility. Research in action ill hospitalized adults. J Am Geriatr Soc 48(5 Suppl): S6-9. and knee: 2000 update. Arthritis Rheum 43(9): 905-5. no. 4. AHRQ Pub. No. 02-0023. Rockville, Md.: U.S. Bartels, S. J., et al. 2004. Improving access to geriatric mental health Department of Health and Human Services. ADA (American Diabetes Association). 2004. Diabetes statistics for services: A randomized trial comparing treatment engagement seniors. Washington, D.C.: American Diabetes Association. ——— 2003a. General questions about the AHRQ QIs. Rockville, with integrated versus enhanced referral care for depression, Md.: U.S. Department of Health and Human Services. anxiety, and at-risk alcohol use. Am J Psychiatry 6(8): 455-62. Adler, G. S. and A. Shatto. 2002. Screening for osteoporosis and colon cancer under Medicare. Health Care Financ Rev 23(4): 89-200. ——— 2003b. AHRQ quality indicators: Guide to patient safety Bartlett, J. G., et al. 2000. Practice guidelines for the management indicators. AHRQ Pub. 03-R203, revision . Rockville, of community-acquired pneumonia in adults. Infectious AGS (American Geriatrics Society). 998. The management of Md.: U.S. Department of Health and Human Services. Diseases Society of America. Clin Infect Dis 3(2): 347-82. chronic pain in older persons: AGS Panel on Chronic Pain in Older Persons. J Am Geriatr Soc 46(5): 635-5. ——— 2004. National healthcare quality report, 2003. Beers, M. H. 997. Explicit criteria for determining potentially AHRQ Pub. No. 04-RG004. Rockville, Md.: U.S. inappropriate medication use by the elderly. An ——— 200. Exercise prescription for older adults Department of Health and Human Services. update. Arch Intern Med 57(4): 53-6. with osteoarthritis pain: Consensus practice recommendations. J Am Geriatr Soc 49(6): 808-23. ——— 2005a. National healthcare disparities report, 2004. Bellomo, R., et al. 2003. A prospective before-and-after trial of AHRQ Pub. No. 05-004. Rockville, Md.: U.S. a medical emergency team. Med J Aust 79(6): 283-7. AGS/BGS/AAOS Panel on Falls Prevention. 200. Guideline for the Department of Health and Human Services. prevention of falls in older persons. American Geriatrics Society- Berenholtz, S. M., et al. 2004. Eliminating catheter-related bloodstream British Geriatric Society-American Academy of Orthopaedic ——— 2005b. National healthcare quality report, 2004. infections in the intensive care unit. Crit Care Med 32(0): 204-20. Surgeons Panel on Falls Prevention. J Am Geriatr Soc 49(5): 664-72. AHRQ Pub. No. 05-003. Rockville, Md.: U.S. Bergstrom, N. I. 997. Strategies for preventing pressure Department of Health and Human Services. AHA (American Heart Association). 2005a. Atrial fibrillation. ulcers. Clin Geriatr Med 3(3): 437-54. Dallas, Tex.: American Heart Association. AMA (American Medical Association). 2003. Complete medical Berry, L. L., et al. 2003. Innovations in access to care: A patient- encyclopedia. Edited by J. B. Leikin and M. S. Lipsky. ——— 2005b. Heart disease and strokes statistics, 2005 update. centered approach. Ann Intern Med 39(7): 568-74. New York, N.Y.: Random House Reference. Dallas, Tex.: American Heart Association. Bezon, J., et al. 999. Nursing outcome indicator: Preventing falls for Andrade, S. E., et al. 2003. Low frequency of treatment of AHRQ (Agency for Healthcare Research and Quality). 992. elderly people. Outcomes Manag Nurs Pract 3(3): 2-6; quiz 6-7. osteoporosis among postmenopausal women following Pressure ulcers in adults: Prediction and prevention. Clinical a fracture. Arch Intern Med 63(7): 2052-7. Bindman, A. B., et al. 995. Preventable hospitalizations and practice guideline no. 3. AHRQ Pub. No. 92-0047. Rockville, access to health care. JAMA 274(4): 305-. Md.: U.S. Department of Health and Human Services. Angeja, B. G., et al. 2002. Predictors of door-to-balloon delay in primary angioplasty. Am J Cardiol 89(0): 56-6. Bishop, C. E., et al. 2002. Medicare spending for injured elders: Are there ——— 996. Urinary incontinence in adults: Acute and chronic opportunities for savings? Health Aff (Millwood) 2(6): 25-23. management. Clinical practice guideline number 2, 996 Antman, E. M., et al. 2004. ACC/AHA guidelines for the management update. AHRQ Pub. No. 96-0682. Rockville, Md.: U.S. of patients with ST-elevation myocardial infarction: A report of Blustein, J. and L. J. Weiss. 998. The use of mammography Department of Health and Human Services. the American College of Cardiology/American Heart Association by women aged 75 and older: Factors related to health, Task Force on Practice Guidelines. Circulation 0(9): e82-292. functioning, and age. J Am Geriatr Soc 46(8): 94-6. ——— 998. Consumer assessment of health plans overview. Rockville, Md.: U.S. Department of Health and Human Services. Asch, S. M., et al. 2000. Measuring underuse of necessary care BMS (Brown Medical School). 2004. Brown atlas of dying. among elderly Medicare beneficiaries using inpatient Providence, R.I.: Brown University, Center for ——— 2002a. AHRQ quality indicators: Guide to inpatient quality and outpatient claims. JAMA 284(8): 2325-33. Gerontology and Health Care Research. indicators. AHRQ Pub. No. 02-RO204. Rockville, Md.: U.S. Department of Health and Human Services. Ayanian, J. Z., et al. 2000. Unmet health needs of uninsured Born, W., et al. 2004. Knowledge, attitudes, and beliefs about adults in the United States. JAMA 284(6): 206-9. end-of-life care among inner-city African Americans ——— 2002b. AHRQ quality indicators: Guide to prevention quality and Latinos. J Palliat Med 7(2): 247-56. indicators. AHRQ Pub. No. 02-R0203, revision 2. Rockville, Bach, P. B., et al. 2004. Primary care physicians who treat Md.: U.S. Department of Health and Human Services. blacks and whites. N Engl J Med 35(6): 575-84. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 175 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 176 Brach, C. and I. Fraser. 2000. Can cultural competency reduce ——— 2000. Updated recommendations from the Advisory Chiquette, E., et al. 998. Comparison of an anticoagulation clinic with racial and ethnic health disparities? A review and Committee on Immunization Practices in response to usual medical care: Anticoagulation control, patient outcomes, conceptual model. Med Care Res Rev 57 Suppl : 8-27. delays in supply of influenza vaccine for the 2000-0 and health care costs. Arch Intern Med 58(5): 64-7. season. MMWR Morb Mortal Wkly Rep 49(39): 888-92. Bratzler, D. W. and P. M. Houck. 2004. Antimicrobial prophylaxis Chobanian, A. V., et al. 2003. The seventh report of the Joint National for surgery: an advisory statement from the National Surgical ——— 2002. Preventive-care practices among persons with Committee on Prevention, Detection, Evaluation, and Infection Prevention Project. Clin Infect Dis 38(2): 706-5. diabetes—United States, 995 and 200. MMWR Treatment of High Blood Pressure. JAMA 289(9): 2560-72. Morb Mortal Wkly Rep 5(43): 965-9. Bratzler, D. W., et al. 2005. Use of antimicrobial prophylaxis for CMS (Centers for Medicare and Medicaid Services). 2000a. The major surgery: Baseline results from the National Surgical ——— 2003a. Facilitating influenza and pneumococcal characteristics and perceptions of the Medicare population: Data Infection Prevention Project. Arch Surg 40(2): 74-82. vaccination through standing orders programs. from the 998 Medicare Current Beneficiary Survey. Baltimore, MMWR Morb Mortal Wkly Rep 52(4): 68-9. Md.: U.S. Department of Health and Human Services. Breen, N., et al. 200. Progress in cancer screening over a decade: Results of cancer screening from the 987, 992, and 998 National ——— 2003b. Public health and aging: Influenza vaccination coverage ——— 2000b. Medicare priorities: Medicare’s health care quality Health Interview Surveys. J Natl Cancer Inst 93(22): 704-3. among adults aged > or =50 years and pneumococcal vaccination improvement program. CMS Pub. No. 056. Baltimore, coverage among adults aged > or =65 years—United States, Md.: U.S. Department of Health and Human Services. Brown, A. F., et al. 2003. Guidelines for improving the care 2002. MMWR Morb Mortal Wkly Rep 52(4): 987-92. of the older person with diabetes mellitus. J Am ——— 2003. Medicare quality improvement priorities. CMS Pub. No. 04. Geriatr Soc 5(5 Suppl Guidelines): S265-80. ——— 2003c. Trends in aging—United States and worldwide. Baltimore, Md.: U.S. Department of Health and Human Services. MMWR Morb Mortal Wkly Rep 52(6): 0-4, 06. Brown, M. L., et al. 200. The burden of illness of cancer: Economic ——— 2004a. The characteristics and perceptions of the cost and quality of life. Annu Rev Public Health 22: 9-3. ——— 2004a. Influenza and pneumococcal vaccination coverage Medicare population: Data from the 2002 Medicare among persons aged > or =65 years and persons aged 8- Current Beneficiary Survey. Baltimore, Md.: U.S. Brumley, R. D., et al. 2003a. Effectiveness of a home-based palliative 64 years with diabetes or asthma—United States, 2003. Department of Health and Human Services. care program for end-of-life. J Palliat Med 6(5): 75-24. MMWR Morb Mortal Wkly Rep 53(43): 007-2. ——— 2004b. Guide to Medicare’s preventive services. CMS Pub. No. 00. ——— 2003b. The palliative care program. Permanente Journal 7(2): 7-2. ——— 2004b. Influenza vaccination and self-reported reasons Baltimore, Md.: U.S. Department of Health and Human Services. Burack, R. C., et al. 998. Health status and mammography use for not receiving influenza vaccination among Medicare ——— 2004c. Summary of draft 8th statement of work. among older women. J Gen Intern Med 3(6): 366-72. beneficiaries aged > or =65 years—United States, 99- Baltimore, Md.: Center for Medicare and Medicaid ——— 2000. Mammography use among women as a 2002. MMWR Morb Mortal Wkly Rep 53(43): 02-5. Services, Office of Clinical Standards and Quality. function of age and patient involvement in decision- ——— 2005. Get smart: know when antibiotics work. ——— 2005a. Home health compare. Rockville, Md.: U.S. making. J Am Geriatr Soc 48(7): 87-2. Background on antibiotic resistance. Atlanta, Ga.: U.S. Department of Health and Human Services. http:// Burgio, K. L., et al. 998. Behavioral vs drug treatment for urge Department of Health and Human Services. www.medicare.gov/HHCompare/Home.asp urinary incontinence in older women: A randomized CDC/MIAH (Centers for Disease Control and Prevention and ——— 2005b. Information on hospital quality now available. Press release. controlled trial. JAMA 280(23): 995-2000. the Merck Institute of Aging and Health). 2004. The Baltimore, Md.: U.S. Department of Health and Human Services. Burwen, D. R., et al. 2003. National and state trends in quality state of aging and health in America 2004. Washington, D.C.: Merck Institute of Aging and Health. ——— 2005c. Medicare personal plan finder. Baltimore, of care for acute myocardial infarction between 994- Md.: U.S. Department of Health and Human 995 and 998-999: The Medicare health care quality Cegala, D. J., et al. 200. The effects of patient communication Services. http://www.medicare.gov improvement program. Arch Intern Med 63(2): 430-9. skills training on the discourse of older patients during a primary care interview. J Am Geriatr Soc 49(): 505-. ——— 2005d. Medicare physician group practice demonstration. Baltimore, Car, J. and A. Sheikh. 2003. Telephone consultations. Md.: U.S. Department of Health and Human Services. BMJ 326(7396): 966-9. Chang, J. T., et al. 2004. Interventions for the prevention of falls in older adults: Systematic review and meta-analysis ——— 2005e. Unprecedented beneficiary support in getting CDC (Centers for Disease Control and Prevention). 995. Knowledge, of randomised clinical trials. BMJ 328(744): 680. the most help from Medicare’s improved coverage. attitudes, and practices of physicians regarding urinary CMS Medicare Reform Issue Papers 5. Baltimore, Md.: incontinence in persons aged > or = 65 years—Massachusetts Chassin, M. R. 99. Quality of care. Time to act. JAMA 266(24): 3472-3. U.S. Department of Health and Human Services. and Oklahoma. MMWR Morb Mortal Wkly Rep 44: 747, 753-4. Chatterji, P., et al. 998. Evaluation of the program of all-inclusive care Covinsky, K. E., et al. 2003. Patient and caregiver characteristics ——— 997. Prevention of pneumococcal disease: Recommendations for the elderly (PACE) demonstration. The impact of PACE on associated with depression in caregivers of patients of the Advisory Committee on Immunization participant outcomes. Cambridge, Mass.: Abt Associates, Inc. with dementia. J Gen Intern Med 8(2): 006-4. Practices. MMWR Recomm Rep 46(RR-8): -24. Chaudhry, S. I., et al. 2004. Systolic hypertension in Crystal, S., et al. 2003. Diagnosis and treatment of depression in ——— 999. Reasons reported by Medicare beneficiaries for not older persons. JAMA 292(9): 074-80. the elderly Medicare population: Predictors, disparities, receiving influenza and pneumococcal vaccinations—United Chin, M. H., et al. 999. Appropriateness of medication selection and trends. J Am Geriatr Soc 5(2): 78-28. States, 996. MMWR Morb Mortal Wkly Rep 48(39): 886-90. for older persons in an urban academic emergency department. Acad Emerg Med 6(2): 232-42. Davis, K., et al. 2002. Medicare versus private insurance: Rhetoric and Ford, E. S., et al. 2003. Serum total cholesterol concentrations and Goldstein, E. 200. CMS’s consumer information reality. Health Aff (Millwood) Suppl Web Exclusives: W3-24. awareness, treatment, and control of hypercholesterolemia among efforts. Health Care Financ Rev 23(): -4. U.S. adults: findings from the National Health and Nutrition DHHS (Department of Health and Human Services). 2000. Older adults Goldstein, E., et al. 200. Medicare managed care CAHPS: A tool for Examination Survey, 999 to 2000. Circulation 07(7): 285-9. and mental health. Mental health: A report of the Surgeon General. performance improvement. Health Care Financ Rev 22(3): 0-7. Washington, D.C.: U.S. Department of Health and Human Services. Fronstin. 2005. The impact of the erosion of retiree health benefits Gonzales, R., et al. 999. Decreasing antibiotic use in ambulatory practice: on workers and retirees. EBRI Issue Brief (279): -9. ——— 2002a. Healthy People 200. Washington, D.C.: U.S. Impact of a multidimensional intervention on the treatment of Washington, D.C.: Employee Benefit Research Institute. Department of Health and Human Services. uncomplicated acute bronchitis in adults. JAMA 28(6): 52-9. Fu, A. Z., et al. 2004. Inappropriate medication use and health ——— 2002b. Tracking Healthy People 200. Washington, D.C.: ——— 200. Principles of appropriate antibiotic use for treatment outcomes in the elderly. J Am Geriatr Soc 52(): 934-9. U.S. Department of Health and Human Services. of acute respiratory tract infections in adults: Background, Fuster, V., et al. 200. ACC/AHA/ESC guidelines for the management specific aims, and methods. Ann Intern Med 34(6): 479-86. ——— 2004. Bone health and osteoporosis: A report of the Surgeon of patients with atrial fibrillation: Executive summary. A General. Rockville, Md.: U.S. Department of Health and ——— 2004. Antibiotic treatment of acute respiratory tract report of the American College of Cardiology/American Human Services, Office of the Surgeon General. infections in the elderly: Effect of a multidimensional Heart Association Task Force on Practice Guidelines and educational intervention. J Am Geriatr Soc 52(): 39-45. Diokno, A. and M. Yuhico, Jr. 995. Preference, compliance and the European Society of Cardiology Committee for Practice initial outcome of therapeutic options chosen by female Guidelines and Policy Conferences. Circulation 04(7): 28-50. Gornick, M. E. 2000. Vulnerable populations and Medicare services: Why patients with urinary incontinence. J Urol 54(5): 727-30. do disparities exist? New York, N.Y.: The Century Foundation Press. Gaba, D. M. 2000. Anaesthesiology as a model for patient Dowell, S. F. and B. Schwartz. 997. Resistant pneumococci: safety in health care. BMJ 320(7237): 785-8. Groeneveld, P. W., et al. 2005. Trends in implantable cardioverter- Protecting patients through judicious use of antibiotics. defibrillator racial disparity: The importance of GAO (Government Accountability Office). 2002a. Medicare Am Fam Physician 55(5): 647-54, 657-8. geography. J Am Coll Cardiol 45(): 72-8. beneficiary use of clinical preventive services. GAO-02-422. Eichner, J. and B. C. Vladeck. 2005. Medicare as a catalyst for reducing Washington, D.C.: Government Accountability Office. Gross, D. A., et al. 998. Patient satisfaction with time spent health disparities. Health Aff (Millwood) 24(2): 365-75. with their physician. J Fam Pract 47(2): 33-7. ——— 2002b. Medicare home health agencies: Weaknesses in federal Englander, F., et al. 996. Economic dimensions of slip and fall injuries. and state oversight mask potential quality issues. GAO-05-77T. Gross, D. L., et al. 2004. The growing pains of integrated J Forensic Sci 4(5): 733-46. Washington, D.C.: Government Accountability Office. health care for the elderly: Lessons from the expansion of PACE. Milbank Q 82(2): 257-82. Epstein, A. J. 200. The role of public clinics in ——— 2004a. Accuracy of responses from the -800-MEDICARE preventable hospitalizations among vulnerable help line should be improved. GAO-05-30. Washington, Grundy, S. M., et al. 2004. Implications of recent clinical trials populations. Health Serv Res 36(2): 405-20. D.C.: Government Accountability Office. for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation 0(2): 227-39. Fang, M. C., et al. 2004. National trends in antiarrhythmic ——— 2004b. Flu vaccine: Recent supply shortages underscore and antithrombotic medication use in atrial ongoing challenges. GAO-05-77T. Washington, Gurwitz, J. H., et al. 2003. Incidence and preventability of fibrillation. Arch Intern Med 64(): 55-60. D.C.: Government Accountability Office. adverse drug events among older persons in the ambulatory setting. JAMA 289(9): 07-6. Fedson, D. S., et al. 990. Hospital-based pneumococcal Gaynes, R. P., et al. 200. Surgical site infection (SSI) rates in the United immunization. Epidemiologic rationale from the States, 992-998: The National Nosocomial Infections Surveillance Hajjar, I. and T. A. Kotchen. 2003. Trends in prevalence, Shenandoah study. JAMA 264(9): 7-22. System basic SSI risk index. Clin Infect Dis 33 Suppl 2: S69-77. awareness, treatment, and control of hypertension in the United States, 988-2000. JAMA 290(2): 99-206. Fick, D. M., et al. 200. Potentially inappropriate medication use in a Geerts, W. H., et al. 2004. Prevention of venous thromboembolism: Medicare managed care population: Association with higher The Seventh ACCP Conference on Antithrombotic and Harper, S. A., et al. 2004. Prevention and control of influenza: costs and utilization. J Managed Care Pharm 7(5): 407-3. Thrombolytic Therapy. Chest 26(3 Suppl): 338S-400S. Recommendations of the Advisory Committee on Immunization Practices. MMWR Recomm Rep 53(RR-6): -40. ——— 2003. Updating the Beers criteria for potentially inappropriate Gleason, P. P., et al. 999. Associations between initial antimicrobial medication use in older adults: Results of a US consensus therapy and medical outcomes for hospitalized elderly HCFO (Changes in Health Care Financing and Organization). panel of experts. Arch Intern Med 63(22): 276-24. patients with pneumonia. Arch Intern Med 59(2): 2562-72. 2004. Medicare Modernization Act offers new end-of-life care provisions. Washington, D.C.: AcademyHealth. FIFARS (Federal Interagency Forum on Aging Related Statistics). Gnanadesigan, N., et al. 2004. The quality of care provided to 2004. Older Americans 2004: Key indicators of well-being. vulnerable older community-based patients with urinary Higashi, T., et al. 2004. The quality of pharmacologic care for Washington, D.C.: U.S. Government Printing Office. incontinence. J Am Med Dir Assoc 5(3): 4-6. vulnerable older patients. Ann Intern Med 40(9): 74-20. Fiscella, K., et al. 2000. Inequality in quality: Addressing socioeconomic, Go, A. S., et al. 200. Prevalence of diagnosed atrial fibrillation in Houck, P. M., et al. 2004. Timing of antibiotic administration and racial, and ethnic disparities in health care. JAMA 283(9): 2579-84. adults: National implications for rhythm management and outcomes for Medicare patients hospitalized with community- stroke prevention. The AnTicoagulation and Risk Factors in acquired pneumonia. Arch Intern Med 64(6): 637-44. ——— 2002. Disparities in health care by race, ethnicity, and language among the insured: Findings from Atrial Fibrillation (ATRIA) Study. JAMA 285(8): 2370-5. a national sample. Med Care 40(): 52-9. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 177 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 178 Hudson, R. B. 2003. Emerging crisis: The geriatric care workforce. Kane, R. L., et al. 2004. Essentials of clinical geriatrics. Landon, B. E., et al. 2004. Comparison of performance of traditional Public Policy & Aging Report 3(2): -26. Washington, New York, N.Y.: McGraw-Hill Companies. Medicare vs. Medicare managed care. JAMA 29(4): 744-52. D.C.: National Academy on an Aging Society. Kanjanarat, P., et al. 2003. Nature of preventable adverse Leatherman, S. and K. Sutherland. 2003. The quest for quality Humphrey, L. L., et al. 2002. Breast cancer screening: A summary drug events in hospitals: A literature review. in the NHS. London: The Nuffield Trust. of the evidence for the U .S. Preventive Services Task Am J Health Syst Pharm 60(7): 750-9. Lebowitz, B. D., et al. 997. Diagnosis and treatment of depression in Force. Ann Intern Med 37(5 Part ): 347-60. Katon, W., et al. 200. A randomized trial of relapse prevention of late life. Consensus statement update. JAMA 278(4): 86-90. Hunt, D., et al. 2004. Fundamentals of Medicare patient safety depression in primary care. Arch Gen Psychiatry 58(3): 24-7. Lewin, S. A., et al. 200. Interventions for providers to promote surveillance: Intent, relevance, and transparency. Baltimore, Katon, W. J., et al. 2003. Increased medical costs of a a patient-centred approach in clinical consultations. Md.: Centers for Medicare and Medicaid Services. population-based sample of depressed elderly Cochrane Database Syst Rev (4): CD003267. IOM (Institute of Medicine). 990. Medicare: A strategy patients. Arch Gen Psychiatry 60(9): 897-903. Lewis, R. 995. The rise of antibiotic-resistant infections. FDA Consumer for quality assurance. Edited by K. T. Lohr. Katz, S. J., et al. 998. Medication management of depression in the Magazine. Washington, D.C.: U.S. Food and Drug Administration. Washington, D.C.: National Academy Press. United States and Ontario. J Gen Intern Med 3(2): 77-85. Lorenz, K. A., et al. 2004a. Accommodating ethnic diversity: A study ——— 993. Access to health care in America. Edited by M. Kaufman, D. W., et al. 2002. Recent patterns of medication of California hospice programs. Med Care 42(9): 87-4. Millman. Washington, D.C.: National Academy Press. use in the ambulatory adult population of the United ——— 2004b. Hospice admission practices: Where does hospice fit ——— 997. Approaching death: Improving care at the States: The Slone survey. JAMA 287(3): 337-44. in the continuum of care? J Am Geriatr Soc 52(5): 725-30. end of life. Edited by M. J. Field and C. K. Cassel. Keehan, S. P., et al. 2004. Age estimates in the national health Washington, D.C.: National Academy Press. Lyke, J. and M. Colon. 2004. Practical recommendations accounts. Health Care Financ Rev (Web Exclusive) (): -6. for ethnically and racially sensitive hospice ——— 999. To err is human: Building a safer health care system. Edited KFF/AHRQ (Kaiser Family Foundation and Agency for services. Am J Hosp Palliat Care 2(2): 3-3. by L. T. Kohn, et al. Washington, D.C.: National Academy Press. Healthcare Research and Quality). 996. Americans as Majumdar, S. R., et al. 2004. A controlled trial to increase ——— 200a. Crossing the quality chasm: A new health system for the health care consumers: The role of quality information. detection and treatment of osteoporosis in older patients 2st century. Washington, D.C.: National Academy Press. Washington, D.C.: Kaiser Family Foundation. with a wrist fracture. Ann Intern Med 4(5): 366-73. ——— 200b. Envisioning the national health care quality report. Edited by Khuri, S. F., et al. 2002. The comparative assessment and Mandell, L. A., et al. 2003. Update of practice guidelines for M. P. Hurtado, et al. Washington, D.C.: National Academy Press. improvement of quality of surgical care in the Department the management of community-acquired pneumonia in of Veterans Affairs. Arch Surg 37(): 20-7. ——— 2003. Unequal treatment: Confronting racial and ethnic immunocompetent adults. Clin Infect Dis 37(): 405-33. disparities in health care. Edited by B. D. Smedley, et Kilker, K., et al. 2000. Hypertension: A common Masoudi, F. A., et al. 2004. National patterns of use and al. Washington, D.C.: National Academy Press. condition for older Americans. Washington, D.C.: effectiveness of angiotensin-converting enzyme inhibitors National Academy on an Aging Society. ——— 2004. The st annual crossing the quality chasm summit: in older patients with heart failure and left ventricular A focus on communities. Edited by K. Adams, et al. Kirkland, K. B., et al. 999. The impact of surgical-site infections in the systolic dysfunction. Circulation 0(6): 724-3. Washington, D.C.: National Academy Press. 990s: Attributable mortality, excess length of hospitalization, McGlynn, E. A. and R. H. Brook. 200. Evaluating the quality care and extra costs. Infect Control Hosp Epidemiol 20(): 725-30. Jack, L., Jr., et al. 2004. The psychosocial aspects of diabetes care. Changing the U.S. Health Care System. Edited by R. M. care. Using collaborative care to manage older adults Kitchman, M., et al. 2002. Seniors and prescription drugs: Findings Andersen, et al. San Francisco, Calif.: Jossey-Bass: 6-7. with diabetes. Geriatrics 59(5): 26-3; quiz 32. from a 200 survey of seniors in eight states. Washington, McWilliams, J. M., et al. 2003. Impact of Medicare D.C.: Kaiser Family Foundation, Commonwealth Jemal, A., et al. 2004. Cancer statistics, 2004. CA Cancer J Clin 54(): 8-29. coverage on basic clinical services for previously Fund, Tufts-New England Medical Center. uninsured adults. JAMA 290(6): 757-64. Jencks, S. F. and G. R. Wilensky. 992. The health care quality Klap, R., et al. 2003. Caring for mental illness in the United States: A improvement initiative. A new approach to quality MedPAC (Medicare Payment Advisory Commission). 2002a. focus on older adults. Am J Geriatr Psychiatry (5): 57-24. assurance in Medicare. JAMA 268(7): 900-3. Coverage beyond the basic benefit package. Report to the Kruspe, R., et al. 2003. Education does pay off: Pneumococcal Congress: Assessing Medicare benefits: 27-40. Washington, Jencks, S. F., et al. 2000. Quality of medical care delivered vaccine screening and administration in hospitalized adult D.C.: Medicare Payment Advisory Commission. to Medicare beneficiaries: A profile at state and patients with pneumonia. J La State Med Soc 55(6): 325-3. national levels. JAMA 284(3): 670-6. ——— 2002b. Report to the Congress: Applying quality Kruzikas, D. T., et al. 2004. Preventable hospitalizations: A window into improvement standards in Medicare. Washington, ——— 2003. Change in the quality of care delivered to Medicare primary and preventive care, 2000. AHRQ Pub. No. 04-0056. D.C.: Medicare Payment Advisory Commission. beneficiaries, 998-999 to 2000-200. JAMA 289(3): 305-2. Rockville, Md.: Agency for Healthcare Research and Quality. ——— 2002c. Report to the Congress: Medicare beneficiaries’ Kamel, H. K., et al. 2000. Failure to diagnose and treat LaMascus, A. M., et al. 2005. Bridging the workforce gap for our aging access to hospice. Washington, D.C.: Medicare osteoporosis in elderly patients hospitalized with society: How to increase and improve knowledge and training. Payment Advisory Commission. hip fracture. Am J Med 09(4): 326-8. Report of an expert panel. J Am Geriatr Soc 53(2): 343-7. ——— 2004a. Home health services. Report to the Congress: ——— 2004a. Health, United States, 2004. Hyattsville, Md.: NOF (National Osteoporosis Foundation). 999, 2003. Physician’s Medicare payment policy. Washington, D.C.: Centers for Disease Control and Prevention. guide to prevention and treatment of osteoporosis. Medicare Payment Advisory Commission. Washington, D.C.: National Osteoporosis Foundation. ——— 2004b. Summary health statistics for the U.S. population: ——— 2004b. Hospice care in Medicare: Recent trends and a review of National Health Interview Survey, 2002. Vital Health Stat 0(220). ——— 2004. Osteoporosis disease statistics. Washington, the issues. Report to the Congress: New approaches in Medicare. Hyattsville, Md.: Centers for Disease Control and Prevention. D.C.: National Osteoporosis Foundation. Washington, D.C.: Medicare Payment Advisory Commission. ——— 2004c. Summary health statistics for U.S. adults: National NPA (National PACE Association). 2002. What can PACE do for ——— 2004c. Quality of care for Medicare beneficiaries. Report Health Interview Survey, 2002. Vital Health Stat 0(222). your state? Alexandria, Va.: National PACE Association. to the Congress: Medicare payment policy. Washington, Hyattsville, Md.: Centers for Disease Control and Prevention. NQMC (National Quality Measures Clearinghouse). 2003a. Measure D.C.: Medicare Payment Advisory Commission. ——— 2005. Incontinence among community dwelling Medicare summary: Antidepressant medication management: Effective acute Meehan, T. P., et al. 997. Quality of care, process, and outcomes in beneficiaries, by age, sex, and race: United States, 992-999. treatment phase (National Committee for Quality Assurance). elderly patients with pneumonia. JAMA 278(23): 2080-4. Hyattsville, Md.: U.S. Department of Health and Human Services. Rockville, Md.: Agency for Healthcare Research and Quality. Mehta, R. H., et al. 2002. Improving quality of care for acute NCIPC (National Center for Injury Prevention and Control). ——— 2003b. Measure summary: Antidepressant medication myocardial infarction: The Guidelines Applied in 2005. Falls and hip fractures among older adults: Fact sheet. management: Effective continuation phase treatment Practice (GAP) Initiative. JAMA 287(0): 269-76. Atlanta, Ga.: Centers for Disease Control and Prevention. (National Committee for Quality Assurance). Rockville, Md.: Agency for Healthcare Research and Quality. Menendez-Jandula, B., et al. 2005. Comparing self-management NCQA (National Committee for Quality Assurance). 2002. of oral anticoagulant therapy with clinic management: HEDIS: What’s in it and why it matters. Washington, ——— 2003c. Measure summary: Antidepressant medication A randomized trial. Ann Intern Med 42(): -0. D.C.: National Committee for Quality Assurance. management: Optimal practitioner contacts (National Committee for Quality Assurance). Rockville, Md.: Mittelman, M. S., et al. 996. A family intervention to delay nursing ——— 2004. The state of managed care quality, 2004. Washington, Agency for Healthcare Research and Quality. home placement of patients with Alzheimer’s disease. A D.C.: National Committee for Quality Assurance. randomized controlled trial. JAMA 276(2): 725-3. ——— 2003d. Measure summary: Beta-blocker treatment after heart ——— 2005. The Health Plan Employer Data and Information Set. attack (National Committee for Quality Assurance). Rockville, ——— 2004. Sustained benefit of supportive intervention for Washington, D.C.: National Committee for Quality Assurance. Md.: Agency for Healthcare Research and Quality. depressive symptoms in caregivers of patients with Nelson, D. E., et al. 2003. A comparison of national estimates from Alzheimer’s disease. Am J Psychiatry 6(5): 850-6. ——— 2003e. Measure summary: Cholesterol management after acute the National Health Interview Survey and the Behavioral Risk cardiovascular events (National Committee for Quality Assurance). Nadel, M. R., et al. 2002. Are people being screened for colorectal Factor Surveillance System. Am J Public Health 93(8): 335-4. Rockville, Md.: Agency for Healthcare Research and Quality. cancer as recommended? Results from the National Neveleff, D. J. 2003. Strategies aim to reduce harm to Health Interview Survey. Prev Med 35(3): 99-206. ——— 2003f. Measure summary: Hypertension (National patients. QualityIndicator.com. Morristown, Committee for Quality Assurance). Rockville, Md.: Naylor, M. D. 2002. Transitional care of older adults. N.J.: Premier Healthcare Resource, Inc. Agency for Healthcare Research and Quality. Annu Rev Nurs Res 20: 27-47. Nickens, H. 986. Report of the Secretary’s Task Force on Black ——— 2003g. Measure summary: Mental illness: Percentage of Naylor, M. D., et al. 2004. Transitional care of older adults and Minority Health. J Natl Med Assoc 78(6): 577-80. members who had an ambulatory or day/night mental hospitalized with heart failure: A randomized, Niederman, M. S., et al. 200. Guidelines for the management of health visit within 7 days of hospital discharge (National controlled trial. J Am Geriatr Soc 52(5): 675-84. adults with community-acquired pneumonia. Diagnosis, Committee for Quality Assurance). Rockville, Md.: NCEP (National Cholesterol Education Program). 200. Executive assessment of severity, antimicrobial therapy, and Agency for Healthcare Research and Quality. Summary of the Third Report of the National Cholesterol prevention. Am J Respir Crit Care Med 63(7): 730-54. ——— 2003h. Measure summary: Mental illness: Percentage of Education Program (NCEP) Expert Panel on Detection, Niefeld, M. R., et al. 2003. Preventable hospitalization members who had an ambulatory or day/night mental Evaluation, and Treatment of High Blood Cholesterol in among elderly Medicare beneficiaries with type health visit within 30 days of hospital discharge (National Adults (Adult Treatment Panel III). JAMA 285(9): 2486-97. 2 diabetes. Diabetes Care 26(5): 344-9. Committee for Quality Assurance). Rockville, Md.: NCHS (National Center for Health Statistics). 997a. Access to health Agency for Healthcare Research and Quality. NIH (National Institutes of Health). 99. Diagnosis and treatment care. Part 2: Working-age adults. Vital Health Stat 0(97). of depression in late life. NIH Consensus Development Parchman, M. L. and S. D. Culler. 999. Preventable hospitalizations Hyattsville, Md.: Centers for Disease Control and Prevention. Program 9(3). Bethesda, Md.: National Institutes of Health. in primary care shortage areas. An analysis of vulnerable ——— 997b. Access to health care. Part 3: Older adults. Medicare beneficiaries. Arch Fam Med 8(6): 487-9. ——— 2004. Osteoporosis overview. Washington, D.C.: Vital Health Stat 0(98). Hyattsville, Md.: Centers National Institute of Health, Osteoporosis and Related Pearson, T. A., et al. 2002. AHA Guidelines for Primary Prevention of for Disease Control and Prevention. Bone Diseases, National Resource Center. Cardiovascular Disease and Stroke: 2002 Update: Consensus ——— 2002. Summary health statistics for U.S. adults: National Panel Guide to Comprehensive Risk Reduction for Adult NIMH (National Institute of Mental Health). 2003. Older Health Interview Survey, 997. Vital Health Stat 0(205). Patients Without Coronary or Other Atherosclerotic Vascular adults: Depression and suicide facts. Bethesda, Hyattsville, Md.: Centers for Disease Control and Prevention. Diseases. American Heart Association Science Advisory Md.: National Institutes of Health. and Coordinating Committee. Circulation 06(3): 388-9. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 179 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 180 Phillips, C. O., et al. 2004. Comprehensive discharge planning with RTI (Research Triangle Institute). 200. Implementation of Simon, G. E., et al. 200. Treatment process and outcomes for postdischarge support for older patients with congestive Medicare CAHPS fee for service survey. Rockville, Md.: managed care patients receiving new antidepressant heart failure: A meta-analysis. JAMA 29(): 358-67. Centers for Medicare and Medicaid Services. prescriptions from psychiatrists and primary care physicians. Arch Gen Psychiatry 58(4): 395-40. Phillips, R. S., et al. 996. Choices of seriously ill patients Rubenstein, L. Z., et al. 2004. Detection and management of about cardiopulmonary resuscitation: Correlates falls and instability in vulnerable elders by community Singer, D. E., et al. 2004. Antithrombotic therapy in atrial fibrillation: and outcomes. SUPPORT Investigators. Study to physicians. J Am Geriatr Soc 52(9): 527-3. The Seventh ACCP Conference on Antithrombotic and Understand Prognoses and Preferences for Outcomes Thrombolytic Therapy. Chest 26(3 Suppl): 429S-456S. Safran, D. G., et al. 2002. Prescription drug coverage and and Risks of Treatments. Am J Med 00(2): 28-37. seniors: How well are states closing the gap? Health Sirovich, B. E., et al. 2003. Screening men for prostate and Pignone, M., et al. 2002. Screening for colorectal cancer in adults at Aff (Millwood) Suppl Web Exclusives: W253-68. colorectal cancer in the United States: Does practice average risk: A summary of the evidence for the U.S. Preventive reflect the evidence? JAMA 289(): 44-20. Samsa, G. P., et al. 2000. Quality of anticoagulation management among Services Task Force. Ann Intern Med 37(2): 32-4. patients with atrial fibrillation: Results of a review of medical Skinner, J., et al. 2003. Racial, ethnic, and geographic Post, D. M., et al. 2002. The other half of the whole: Teaching patients records from 2 communities. Arch Intern Med 60(7): 967-73. disparities in rates of knee arthroplasty among to communicate with physicians. Fam Med 34(5): 344-52. Medicare patients. N Engl J Med 349(4): 350-9. Sangl, J., et al. 2005. Challenges in measuring nursing home and Promoting Excellence. 200. Promoting excellence in home health quality: Lessons from the First National Sloss, E. M., et al. 2000. Selecting target conditions for end of life care: Interim report. Missoula, Mont.: Healthcare Quality Report. Med Care 43(3 Suppl): I24-32. quality of care improvement in vulnerable older University of Montana Practical Ethics Center. adults. J Am Geriatr Soc 48(4): 363-9. Schoeni, P. Q. 2002. Accelerating change today for America’s Pronovost, P. J. and S. M. Berenholtz. 2004. Improving sepsis care health. Washington, D.C.: National Coalition on Health Solomon, D. H., et al. 2003. Underuse of osteoporosis medications in in the intensive care unit: An evidence-based approach. Care and the Institute for Healthcare Improvement. elderly patients with fractures. Am J Med 5(5): 398-400. VHA Research Series. Irving, Tex.: VHA, Inc. SCIP (Surgical Care Improvement Project). 2004. Surgical care Sood, N., et al. 2000. Determinants of antidepressant treatment Reason, J. 997. Managing the risks of organizational accidents. improvement project: A national quality partnership. Rockville, outcome. Am J Manag Care 6(2): 327-36. Burlington, Vt.: Ashgate Publishing Co. Md.: Centers for Medicare and Medicaid Services. Sprague, L. 2002. Contracting for quality: Medicare’s quality Reuben, D. B., et al. 2003a. Quality of care for older persons at the dawn Seeff, L. C., et al. 2004. Patterns and predictors of colorectal cancer test improvement organizations. NHPF Issue Brief (774): -5. of the third millennium. J Am Geriatr Soc 5(7 Suppl): S346-50. use in the adult U.S. population. Cancer 00(0): 2093-03. Starfield, B. 998. Primary care: Balancing health needs, services, ——— 2003b. Restructuring primary care practices to Seppala, H., et al. 997. The effect of changes in the consumption and technology. New York, N.Y.: Oxford University Press. manage geriatric syndromes: The ACOVE-2 of macrolide antibiotics on erythromycin resistance in Stewart, M., et al. 2000. The influence of older patient- intervention. J Am Geriatr Soc 5(2): 787-93. group A streptococci in Finland. Finnish Study Group for physician communication on health and health-related Antimicrobial Resistance. N Engl J Med 337(7): 44-6. Richard, A. A., et al. 2000. Using OASIS for outcome-based outcomes. Clin Geriatr Med 6(): 25-36, vii-viii. quality improvement. Home Healthc Nurse 8(4): 232-7. Shaughnessy, P. W., et al. 2002a. Improving patient outcomes of home Stewart, M. A. 995. Effective physician-patient communication health care: Findings from two demonstration trials of outcome- Ries, L. A., et al. 2000. The annual report to the nation on and health outcomes: A review. CMAJ 52(9): 423-33. based quality improvement. J Am Geriatr Soc 50(8): 354-64. the status of cancer, 973-997, with a special section Stone, E. G., et al. 2002. Interventions that increase use of on colorectal cancer. Cancer 88(0): 2398-424. ——— 2002b. OASIS and outcome-based quality improvement in adult immunization and cancer screening services: A home health care: Research and demonstration findings, Rizzo, J. A., et al. 996. The cost-effectiveness of a multifactorial meta-analysis. Ann Intern Med 36(9): 64-5. policy implications, and considerations for future change. targeted prevention program for falls among community Denver, Colo.: University of Colorado Health Science SUN (CAHPS Survey Users Network). 2004. An Introduction to CAHPS. elderly persons. Med Care 34(9): 954-69. Center, Center for Health Services Research. Rockville, Md.: Agency for Healthcare Research and Quality. Robinson, K. A., et al. 200. Epidemiology of invasive Shi, L. and G. D. Stevens. 2004. Vulnerability and unmet health Tan, E. J., et al. 2003. Differences in mortality of black and Streptococcus pneumoniae infections in the United care needs. J Gen Intern Med 20: forthcoming. white patients enrolled in the program of all-inclusive States, 995-998: Opportunities for prevention in the care for the elderly. J Am Geriatr Soc 5(2): 246-5. conjugate vaccine era. JAMA 285(3): 729-35. Shojania, K. G., et al., Eds. 2004. Diabetes mellitus care. Closing the Quality Gap: A Critical Analysis of Quality Improvement Tang, S. T. 2003. When death is imminent: Where terminally ill patients Rothschild, J. M., et al. 2000. Preventable medical injuries in Strategies, vol. 2. AHRQ Pub. No. 04-005-2. Rockville, with cancer prefer to die and why. Cancer Nurs 26(3): 245-5. older patients. Arch Intern Med 60(8): 277-28. Md.: Agency for Healthcare Research and Quality. Teno, J. M., et al. 200. Patient-focused, family-centered end-of- Rozich, J. D., et al. 2003. Adverse drug event trigger tool: Siebenhofer, A., et al. 2004. Systematic review of studies life medical care: Views of the guidelines and bereaved A practical methodology for measuring medication of self-management of oral anticoagulation. family members. J Pain Symptom Manage 22(3): 738-5. related harm. Qual Saf Health Care 2(3): 94-200. Thromb Haemost 9(2): 225-32. ——— 2004. Family perspectives on end-of-life care at the last place of care. JAMA 29(): 88-93. Teunissen, T. A., et al. 2004. Treating urinary incontinence ——— 2002c. Screening for depression: Recommendations Williams, R. D., 2nd. 2004. The unique needs of Medicare in the elderly—conservative therapies that work: A and rationale. Ann Intern Med 36(0): 760-4. beneficiaries. Medicare Brief (0). Washington, systematic review. J Fam Pract 53(): 25-30, 32. D.C.: National Academy of Social Insurance. ——— 2002d. Screening for osteoporosis in postmenopausal women: Thomas, E. J. and T. A. Brennan. 2000. Incidence and types of Recommendation and rationale. Ann Intern Med 37(6): 526-8. Wilson, S. J., et al. 2003. Comparing the quality of oral anticoagulant preventable adverse events in elderly patients: Population management by anticoagulation clinics and by family physicians: ——— 2002e. Screening for prostate cancer: Recommendation based review of medical records. BMJ 320(7237): 74-4. A randomized controlled trial. CMAJ 69(4): 293-8. and rationale. Ann Intern Med 37(): 95-6. Thompson, W. W., et al. 2003. Mortality associated with Wilt, V. M., et al. 995. Outcome analysis of a pharmacist-managed Virnig, B., et al. 2004. Does Medicare managed care influenza and respiratory syncytial virus in the anticoagulation service. Pharmacotherapy 5(6): 732-9. provide equal treatment for mental illness across United States. JAMA 289(2): 79-86. races? Arch Gen Psychiatry 6(2): 20-5. Winawer, S., et al. 2003. Colorectal cancer screening and Tinetti, M. E. and C. S. Williams. 997. Falls, injuries due to falls, and the surveillance: Clinical guidelines and rationale—update Virnig, B. A., et al. 2002. Racial variation in quality of care among risk of admission to a nursing home. N Engl J Med 337(8): 279-84. based on new evidence. Gastroenterology 24(2): 544-60. Medicare+Choice enrollees. Health Aff (Millwood) 2(6): 224-30. ——— 998. The effect of falls and fall injuries on Wittkowsky, A. K., et al. 2005. Barriers to patient self-testing of Walsh, J., et al., Eds. 2005. Hypertension care. Closing the functioning in community-dwelling older persons. prothrombin time: National survey of anticoagulation Quality Gap: A Critical Analysis of Quality Improvement J Gerontol A Biol Sci Med Sci 53(2): M2-9. practitioners. Pharmacotherapy 25(2): 265-9. Strategies, vol. 3. AHRQ Pub. No. 04-005-3. Rockville, Tinetti, M. E., et al. 2004. Potential pitfalls of disease-specific guidelines Md.: Agency for Healthcare Research and Quality. Wolff, J. L., et al. 2002. Prevalence, expenditures, and for patients with multiple conditions. N Engl J Med 35(27): 2870-4. complications of multiple chronic conditions in the Walter, L. C. and K. E. Covinsky. 200. Cancer screening elderly. Arch Intern Med 62(20): 2269-76. Trude, S. and P. B. Ginsburg. 2005. An update on Medicare in elderly patients: A framework for individualized beneficiary access to physician services. Issue Brief decision making. JAMA 285(2): 2750-6. Zhan, C., et al. 200. Potentially inappropriate medication use in Cent Stud Health Syst Change (93): -4. the community-dwelling elderly: Findings from the 996 Walter, L. C., et al. 2004. Relationship between health status and use Medical Expenditure Panel Survey. JAMA 286(22): 2823-9. Tu, G. S., et al. 2004. Which strategies facilitate improvement of screening mammography and Papanicolaou smears among in quality of care for elderly hospitalized pneumonia women older than 70 years of age. Ann Intern Med 40(9): 68-8. Zhan, C. and M. R. Miller. 2003. Excess length of stay, patients? Jt Comm J Qual Saf 30(): 25-35. charges, and mortality attributable to medical injuries Wegman, A., et al. 2004. Nonsteroidal antiinflammatory drugs or during hospitalization. JAMA 290(4): 868-74. Tutty, S., et al. 2000. Telephone counseling as an adjunct acetaminophen for osteoarthritis of the hip or knee? A systematic to antidepressant treatment in the primary care review of evidence and guidelines. J Rheumatol 3(2): 344-54. Zhan, C., et al. 2005. Suboptimal prescribing in elderly system. A pilot study. Eff Clin Pract 3(4): 70-8. outpatients: Potentially harmful drug-drug and drug- Weingart, S. N., et al. 2005. Patient-reported medication symptoms disease combinations. J Am Geriatr Soc 53(2): 262-7. U.S. Census Bureau. 2005. Health insurance coverage status and in primary care. Arch Intern Med 65(2): 234-40. type of coverage by selected characteristics: 2003. Table Wenger, N. S., et al. 2003. The quality of medical care HI0. Washington, D.C.: U.S. Census Bureau, Housing provided to vulnerable community-dwelling older and Household Economic Statistics Division. patients. Ann Intern Med 39(9): 740-7. Unutzer, J., et al. 997. Depressive symptoms and the cost of ——— 2005. A controlled trial of a practice-based intervention health services in HMO patients aged 65 years and older. to improve primary care for falls, incontinence, and A 4-year prospective study. JAMA 277(20): 68-23. dementia. SGIM 28th Annual Meeting, New Orleans, ——— 200. Improving primary care for depression in late life: The La.: Society for General Internal Medicine. design of a multicenter randomized trial. Med Care 39(8): 785-99. Wennberg, J. E., et al. 2004a. Geography and the debate over Medicare ——— 2002. Collaborative care management of late-life reform. Health Aff (Millwood) Suppl Web Exclusive: W96-4. depression in the primary care setting: A randomized ——— 2004b. Use of Medicare claims data to monitor provider- controlled trial. JAMA 288(22): 2836-45. specific performance among patients with severe chronic ——— 2003. Depression treatment in a sample of ,80 depressed illness. Health Aff (Millwood) Suppl Web Exclusive: VAR5-8. older adults in primary care. J Am Geriatr Soc 5(4): 505-4. White, A. 998. Evaluation of the program of all-inclusive care for the USPSTF (U.S. Preventive Services Task Force). 2002a. elderly (PACE) demonstration. The effect of PACE on costs to Screening for breast cancer: Recommendations and Medicare: A comparison of Medicare capitation rates to projected rationale. Ann Intern Med 37(5 Part ): 344-6. costs in the absence of PACE. Cambridge, Mass.: Abt Associates, Inc. ——— 2002b. Screening for colorectal cancer: Recommendation Whitney, C. G., et al. 2000. Increasing prevalence of and rationale. Ann Intern Med 37(2): 29-3. multidrug-resistant Streptococcus pneumoniae in the United States. N Engl J Med 343(26): 97-24. Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 181 Leatherman and McCarthy, Quality of Health Care for Medicare Beneficiaries: A Chartbook, 2005. The Commonwealth Fund 182 About the Authors Sheila Leatherman is a research professor at the School of Public Douglas McCarthy is president of Issues Research, Inc., in Durango, Health, The University of North Carolina at Chapel Hill, and a Colorado. He has 20 years of experience in public and private sector Senior Associate of The Judge Institute of Management (996) and research, policymaking, and management. He has authored or Distinguished Associate of Darwin College at the University of coauthored reports and articles on a range of topics including health Cambridge, England. She is an elected member of the Institute of care coverage and quality, information privacy, technology assess- Medicine of the U.S. National Academy of Sciences (2002) and of ment, corporate philanthropy, and public performance reporting. the National Academy of Social Insurance (997). He was previously a research director at a health services Professor Leatherman conducts research and policy analysis research center affiliated with a national health care company, where in the United States and the United Kingdom, focusing on quality he studied health system performance and implemented quality- of care, health systems reform, performance measurement and evaluation tools in health plans nationally. He began his career as improvement, and the economic implications of implementing an internal consultant for a local government, where he supported quality-enhancing interventions in health care delivery. She was quality improvements through operations research and information appointed by President Clinton in 997 to the President’s Advisory systems development. Commission on Consumer Protection and Quality in the Health He received his bachelor’s degree with honors from Yale Care Industry, chairing the sub-committee to develop a national College and a master’s degree in health care management from strategy for quality measurement and reporting. She is coauthor the University of Connecticut. During 996–997, he was a of a series of chartbooks on quality of health care in the United public policy fellow at the Humphrey Institute of Public Affairs States, commissioned by The Commonwealth Fund. In the United at the University of Minnesota. He serves on the advisory board Kingdom, she was commissioned by The Nuffield Trust to assess for a local community health center and on a citizens’ advisory the British Government’s proposed quality reforms for the National committee that is working to improve the accessibility of health Health Service in 997–98 and evaluated the mid-term impact of care in his rural community. the 0-year quality agenda in the NHS, resulting in publication of the book Quest for Quality in the NHS (2003) and a forthcoming sequel, Quest for Quality in the NHS: A Chartbook on Quality in the UK (June 2005). She has a broad background in health care management in state and federal health agencies, as chief executive of an HMO, and as senior executive of a large national managed care company in the United States. She is a senior advisor to The Nuffield Trust and to The Health Foundation in the United Kingdom, a trustee of the American Board of Medicine Foundation, and she serves on the board of directors of the international organization Freedom From Hunger. The Commonwealth Fund The Commonwealth Fund is a private foundation established in 1918 by Anna M. Harkness with One East 75th Street the broad charge to enhance the common good. The Fund carries out this mandate by supporting New York, NY 10021-2692 efforts that help people live healthy and productive lives, and by assisting specific groups with Telephone 2 1 2 . 6 0 6 . 3 8 0 0 serious and neglected problems. The Fund supports independent research on health and social Facsimile 2 1 2 . 6 0 6 . 3 5 0 0 E-mail c mw f @ c mw f. o rg issues and makes grants to improve health care practice and policy. Web w w w. c mw f. o rg The Fund’s two national program areas are improving health insurance coverage and access to care and improving the quality of health care services. The Fund is dedicated to helping people become more informed about their health care, and improving care for vulnerable populations such as children, elderly people, low-income families, minority Americans, and the uninsured. In addition, an international program in health policy is designed to stimulate innovative policies and practices in the United States and other industrialized countries. In its own community, New York City, the Fund also makes grants to improve health care.