Aiming Higher Results from a State Scorecard on Health System Performance T h e C o m m o n w e a lt h F u n d C o m m i s s i o n o n a H i g h P e r f o r m a n c e H e a lt h S y s t e m June 2007 t h e c o m m o n w e a lt h f u n d c o m m i s s i o n o n a h i g h p e r f o r m a n c e h e a lt h s y s t e m Membership James J. Mongan, M.D. Cleve L. Killingsworth Chair of the Commission President and CEO President and CEO Blue Cross Blue Shield of Massachusetts Stephen C. Schoenbaum, M.D. Partners HealthCare System, Inc. Executive Director Sheila T. Leatherman Executive Vice President for Programs Maureen Bisognano Research Professor The Commonwealth Fund Executive Vice President & COO School of Public Health Institute for Healthcare Improvement University of North Carolina Anne K. Gauthier Judge Institute Senior Policy Director Christine K. Cassel, M.D. University of Cambridge The Commonwealth Fund President and CEO American Board of Internal Medicine Gregory P. Poulsen Cathy Schoen and ABIM Foundation Senior Vice President Research Director Intermountain Health Care Senior Vice President for Michael Chernew, Ph.D. Research and Evaluation Professor Dallas L. Salisbury The Commonwealth Fund Department of Health Policy President & CEO Harvard Medical School Employee Benefit Research Institute Allison Frey Program Associate Patricia Gabow, M.D. Sandra Shewry The Commonwealth Fund CEO and Medical Director Director Denver Health California Department of Health Services Robert Galvin, M.D. Glenn D. Steele, Jr., M.D., Ph.D. Director, Global Health President and CEO General Electric Company Geisinger Health System Fernando A. Guerra, M.D. Mary K. Wakefield, Ph.D., R.N. Director of Health Associate Dean San Antonio Metropolitan Health District School of Medicine Health Sciences Director and Professor Glenn M. Hackbarth, J.D. Center for Rural Health Chairman University of North Dakota MedPAC Alan R. Weil, J.D. George C. Halvorson Executive Director Chairman and CEO National Academy for State Health Policy Kaiser Foundation Health Plan, Inc. President Center for Health Policy Development Robert M. Hayes, J.D. President Steve Wetzell Medicare Rights Center Vice President HR Policy Association t h e c o m m o n w e a lt h f u n d The Commonwealth Fund, among the first private foundations most vulnerable, including low-income people, the uninsured, started by a woman philanthropist—Anna M. Harkness—was minority Americans, young children, and elderly adults. established in 1918 with the broad charge to enhance the The Fund carries out this mandate by supporting independent common good. research on health care issues and making grants to improve The mission of The Commonwealth Fund is to promote a health care practice and policy. An international program in health high performing health care system that achieves better access, policy is designed to stimulate innovative policies and practices improved quality, and greater efficiency, particularly for society’s in the United States and other industrialized countries. COVER PHOTOS Top left: Roger Carr Top right: Martin Dixon Bottom left: Paula Photographic Bottom right: Roger Carr Aiming Higher R e s u lt s f r o m a S tat e S c o r e c a r d o n H e a lt h S y s t e m P e r f o r m a n c e Joel C. Cantor and Dina Belloff Rutgers University Center for State Health Policy Cathy Schoen, Sabrina K. H. How, and Douglas McCarthy The Commonwealth Fund On behalf of the Commonwealth Fund Commission on a High Performance Health System June 2007 ABSTRACT: Developed to follow the National Scorecard on U.S. Health System Performance, published in 2006, the State Scorecard assesses state variation across key dimensions of health system performance: access, quality, avoidable hospital use and costs, equity, and healthy lives. The findings document wide variation among states and the potential for substantial improvement—in terms of access, quality, costs, and lives—if all states approached levels achieved by the top states. Leading states outperform lagging states on multiple indicators and dimensions; yet, all states have room to improve. The report presents state performance on 32 indicators, with overall rankings as well as ranks on each dimension. The findings underscore the need for federal and state action in key areas to move all states to higher levels of performance and value. Support for this research was provided by The Commonwealth Fund. The views presented here are those of the authors and not necessarily those of The Commonwealth Fund or its directors, officers, or staff, or of The Commonwealth Fund Commission on a High Performance Health System or its members. This report, related state tables, and other Fund publications are available online at www.commonwealthfund.org. To learn about new publications when they become available, visit the Fund Web site and register to receive e-mail alerts. Commonwealth Fund pub. no. 1030. Photo: Roger Carr Contents Preface 5 Acknowledgments 6 List of Exhibits 7 Executive Summary 8 Introduction 15 Access 18 Quality 23 Potentially Avoidable Use of Hospitals and Costs of Care 29 Equity 34 Healthy Lives 38 Cross-Cutting Findings 42 Impact of Improved Performance 44 Moving Forward: The Need for Action to Improve Performance 45 Notes 49 Appendices 51 About the Authors 73 Further Reading 74 Photo: Jeff Lee/Redux Plus Preface In 2006, the Commission published Why Not the Best? Results from a National Scorecard on T he Commonwealth Fund Commission U.S. Health System Performance to comprehen- on a High Performance Health System is sively assess how well the U.S. health system is pleased to sponsor this first State Scorecard performing across key indicators of health care on Health System Performance in the hope that it outcomes, quality, access, efficiency, and equity. will help meet the growing need for comparative Findings of the National Scorecard indicate that America’s health system falls far short of achiev- state health system performance information and able benchmarks, especially given the resources contribute to positive action among the states. the nation invests. Based on these and other data, In the U.S. federal system, the states maintain the Commission believes that transformation significant authority over many health and regula- of the U.S. health system is urgently needed to tory policies that influence health system perfor- achieve optimal health care for all Americans while mance and health outcomes. States organize and improving value for society’s investment in health deliver population health services, regulate health care. States and their health delivery systems vary insurance markets, provide Medicaid coverage for and include models and centers of excellence. In the poor and State Children’s Health Insurance many instances even top-performing states do not Program (SCHIP) coverage for low-income reach as high a level as should be achievable—and children, purchase coverage for their employees and all have substantial room to improve. Nonethe- retirees, license and monitor health care providers, less, focusing on how top-performing states and and finance charity care for the uninsured. Given organizations achieve high levels of performance these activities and levers, state policymakers across will enable the entire country to improve. The State the country are realizing the tremendous oppor- Scorecard underscores the need for national as tunity they have to shape and improve health care well as state action in key areas to move all states at the local level for their populations. to higher levels of performance and value. James J. Mongan, M.D. Stephen C. Schoenbaum, M.D. Chairman Executive Director The Commonwealth Fund Commission on a High Performance Health System Acknowledgments Commission members reviewed the State Scorecard methodology and drafts of the report. T he authors would like to thank the members We especially thank Alan Weil, J.D., M.P.P., of the Commonwealth Fund Commission executive director, National Academy for State on a High Performance Health System for Health Policy, and Mary Wakefield, Ph.D., their invaluable guidance. We also owe our sincere associate dean, University of North Dakota appreciation to all of the researchers who developed School of Medicine. Additionally, we thank indicators and conducted data analyses for the Trish Riley, M.S., director, Maine Governor’s State Scorecard. We particularly thank Katherine Office of Health Policy and Finance, and Joseph Hempstead, Ph.D., Rutgers University Center for Thompson, M.D., director, Arkansas Center for State Health Policy, and Ellen Nolte, Ph.D., at the Health Improvement and Surgeon General, State London University School of Hygiene and Tropical of Arkansas, for their review. We also thank Medicine, for their analysis of U.S. mortality data, senior staff of The Commonwealth Fund and upon which we based our estimates of mortality the Commission on a High Performance Health amenable to health care, with variation by race, for System, including Karen Davis, Stephen Schoen- all 50 states and the District of Columbia. Vincent baum, and Anne Gauthier, and the Fund’s com- Mor, Ph.D., at Brown University’s Department munications team, including Barry Scholl, Chris of Community Health, provided the analysis of nursing home admission and readmission rates; Hollander, Martha Hostetter, Mary Mahon, Paul Fronstin, Ph.D., at the Employee Benefit Christine Haran, and Paul Frame, for their Research Institute, provided analysis of uninsured guidance, editorial and production support, and rates derived from the Current Population Survey; public dissemination efforts. Finally, we thank and Gerard Anderson, Ph.D., at the Johns Hopkins Margaret Koller, M.S., of the Rutgers Center University Bloomberg School of Public Health, for State Health Policy, for her work overseeing provided analysis of data on potentially preventable the production of the State Scorecard, and Jim hospital admissions of Medicare patients, as well Walden of Walden Creative for working with us as 30-day hospital readmissions. to design and produce the final report. List of Exhibits E x hibit 1 List of 32 Indicators in State Scorecard on Health System Performance E x hibit 2 State Scorecard Summary of Health System Performance Across Dimensions E x hibit 3 State Ranking on Access and Quality Dimensions E x hibit 4 State Ranking on Health System Performance by Dimension Access E x hibit 5 State Ranking on Access Dimension E x hibit 6 Percent of Adults Ages 18–64 Uninsured by State from 1999–2000 to 2004–2005 E x hibit 7 Percent of Children Ages 0–17 Uninsured by State from 1999–2000 to 2004–2005 E x hibit 8 Percent of Children and Adults Uninsured by State, 2004–2005 E x hibit 9 Median Income, Health Insurance Premiums as Percent of Income, and Percent of Adults Uninsured by State Quality E x hibit 1 0 State Ranking on Quality Dimension E x hibit 1 1 State Variation: Ambulatory Care Quality Indicators E x hibit 1 2 State Variation: Hospital Care Quality Indicators E x hibit 1 3 State Variation: Surgical Infection Prevention E x hibit 1 4 State Variation: Coordination of Care Indicators Potentially Avoidable Use of Hospitals and Costs of Care E x hibit 1 5 State Ranking on Potentially Avoidable Use of Hospitals and Costs of Care Dimension E x hibit 1 6 State Rates of Ambulatory Care Sensitive Hospital Admissions Among Medicare Beneficiaries E x hibit 1 7 State Variation: Hospital Admissions Indicators E x hibit 1 8 Medicare Reimbursement and 30-Day Readmissions by State Equity E x hibit 1 9 Equity Dimension and Equity Type Ranking E x hibit 2 0 Lack of Recommended Preventive Care by Income and Insurance E x hibit 2 1 Absence of a Medical Home by Income and Insurance E x hibit 2 2 Quality and Access Indicators by Race/Ethnicity, National Averages Healthy Lives E x hibit 2 3 State Ranking on Healthy Lives Dimension E x hibit 2 4 Mortality Amenable to Health Care by State E x hibit 2 5 Mortality Amenable to Health Care by Race, National Average and State Variation E x hibit 2 6 National Cumulative Impact If All States Achieved Top State Rates Executive Summary to care, health care quality, reduced costs, and healthier lives. T he rich geographical diversity of the United The analysis of the range of state performance States is part of its appeal. The diverse per- points to five cross-cutting findings: formance of the health care system across • There is wide variation among states. This means the U.S., however, is not. People in the United that the potential exists for the country to do States, regardless of where they live, deserve the much better. best of American health care. The State Scorecard • Leading states consistently outperform lagging is intended to assist states in identifying oppor- states. The patterns indicate that federal and state tunities to better meet their residents’ current policies and local and regional health systems and future health needs and enable them to live make a difference. long and healthy lives. With rising health costs • Across states, better access is closely associated squeezing the budgets of businesses, families, with better quality. and public programs, there is a pressing need to • There are significant opportunities to reduce improve performance and reap greater value from costs as well as improve access to and quality the health system. of care. Higher quality is not associated with The State Scorecard offers a framework through higher costs across states. which policymakers and other stakeholders can • All states have substantial room to improve. gauge efforts to ensure affordable access to high- highlights and key findings quality, efficient, and equitable care. With a goal of focusing on opportunities to improve, the Health care access, quality, cost, and efficiency analysis assesses performance relative to what is vary widely across the United States. achievable, based on benchmarks drawn from the The range of performance is often wide across range of state health system performance. states, with a two- to threefold or greater spread Currently, where you live in the United States from top to bottom. The variability extends to matters for quality and care experiences. The many of the 32 indicators across five dimensions widely varying performance across states and of health system performance: access; quality; sharp differences between top and bottom state potentially avoidable use of hospitals and costs of rates on the 32 indicators included in the State care; equity; and the ability to live long and healthy Scorecard highlight broad opportunities to improve. If all states approached levels achieved lives (referred to as “healthy lives”) (Exhibit 1). by the top states, the cumulative result would Improving performance across the nation to rates be substantial improvement in terms of access achieved by the leading states could save thousands of lives, improve quality of life for millions, and enhance the value gained from our substantial Note: This report summarizes results of the State Scorecard and presents overall state rankings and rankings on each investment in health care. of the five dimensions of health system performance. If all states could approach the low levels of Appendices present state-level data for all indicators. mortality from conditions amenable to care achieved State Scorecard Data Tables with data and state rankings on the 32 health system indicators and data for all equity by the top state, nearly 90,000 fewer deaths before comparisons can be downloaded from the Commonwealth the age of 75 would occur annually. If insurance Fund Web site at www.commonwealthfund.org. The Web rates nationwide reached that of the top states, the site also provides individual state performance profiles uninsured population would be halved. Matching that compare the state to the top state, top five states, and state median rates on all indicators. Also available the performance of the best states on chronic care on the Web site is an analysis of the impact on access, would enable close to four million more diabetics costs, and lives for each state if it were to achieve the top across the nation to receive basic recommended level of performance on each of 11 key indicators. State- specific profiles can be downloaded from the Web site. care and avoid preventable complications, such as renal failure or limb amputation. By matching levels EXHIBIT 1 List of 32 Indicators in State Scorecard on Health System Performance Range of State All States Performance Top Access Year Median (Bottom – Top) State 1. Adults under age 65 insured 2004–2005 81.5 69.6 – 89.0 MN 2. Children insured 2004–2005 91.1 79.8 – 94.9 VT 3. Adults visited a doctor in past two years 2000 83.4 73.9 – 91.5 DC 4. Adults without a time when they needed to see 2004 87.2 80.1 – 96.6 HI a doctor but could not because of cost Quality 5. Adults age 50 and older received recommended 2004 39.7 32.6 – 50.1 MN screening and preventive care 6. Adult diabetics received recommended preventive care 2004 42.4 28.7 – 65.4 HI 7. Children ages 19–35 months received all 2005 81.6 66.7 – 93.5 MA recommended doses of five key vaccines 8. Children with both medical and dental preventive care visits 2003 59.2 45.7 – 74.9 MA 9. Children with emotional, behavioral, or developmental 2003 61.9 43.4 – 77.2 WY problems received mental health care 10. Hospitalized patients received recommended care for acute 2004 83.4 79.0 – 88.4 RI myocardial infarction, congestive heart failure, and pneumonia 11. Surgical patients received appropriate timing 2004–2005 69.5 50.0 – 90.0 CT of antibiotics to prevent infections 12. Adults with a usual source of care 2004 81.1 66.3 – 89.4 DE 13. Children with a medical home 2003 47.6 33.8 – 61.0 NH 14. Heart failure patients given written instructions at discharge 2004–2005 49 14 – 67 NJ 15. Medicare patients whose health care provider always listens, 2003 68.7 63.1 – 74.9 VT explains, shows respect, and spends enough time with them 16. Medicare patients giving a best rating for health care received 2003 70.2 61.2 – 74.4 MT 17. High-risk nursing home residents with pressure sores 2004 13.2 19.3 – 7.6 ND 18. Nursing home residents who were physically restrained 2004 6.2 15.9 – 1.9 NE Potentially Avoidable Use of Hospitals & Costs of Care 19. Hospital admissions for pediatric asthma per 100,000 children 2002 176.7 314.2 – 54.9 VT 20. Asthmatics with an emergency room or urgent care visit 2001–2004 15.5 29.4 – 9.1 IA 21. Medicare hospital admissions for ambulatory care 2003 7,278 11,537 – 4,069 HI sensitive conditions per 100,000 beneficiaries 22. Medicare 30-day hospital readmission rates 2003 17.6 23.8 – 13.2 VT 23. Long-stay nursing home residents with a hospital admission 2000 16.1 24.9 – 8.3 UT 24. Nursing home residents with a hospital readmission within three months 2000 11.7 17.5 – 6.7 OR 25. Home health patients with a hospital admission 2004 26.9 46.4 – 18.3 UT 26. Total single premium per enrolled employee at private- 2004 $3,706 $4,379 – 3,034 UT sector establishments that offer health insurance 27. Total Medicare (Parts A & B) reimbursements per enrollee 2003 $6,070 $8,076 – 4,530 HI Healthy Lives 28. Mortality amenable to health care, deaths per 100,000 population 2002 96.9 160.0 – 70.2 MN 29. Infant mortality, deaths per 1,000 live births 2002 7.1 11.0 – 4.3 ME 30. Breast cancer deaths per 100,000 female population 2002 25.3 34.1 – 16.2 HI 31. Colorectal cancer deaths per 100,000 population 2002 20.0 24.6 – 15.3 UT 32. Adults under age 65 limited in any activities because 2004 15.3 22.8 – 10.8 DC of physical, mental, or emotional problems Note: All values are expressed as percentages unless labeled otherwise. See Appendices B1 and B2 for data source and definition of each indicator. SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 achieved in the best-performing states, the nation could save billions of dollars a year by reducing potentially preventable hospitalizations or readmis- sions, and by improving care for frail nursing home residents. If annual per-person costs for Medicare in higher-cost states came down to median rates or those achieved in the lowest quartile of states, the nation would save $22 billion to $38 billion per year. While some savings would be offset by the costs of interventions and insurance coverage expansions, there would be a net gain in value from a higher-performing health care system. Leading states consistently outperform lagging states on multiple indicators and dimensions. Thirteen states—Hawaii, Iowa, New Hampshire, Vermont, Maine, Rhode Island, Connecticut, Mas- sachusetts, Wisconsin, South Dakota, Minnesota, Nebraska, and North Dakota—emerge at the top quartile of the overall performance rankings (Exhibit 2). These states generally ranked high on multiple indicators in each of the five dimensions assessed by the State Scorecard. Many have been leaders in reforming and improving their health systems and have among the lowest uninsured rates in the nation. Conversely, the 13 states at the bottom quartile of the overall performance ranking—Califor- nia, Tennessee, Alabama, Georgia, Florida, West Virginia, Kentucky, Louisiana, Nevada, Arkansas, Texas, Mississippi, and Oklahoma—lag well behind their peers on multiple indicators across dimensions. Uninsured rates for adults and children in these states are well above national averages and more than double those in the quartile of states with the lowest rates. The rates for receipt of recommended preventive care are generally low, and mortality rates from condi- tions amenable to health care often high. Health system performance often varies re- gionally. Across dimensions, states in the Upper Midwest and Northeast often rank in the highest quartile of performance, with those in the lowest quartile concentrated in the South. States can look to each other for evidence of effective policies and strategies associated with higher performance. For example, in 1974, 10 Hawaii became the first state to enact legislation Better access is associated with better requiring employers to provide health insurance quality across states; insurance matters. to full-time workers; it now ranks first in terms of Across states, better access to care and higher rates access to care. For the past decade, Rhode Island of insurance are closely associated with better has provided incentive payments to Medicaid quality (Exhibit 3). States with the lowest rates managed care plans that reach quality targets; of uninsured residents tend to score highest on it now ranks first on measures of the quality measures of preventive and chronic disease care, of care. Maine, Massachusetts, and Vermont as well as other quality indicators. lead in providing equitable health systems; the Four of the five leading states in the access three states are recognized for their innovation dimension—Massachusetts, Iowa, Rhode Island, and leadership on expanding health insurance and Maine—also rank among the top five states coverage and benchmarking for quality. in terms of quality. Moreover, states with low quality rankings tend to have high rates of The patterns indicate that federal and state uninsured. This cross-state pattern points to policies plus local and regional health care the importance of affordable access as a first systems make a difference. Leading states out- step to ensure that patients obtain essential care perform lagging states on multiple indicators and receive care that is well coordinated and that span the dimensions of access, quality, cost, patient-centered. In states where more people equity, and healthy lives. are insured, adults and children are more likely 11 to have a medical home and receive recom- state variation: mended preventive and chronic care. Identifying highlights by dimension care system practices as well as state policies that Access promote access to care is essential to improving quality and lowering costs. • The percent of adults under age 65 who were The number of uninsured children has uninsured in 2004–2005 ranges from a low of declined following enactment of federal Medicaid 11 percent in Minnesota to a high of 30 percent and State Children’s Health Insurance Program in Texas. The percent of uninsured children (SCHIP) expansions for children. Yet, the high varies fourfold, from 5 percent in Vermont to and rising rates of uninsured adults put states 20 percent in Texas. and the nation at risk as adults lose affordable • Over the past five years, the number of states access and financial security. The deterioration with more than 16 percent of children uninsured in coverage and the relationship between better declined from 10 to three. In contrast, the coverage and better care point to a pressing need number of states with 23 percent or more of for national action to expand insurance coverage adults uninsured increased from four to 12. and ensure access to care. • In all but six states, the percent of adults uninsured increased. Notable exceptions include Higher quality does not mean higher costs. Maine and New York, which have both expanded Annual costs of care vary widely across states, with programs to insure low-income adults. no systematic relationship to insurance coverage • Across states, three of four uninsured adults age 50 or older did not receive basic preventive or ability to pay as measured by median incomes. care, including cancer screening. The percent of Moreover, there is no systematic relationship adults who reported going without care because between the cost of care and quality across states. of costs is up to five times greater in states with Some states achieve high quality at lower costs. high rates of uninsured adults than in states with States with higher medical costs tend to have the lowest uninsured rates. higher rates of potentially preventable hospital • The nation would insure 22 million more adults use, including high rates of readmission within and children if all states moved to the level of 30 days of discharge and high rates of admission coverage provided in the top-performing states. for complications of diabetes, asthma, and other chronic conditions. Reducing the use of Quality expensive hospital care by preventing compli- • Even in the best states, performance falls far cations, controlling chronic conditions, and short of optimal standards. The percent of adults providing effective transitional care following age 50 or older receiving all recommended discharge has the potential to improve outcomes preventive care ranges from a high of 50 percent and lower costs. in Minnesota to 33 percent in Idaho. The percent There is room to improve in all states. of diabetics receiving basic preventive care services varies from 65 percent in Hawaii to 29 All states have substantial room to improve. On percent in Mississippi. some indicators, even the top rates are well below • Childhood immunization rates range from 94 what should be achievable. There are also substan- percent in Massachusetts to less than 75 percent tial variations in performance within states. in the bottom five states. The percent of children Among the top-ranked states, each had some with a medical home that helps coordinate indicators in the bottom quartile or bottom half care ranges from a high of 61 percent in New of the performance distribution. Understanding Hampshire to less than 40 percent in the bottom how underlying care system features and popula- 10 states. tion factors contribute to performance variations • Discharge planning varies markedly. The percent will help inform efforts to improve. of congestive heart failure patients receiving 12 complete hospital discharge instructions ranges savings. If annual per-person costs for Medicare from 33 percent or less in the bottom five states in higher-cost states came down to median rates to 67 percent in New Jersey. or the lowest quartile, the nation would save $22 • If all states reached the levels achieved among billion to $38 billion per year. the top-ranked states, almost nine million Equity more older adults would receive recommended preventive care, and almost four million more • Equity gaps by income and insurance status on diabetics would receive care to help prevent quality indicators exist in most states. The gaps disease complications. Likewise, about 33 are widest in states that perform poorly overall million more adults and children would have on quality and access indicators. a usual source of care or medical home to help • On average, 78 percent of uninsured and 71 coordinate care. percent of low-income adults age 50 and older did not receive recommended preventive Potentially Avoidable Use of services. By comparison, 59 percent of insured Hospitals and Costs of Care adults and 54 percent of higher-income adults • State rates of hospital admission for childhood failed to receive such care. asthma range from a low of 55 per 100,000 • The pattern extends to diabetics. On average, children in Vermont to more than 300 per 67 percent of low-income diabetics did not 100,000 in South Carolina. receive basic care according to guidelines for • Rates of potentially preventable hospital their condition. admission among Medicare beneficiaries range • In some states, equity rankings were low as from more than 10,000 per 100,000 beneficiaries a result of large disparities among minority in the five states with the highest rates to less groups that comprise relatively small shares than 5,000 per 100,000 in the five with the of the state population. For example, in lowest rates (Hawaii, Utah, Washington, Alaska, Minnesota, indicators of health care quality and Oregon). were often low for a group that included Asian • Similarly, there is a twofold variation in rates Americans and Native Americans. A focus of hospital readmission within 30 days among on these groups would have a high return in Medicare beneficiaries (from 24 percent in reducing health disparities. Louisiana and Nevada to only 13 percent in Healthy Lives Vermont and Wyoming) and a threefold range in rates of hospital admission among nursing • There is a twofold range across states in the rate of home residents, from 25 percent (Louisiana) to deaths before age 75 from conditions that might only 8 percent (Utah). have been prevented with timely and appropriate • High rates of potentially avoidable hospital use health care. Potentially preventable death rates in and repeat admissions are closely correlated with the states with the lowest mortality (Minnesota, high costs of care. States with the highest rates Utah, Vermont, Wyoming, and Alaska) are 50 of readmission have annual Medicare costs per percent below rates in the District of Columbia person 38 percent higher than states with the and states with the highest rates (Tennessee, lowest rates. Arkansas, Louisiana, and Mississippi). • If all states reached the low levels of potentially • There are wide differences in this dimension preventable admissions and readmissions, among racial groups. For example, age- hospitalizations could be reduced by 30 to 47 standardized death rates for conditions amenable percent and save Medicare $2 billion to $5 billion to health care are twice as high for blacks as for each year. Potential savings would be still greater whites nationwide (194 versus 94 per 100,000 if the interventions applied to all patients. population). Southern states and some states • Improving care and developing more efficient in the Midwest with large black populations care systems have the potential to generate major have the greatest racial disparities, with more 13 than 100 additional deaths per 100,000 black The findings point to the need for action in the residents above the overall national average. following key areas: Yet, racial disparities exist even in states with • Universal coverage: This is critical for improving narrower gaps. quality and delivering cost-effective care, as • Potentially preventable mortality rates for whites well as ensuring access. Federal action as well also vary significantly across states, ranging as state initiatives will be essential for progress from a low of 67.6 per 100,000 population nationwide. (Minnesota) to a high of 118.3 (West Virginia). • More information to assess performance and In general, white rates are highest in states with identify benchmarks: It takes information high overall rates. to guide and drive change. We need more • If death rates in all states improved to levels sophisticated information systems and better achieved by the best state (Minnesota, with information on practices and policies that 70.2 deaths per 100,000), about 90,000 fewer contribute to high or varying performance. premature deaths would occur annually. • Analyses to determine the key factors that • Health system performance is only one of many contribute to variations: States can use such forces that shape health status and longevity. information to develop evidence-based strategies Family history and immigration status can for improvement. affect state-level population health indicators. • National leadership and collaboration across Risk factors, such as smoking and obesity, vary public and private sectors: This is essential for across states. Public health policies, including coherent, strategic, and ultimately effective workplace and environmental regulations, are improvement efforts. thus critical components for long and healthy Benchmarks set by leading states, as well as lives. The indicators in this dimension are likely exemplary models within the United States and to be sensitive to health system performance other countries, show that there are broad oppor- broadly defined, modifiable through both tunities to improve and achieve better and more improved care and public health policies. affordable health care. With health costs rising summary and implications faster than incomes and straining family, business, state, and federal budgets, with access deteriorat- The view of health system performance across the ing, and with startling evidence of variable quality nation reveals startlingly wide gaps between leading and inefficient care, all states and the nation have and lagging states on multiple indicators. The gaps much to gain from aiming higher. All states can represent illnesses that could have been prevented do better; and all should continually ask, “Why or better managed, as well as costs that could have not the best?” been saved or reinvested to improve population health. The State Scorecard indicates that we have much to gain as a nation by aiming higher with a coherent set of national and state policies that respond to the urgent need for action. States play many roles in the health system—as purchasers of public coverage and coverage for their employees, regulators of providers and insurers, advocates for the public health, and, increasingly, conveners and collaborators with other stake- holders. States also can provide a source of public reports on quality and costs. These roles provide potential leverage points to promote better access and quality and to address rising costs. 14 Introduction system performance that spans all core dimen- sions of system performance. The central goal G rowing public and business concerns of the state-level analysis is to inform action to about the affordability of health care, ensure that residents of every state have access eroding health insurance coverage, and to high-quality and efficient care in systems that broad evidence of variable quality and inefficient strive to improve population health. Toward this care have sparked renewed calls for state and goal, the State Scorecard offers a tool for national national policy leadership. States are increasingly and state policymakers and other stakeholders initiating reforms that seek to improve health care to gauge efforts to improve performance and access and quality and, at the same time, address identify targets for change. the high and rising costs of care. Highly variable The State Scorecard includes 32 key indicators, performance across geographic regions of the grouped into five dimensions of performance: United States attests to the potential to improve. access to care, quality, avoidable hospital use and As states confront shared challenges of how to costs of care, equity, and the ability to live long meet their populations’ health needs and achieve and healthy lives (referred to as “healthy lives”). higher-value health systems, benchmarks drawn The analysis examines the range of variation from the range of achieved performance across across states and assesses performance relative states can provide targets and focus attention on to what has already been achieved by individual opportunities to improve. states. The scorecard ranks all 50 states and the The Commonwealth Fund National Scorecard District of Columbia on each indicator and on on U.S. Health System Performance, published in each of the five dimensions of performance. September 2006, assessed national health system The dimension rankings are then used to derive performance across core dimensions of access, an overall ranking. (The box below explains quality, efficiency, equity, and long and healthy the State Scorecard methodology and describes lives.1 The findings documented striking varia- limitations on data currently available at the tions across geographic regions of the United state level.) States. and highlighted the potential national Summary exhibits show indicators, the range gain if all areas of the country could achieve the of variation across states, and overall state performance levels of leading geographic areas rankings and ranks within dimensions. Exhibit 1 or health care systems. lists the indicators included in each dimension of This State Scorecard on Health System Per- performance and illustrates the range of perfor- formance builds on the National Scorecard and mance across states. Exhibit 2 shows the overall provides a framework for assessing state health state ranking by quartile. Exhibit 3 compares access and quality rankings. Note: This report summarizes results of the State Scorecard Exhibit 4 shows overall state rankings and and presents overall state rankings and rankings on each where each state ranks in the five dimensions. of the five dimensions of health system performance. The appendix to this report provides data for all Appendices present state-level data for all indicators. State Scorecard Data Tables with data and state rankings of the indicators organized by dimension. The on the 32 health system indicators and data for all equity appendix also includes demographic tables that comparisons can be downloaded from the Commonwealth profile states by income, incidence of poverty, Fund Web site at www.commonwealthfund.org. The Web and health risks. site also provides individual state performance profiles that compare the state to the top state, top five states, In the sections that follow, we present the and state median rates on all indicators. Also available State Scorecard results, organized by the five di- on the Web site is an analysis of the impact on access, mensions of performance. The discussion focuses costs, and lives for each state if it were to achieve the top on key indicators and gains possible within each level of performance on each of 11 key indicators. State- specific profiles can be downloaded from the Web site. dimension if all states were to achieve the per- formance level of the top states. 15 EXHIBIT 4 State Ranking on Health System Performance by Dimension = State in top quartile * Final rank for overall health system performance across five dimensions SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 16 Looking across dimensions, the summary section of the report discusses the primary, W hat the S tate S corecard M easures cross-cutting findings based on state patterns Dimensions and Indicators and variations. These include: The State Scorecard measures health system performance for all 50 states and the District of Columbia using 32 key • There is wide variation among states. The indicators (Exhibit 1). It organizes indicators by five broad dimensions that capture critical aspects of health system variations attest to the potential for the performance: country as a whole to do much better. • Access includes rates of insurance coverage for adults • Leading states consistently outperform and children and indicators of access and affordability of care. lagging states on multiple indicators that span • Quality includes indicators that measure three related dimensions of health system performance. components: receipt of the “right care,” coordinated care, and patient-centered care. The patterns indicate that federal and state • Potentially avoidable use of hospitals and costs of policies and local and regional health systems care includes indicators of hospital care that might have make a difference. been prevented with appropriate care and follow-up, as well as the annual costs of Medicare and private health • Across states, better access is closely associated insurance premiums. with better quality. • Equity includes differences in performance associated • Yet, higher quality is not systematically with patients’ income level, type of insurance, or race or ethnicity. associated with higher costs. There are significant • Healthy lives includes indicators that measure the degree opportunities to reduce costs as well as improve to which a state’s residents enjoy long and healthy lives. access to and quality of care. Throughout the text, lists of states appear in order of their • All states have substantial room to improve. ranking on the indicator being discussed. Whenever possible, indicators were selected to be equivalent The final sections of the report examine the to those used in the National Scorecard on U.S. Health System potential impact of improving performance Performance. However, comparable state-level data were not available for some important topics covered by the and implications for policy action. The analysis National Scorecard. In particular, as a nation, we lack state- includes estimates of the cumulative gain if all level indicators to measure how well patients and their doctors are controlling chronic diseases and how often states were to achieve the top level of perfor- patients experience adverse effects from their treatment, mance within the current range of state variation as well as other safety indicators. We also lack state-level data on duplicative services, receipt of inappropriate care, on each of 11 key indicators. insurance administrative overhead, and information system The State Scorecard, overall, indicates that capacity. Moreover, many quality metrics are still in the early stages of development and thus are limited in scope. Hence, we have much to gain as a nation from national State Scorecard indicators should be considered a “starter and state policies that aim higher. The conclud- set” to be expanded over time. See Appendices B1 and B2 for data sources and descriptions for each of the indicators ing remarks outline key areas in which state and included in the State Scorecard. federal action will be critical to move forward. Scorecard Ranking Methodology The State Scorecard first ranks states from best to worst on each of the 32 performance indicators. We averaged rankings for those indicators within each of the five dimensions to determine a state’s dimension rank and then averaged the dimension rankings to arrive at an overall ranking on health system performance. This approach gives each dimension equal weight and, within dimensions, weights indicators equally. We use average state rankings for the State Scorecard because we believe that this approach is easily understandable. This method follows that used by Stephen Jencks and colleagues when assessing quality of care for Medicare beneficiaries at the state level across multiple indicators.2 For the equity dimension, we ranked states based on the difference between the most vulnerable subgroup (i.e., low- income, uninsured, or racial/ethnic minority) and the U.S. national average on selected indicators. The gap indicates how the vulnerable subgroup fares compared with the U.S. average—an absolute standard. 17 Access primary care, health centers, and other safety A net resources. ccess to health care is the foundation The State Scorecard includes four indicators and hallmark of a high performance of access: the percent of adults and children who health system. The foremost factor in are covered by health insurance, the percent of determining whether people have access to care adults who have visited a doctor in the last two when needed is having insurance that covers years, and the percent of adults who reported essential care. The extent to which insurance going without care because of costs. provides affordable access also depends on the These insurance coverage and access indi- design of benefits, and whether provider payment cators vary significantly across states. With a policies secure adequate networks of primary and few exceptions, states in the Upper Midwest specialized care. and the Northeast, along with Hawaii, lead the States can do much to improve both afford- nation on access, ranking in the top quartile able access and efficiency in the organization of all states. States in the South-Central and of insurance and delivery of care through their Western United States have the largest gaps in oversight of health insurance markets, purchase access (Exhibit 5). of insurance for state employees, and support The best-performing states in the access of public insurance initiatives. States also can dimension of performance are among those enhance access in low-income, rural, and other with the most expansive eligibility polices for underserved communities by investing in public health insurance coverage. For example, 18 in Hawaii and Massachusetts, the leading states The top-ranked state, Hawaii, which in 1974 on this dimension, children in families with enacted legislation mandating employers to incomes up to three times the federal poverty provide health coverage, has long been a leader level can enroll in the State Children’s Health in state health policy innovation. Insurance Program (SCHIP). In addition, the five universal par ticipation top-ranked states—Hawaii, Massachusetts, Iowa, Rhode Island, and Maine—provide higher-than- As rising premium costs squeeze workers with average public coverage eligibility for parents. low or modest incomes out of private insurance 19 markets—average family premiums now exceed • Although in all states children are more likely $11,000 per year 3—the proportion of uninsured than nonelderly adults to have health insurance, adults under age 65 has risen dramatically over the proportion of uninsured children varies from the five-year period 1999–2000 to 2004–2005 a low of 5 percent in Vermont to a high of 20 (Exhibit 6). Based on annual census data, the percent in Texas—a rate four times higher. number of states where 23 percent or more of the • Reflecting differences in state coverage policies, adult population is uninsured tripled, from four to trends in coverage for adults and children have 12.4 In sharp contrast, children fared much better diverged sharply over the past five years. In all during the same time period (Exhibit 7). Thanks but 12 states, the uninsured rate for children has to federal support of Medicaid and state expan- declined. In all but six states, the uninsured rate sions through the SCHIP program, the percent of for adults under 65 has increased. children uninsured declined in most states. In only • Alabama stands out in the South for its three states were more than 16 percent of children particularly low uninsured rates for children. In fact, along with Vermont, Massachusetts, uninsured in 2004–2005, compared with 10 states Hawaii, Iowa, Michigan, and Nebraska, it is in 1999–2000. one of the seven states with the lowest rates Access for low- and modest-income families of uninsured children (Exhibit 8). Alabama’s depends critically on where families live. Insurance success in covering children, despite being coverage rates differ sharply across states. relatively poor and having low levels of private, • Among the states, there is a nearly threefold job-based insurance coverage, reflects its variation in the percent of adults under age 65 decision early on to expand SCHIP coverage who were uninsured in 2004–2005, ranging for children in families with incomes up to from a low of 11 percent in Minnesota to a high 200 percent of the poverty level and to pursue of 30 percent in Texas. aggressive enrollment policies. 20 Gaps in insurance coverage create substantial exceed rates in New York, based on national barriers to care and expose people to financial in- employer surveys ($4,116 for the average annual security. If all states achieved the level of coverage single premium in Maine versus $3,858 in New in leading states, 17.2 million more adults and 4.4 York; see Appendix Exhibit A8). million more children would have insurance. The Affordability and coverage are at risk when number of uninsured across the nation would insurance premiums are high relative to family be halved. income or when coverage fails to provide adequate financial protection relative to income. The average access: use and cost barriers cost of employer-sponsored health plans as a percent In addition to insurance, use of health care services of median state income ranges from a low of 11.8 provides another marker of access to care. In some percent in Maryland to 19.3 percent and 19.7 percent instances, poor access to preventive and primary in West Virginia and New Mexico, respectively care can actually increase utilization of hospital (Exhibit 9). Uninsured rates are generally higher in services (see Potentially Avoidable Use of Hospitals states where premiums are higher relative to average on page 30). For most people, good access to care income: 26 percent of adults are uninsured in the should include at least some regular contact with five states with the least affordable premiums, while an ambulatory care provider. The proportion of 16 percent are uninsured in the five states with the adults who have visited a doctor at least once in the lowest premiums. prior two years provides a marker of such access. Historically, the nation has relied on employ- Infrequent physician contact might be a red flag for ment-based insurance to cover the under-65 access barriers, indicating possible gaps in insurance population. In all states, low-wage jobs are the coverage, inadequate benefits, or shortages of ac- least likely to have job-based health benefits. As a cessible sources of care. Overall, higher rates of result, low-income families—those with incomes insurance coverage are associated with higher rates below 200 percent of poverty—are at the highest of physician contact (see Appendix Exhibit A3). risk of being uninsured. Thus, uninsured rates tend to be highest in states with low average affordability incomes and a high percentage of poor or “near- poor” (those with incomes under 200 percent of Insurance is critical for affordable access. The the federal poverty level) residents. percent of adults who go without needed care Yet, there are notable exceptions that serve to because of costs is up to five times greater in states underscore the important role of public policy. that have high rates of uninsured adults, compared Maine’s uninsured rates are among the lowest in with states with the lowest uninsured rates. Notably, the country, despite insurance premiums that are only about 4 percent of Hawaii’s adult population high relative to incomes. In contrast, uninsured reported they did not see a doctor when needed rates are relatively high in Colorado, New Jersey, because of costs. In contrast, nearly 20 percent of and Utah, even though statewide median incomes adults—one of five—in Mississippi, West Virginia, are higher there and premiums as a percent of and Texas went without care because of costs. These median income are lower than those found in states have among the highest rates of uninsured Maine (Exhibit 9). adults in the nation. state and federal polic y The ability to afford health care and health insurance depends on family income as well as State policies can help make insurance coverage the breadth and comprehensiveness of insurance more affordable for low-income families and busi- coverage. Premiums vary narrowly across the nesses. Several states have undertaken coverage country; often they are nearly as high, or higher, expansions that target small businesses and in- in low-income states as in high-income states. dividuals with moderate incomes who cannot For example, Maine’s average premiums for afford to purchase private or publicly sponsored employer-sponsored insurance rival and even coverage without a subsidy. For example, despite 21 ACCESS EXHIBIT 9 Median Income, Health Insurance Premiums as Percent of Income, and Percent Adults Uninsured by State Employer-based insurance premiums as percent Percent of adults under Median household income of median Income age 65 uninsured 2004–2005 Rank 2003 Rank 2004–2005 Rank United States $46,071 14.9 20.5 Alabama 37,502 45 14.9 26 20.2 34 Alaska 56,398 6 15.5 33 23.0 40 Arizona 45,279 25 16.3 39 24.0 43 Arkansas 36,406 48 17.3 44 24.4 45 California 51,312 12 14.8 24 24.5 46 Colorado 51,518 11 13.8 13 20.1 32 Connecticut 56,889 5 12.6 6 14.9 12 Delaware 50,445 14 15.4 32 16.7 19 District of Columbia 44,949 27 16.9 42 16.7 19 Florida 42,440 36 16.2 37 26.9 50 Georgia 44,140 30 14.9 27 23.4 41 Hawaii 58,854 3 12.1 2 12.8 3 Idaho 45,009 26 15.5 34 20.1 33 Illinois 48,008 18 14.7 23 18.1 23 Indiana 43,091 34 15.0 28 18.6 27 Iowa 45,671 24 13.1 9 12.2 2 Kansas 42,233 37 14.5 19 14.8 10 Kentucky 36,750 47 16.8 40 18.4 25 Louisiana 37,442 46 17.8 47 25.3 47 Maine 43,317 31 17.7 46 13.7 5 Maryland 59,762 2 11.8 1 18.5 26 Massachusetts 54,888 8 12.4 5 14.6 8 Michigan 44,801 28 14.7 22 15.9 16 Minnesota 56,098 7 12.9 7 11.0 1 Mississippi 34,396 51 16.8 40 22.3 37 Missouri 43,266 32 14.1 16 16.5 18 Montana 36,202 49 17.8 48 23.7 42 Nebraska 46,587 20 14.4 18 15.8 15 Nevada 48,496 17 15.0 29 22.5 39 New Hampshire 57,850 4 12.3 4 14.0 6 New Jersey 60,246 1 12.2 3 18.9 29 New Mexico 39,916 42 19.7 51 26.1 49 New York 46,659 19 15.1 31 17.9 21 North Carolina 41,820 39 15.6 35 20.4 35 North Dakota 41,362 40 13.3 10 14.7 9 Ohio 44,349 29 14.5 20 15.6 14 Oklahoma 39,292 44 17.1 43 25.5 48 Oregon 43,262 33 15.1 30 22.0 36 Pennsylvania 45,941 22 13.8 12 14.2 7 Rhode Island 49,511 16 14.2 17 14.8 10 South Carolina 40,107 41 16.2 38 22.4 37 South Dakota 42,816 35 14.6 21 16.3 17 Tennessee 39,376 43 17.4 45 18.7 28 Texas 42,102 38 18.4 49 30.4 51 Utah 53,693 9 14.0 14 19.3 31 Vermont 49,808 15 14.1 15 15.5 13 Virginia 52,383 10 12.9 8 18.3 24 Washington 51,119 13 13.7 11 18.0 21 West Virginia 35,467 50 19.3 50 24.1 44 Wisconsin 45,956 21 14.8 24 13.5 4 Wyoming 45,817 23 16.0 36 19.0 29 DATA: Median income - 2005 and 2006 Current Population Survey; Premium as percent of income - 2003 Medical Expenditure Panel Survey Insurance Component (premium) and 2004 and 2005 Current Population Surveys (income); Adults uninsured - 2005 and 2006 Current Population Survey SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 22 being a low-income state with premiums well Quality above the national average, Maine has among the P lowest adult uninsured rates in the nation, thanks atients and families seeking health to public expansion efforts in that state. and medical care expect that their care Differences in uninsured rates for adults and providers will recommend and give them children reflect eligibility criteria for public coverage as well as the extent of private coverage the right services, that their care will be well through employers. More than two-thirds of coordinated, and that those delivering services states extend SCHIP coverage to children with will be responsive to their needs. The organization family incomes up to 200 percent of the federal and delivery of health care by public and private poverty level or higher (up to 300%). In stark providers play critical roles in shaping the quality contrast, 35 states set the Medicaid eligibility and responsiveness of health services. States can threshold for parents below 100 percent of the create incentives for quality and join with private poverty level—in 14 states a family would have and public sector leaders and other payers to to have income at a level less than 50 percent of promote a more responsive and effective health poverty before parents would qualify.5 Moreover, care delivery system in many ways, such as by: most states cover childless adults only if they are blind or disabled. This sharp contrast between • sponsoring public programs and initiatives (e.g., adults’ and children’s coverage underscores how vaccine delivery programs and registries, pay- federal policy (in this case, federal funding for for-performance reimbursement strategies); SCHIP) stimulates and supports state action. • promoting quality in public coverage and public Other state policies and strategies are also employee programs through contracting and important, such as simplifying the enrollment participation requirements for health plans and process for public coverage and engaging in providers; outreach to ensure that all who qualify partici- • collaborating with private and public providers pate in programs intended for them. in quality initiatives; and States with high rates of uninsured adults also • monitoring and benchmarking performance tend to have historically low rates of employer- through public reporting, all-payer databases, based coverage. Many of the states with high rates and regulatory/licensing standards. of uninsured and low employer coverage also have a high percent of the working-age population with The State Scorecard includes 14 indicators in the earnings at or below 200 percent of poverty. Given quality dimension. These include: seven assessing these characteristics, federal action to support the extent to which adults and children receive the expansion and raise the floor of income eligibility “right care” (preventive care and care according for public programs will likely be necessary to to medical guidelines when hospitalized); three stimulate substantial progress nationwide. assessing care coordination; and four assessing patient-centered care for elderly or long-term care patients (see Exhibit 1 for indicators). As with other dimensions in the State Scorecard, there are wide variations in quality performance across states, as well as variation among indicators within states. There are also distinct geographic patterns in states’ overall rankings on quality (Exhibit 10). With some exceptions, states in the South, Southwest, and West ranked lowest on this dimension, while states in the Upper Midwest and Northeast ranked highest. The five highest-ranking states 23 on quality in rank order were Rhode Island, where improvement requires strategies focused on Maine, Massachusetts, Connecticut, and Iowa specific conditions, populations, or care settings, (see Appendix Exhibit A4). These states ranked in states can look to peers that perform well on par- the top quartile across eight to 12 of the 14 quality ticular indicators. indicators, with generally high performance on getting the right care indicators of the right care and coordinated care. However, performance on indicators of patient- On average across seven indicators of right care, less centered care, which are based on the experiences than two-thirds of children and adults receive care of Medicare beneficiaries, did not consistently consistent with established guidelines and profes- track that of right care and coordinated care. sional recommendations, ranging from about three- Understanding the health system and policy quarters in the top five states (ranked separately for factors that contribute to higher rates in the each indicator) to only about one-half in the bottom leading states may offer insights for achieving five states. Bottom-ranking states would need to higher quality, to the degree that improvement improve their performance by about 40 percent, strategies are transferable across indicators and on average, to reach the level of the top-ranking states. Yet, even leading states did not perform states. Performance on these indicators varies not consistently well across all 14 indicators, demon- only by state but also by subpopulation and setting strating that there are opportunities for all states to of care, with substantial room for improvement improve (Appendix Exhibits A4 and A5). In cases across all states on most indicators. 24 • Two indicators of ambulatory care quality for • A high proportion of hospitalized adults receive adults reveal especially large gaps. Among top- evidence-based treatment for heart attack, heart performing states, only one-half of adults age 50 failure, and community-acquired pneumonia. or older receive recommended cancer screenings On a composite of 10 quality indicators, state and vaccinations and only about two-thirds of rates range from a high of 88 percent in Rhode diabetics receive three recommended services Island to about 80 percent in the lowest- (blood screening, foot exams, and eye exams). performing states (Exhibit 12).6 Yet, quality Rates of diabetes management in the bottom indicators for pneumonia and congestive five states, where only about one-third of adults heart failure show lower average performance receive recommended care, would have to more and greater state-to-state variation than do than double to reach the level achieved by the indicators for heart attack treatment. Moreover, top states (Exhibit 11). states performing well on one condition do not • Performance on indicators of ambulatory necessary perform well on others, suggesting care quality is better for children than for a need for systematic approaches to achieve adults, although gaps remain. Rates of five key consistently high performance (see Appendix childhood vaccinations, for instance, range Exhibit A6).7 from a high of 94 percent in Massachusetts to • There is wide variation in performance on an average of 71 percent in the five lowest states one measure of patient safety included in the (Exhibit 11). There is greater variability among State Scorecard, the provision and appropriate states on annual preventive health and dental timing of antibiotics to prevent infections among visits for children. surgical patients (Exhibit 13). Rates range across 25 26 states from 50 percent in Nevada to 90 percent in congestive heart failure—shows states falling Connecticut, indicating substantial opportunity seriously behind a care standard (Exhibit 14). for improvement in poorly performing states. Congestive heart failure is a complex condition that frequently requires hospitalization when care coordinated care management fails. This condition is particularly Errors and omissions in health care can occur sensitive to poor care coordination, as patients because of coordination gaps that result in poor and their doctors often must manage multiple access to patient information, delays, and failures medications as well as complex diet and physical to communicate and exchange critical medical activity regimens. In 2004–2005, more than 50 history, test results, or medication information percent of heart failure patients in over half of when patients move from one health care setting the states did not receive complete discharge to another. Poor coordination can also lead to un- instructions. About two-thirds of patients in the necessary costs and exposure of patients to risk top-ranked states received discharge instructions, because of redundant and unnecessary medical compared with only about one-quarter in the testing. Data on three indicators of coordinated care bottom-ranked states. reveal some of the greatest regional and interstate variation and performance gaps among all quality All states have substantial room for improve- indicators (Exhibit 14). ment on right care and coordination of care metrics. Even the best rates are often low. Yet, the gaps in • Having a “usual source of care” (a personal doctor quality between leading and other states point to or other provider) is one well-accepted marker substantial missed opportunities for primary and of continuity and coordination of care. Most preventive care. For example: adults have a usual source, but the proportion varies from 89 percent in Pennsylvania to only • If performance in all of the states on screening 71 percent in the five bottom-ranking states rates for older adults or basic care for diabetics (Exhibit 14). reached levels achieved by the top states, 8.6 • A more comprehensive indicator of the provider– million additional older adults and 3.6 million patient relationship is available for children. For diabetics would receive basic care according to this measure, a “medical home” is defined as clinical guidelines. having an accessible primary source of care to • If adults and children were connected with usual help coordinate care and receiving all needed sources of care at the rates achieved in top states, care, including at least one preventive care visit an additional 33 million would have a primary in the prior year.8 Only 61 percent of children care connection. have care arrangements meeting this standard patient- centered care in New Hampshire, the best-performing state. The range is wide: just over one of three Patient-centered care takes into account patients’ children in the five bottom-ranked states report preferences, needs, and values. Patients’ experi- having such a medical home (Exhibit 14). The ences can, in turn, influence the way they use and lower performance on this measure of care benefit from the health care system and manage coordination, as compared with adults having a their conditions. usual source of care, probably reflects the stricter Patients are most likely to form partnerships with standard being measured. This underscores the their care providers when providers are responsive need for comparable data on this important to concerns and explain medical information in construct for the adult population. ways patients can understand. Among Medicare • One marker of coordination of care for fee-for-service patients, the proportion reporting hospitalized adults—provision of written that their providers always communicated well did discharge instructions for patients with not vary greatly across states, ranging narrowly 27 from 73 percent in top-ranked states (Vermont, Medicare patient reports likely represent a more Maine, Rhode Island, Louisiana, and Montana) positive range of experiences than would a full to 65 percent in bottom-ranked states (Colorado, population survey, as all Medicare beneficiaries are Florida, New Mexico, Utah, and Arizona). Medicare insured for at least a minimum scope of benefits beneficiary ratings of their overall care experiences and elderly patients tend to give higher ratings showed a similar range (see Appendix Exhibit A5). than nonelderly adults.9 The range of performance on these measures was Two measures of patient-centered care for the narrowest among quality-of-care indicators in vulnerable nursing home residents show wide the State Scorecard. interstate variability. Currently, information about patient experi- • Pressure sores can result from inadequate care ences among the under-65 population at the state of patients who have limited ability to move level is not generally available. What information (although pressure sores sometimes occur even exists is spotty. Although, some private health with the best care). About 13 percent of high- plans report CAHPS (Consumer Assessment risk nursing home residents (i.e., comatose of Healthcare Providers and Systems) patient residents and others who cannot move or change survey data to the National Committee for Quality position on their own, as well as residents who Assurance and some states collect information for do not get or absorb the nutrients they need) some of their Medicaid beneficiaries, the current had pressure sores during 2004 across all states. level of reporting is inadequate to view statewide The 17.7 percent pressure sore rate in the worst- experiences for the under-65 population. The performing states (Illinois, Oklahoma, Louisiana, 28 New Jersey, and the District of Columbia) was This State Scorecard dimension also includes more than double the 8.1 percent rate achieved two important indicators of health care cost— by states with the lowest rates (North Dakota, average private health insurance premiums and Montana, Maine, Idaho, and Iowa). Medicare annual spending per enrollee. Higher • Use of physical restraints on nursing home cost is not necessarily a marker of inefficiency residents can contribute to increased if the health system delivers greater access, prevalence of pressure sores and social isolation improved quality, and better outcomes in return. Yet, the National Scorecard and other studies of residents. Federal regulations assert that have found little systematic relationship between nursing home residents have the right to be higher costs and higher quality within the United free of restraints that are not required to treat States. Moreover, international comparisons medical symptoms. Restraints are rarely used indicate that the U.S. health system as a whole is on nursing home residents in some states not delivering higher value commensurate with (e.g., only 1.9% of residents were restrained the much higher level of spending in the United in Nebraska), suggesting that it is possible States compared with several other nations. to substantially reduce their use. Physical Broad-based evidence from a rich array of restraints are used much more frequently (on studies of health outcome and cost variations an average of 14.1% of residents) in the five states within the United States indicates the potential with the worst performance on this measure for states and the nation to achieve higher-value (Utah, Oklahoma, Louisiana, California, and health systems, supporting better outcomes, higher Arkansas) (Appendix Exhibit A5). quality, improved access, and savings compared with current trends.10 As in the National Scorecard, Some states that are ranked high on the overall State Scorecard indicators in this dimension illus- quality dimension perform poorly on patient- trate the need for policies to move toward high- centered care measures. For instance, Rhode Island, value, efficient health systems that aim higher for the top-ranked state on quality, is in the bottom all core dimensions of performance. quartile in terms of the proportion of high-risk Clinicians and health systems have the nursing home residents with pressure sores. Con- capacity to achieve greater efficiency in patient versely, some states such as Hawaii and Montana care, but their efforts may not be rewarded by the that have lower overall quality scores perform better payment methods in private insurance and public on patient-centered care. programs.11 Policymakers seeking to enhance health system efficiency can use several levers, including recalibrating and aligning incentives Potentially Avoidable Use embedded in Medicare and Medicaid payment systems, regulating the supply of health care facili- of Hospitals and Costs of Care ties, promoting enhanced primary care capacity, E and adopting policies to support better integration fficient health systems should ensure and coordination of care. For example, many states quality, access, and healthy outcomes are exploring the potential of health informa- while minimizing the costs of care. State tion systems to improve quality and efficiency Scorecard indicators in this dimension focus on by linking care across sites and giving physicians an important measure of efficiency: rates of po- and other providers decision-support tools. Public tentially avoidable and expensive hospital care. health initiatives can further address long-term A comprehensive evaluation of health system population health trends (such as rising rates of efficiency would compare broader measures of obesity and chronic diseases) that contribute to inappropriate care, waste, and administrative increased use of health care resources. overhead, but such measures are not currently Rates of potentially avoidable hospital admission available at the state level. from complications of chronic disease and rates of 29 hospital readmission are indicators of health care patients, hospitalizations and re-hospitalizations access and quality as well as costs. This indicator often can be prevented by careful hospital discharge set also serves as symptoms of failures to get the and transition care. Monitoring patients for signs right care, gaps in access to care, and/or poor of decline and stepping up care when needed can coordination of care during transitions. Thus, low also help to avoid complications. rates on these potentially preventable hospital use indicators point to systems that may be achieving • There is a twofold spread across states in the better outcomes through more effective care man- rates of potentially preventable admission for agement in primary care practices and more timely ambulatory care sensitive conditions among access, as well as through basic preventive and Medicare beneficiaries, ranging from more than public health efforts to prevent chronic disease. 10,000 per 100,000 beneficiaries in the highest- Overall, states in the Upper Midwest, Southwest, rate states (all in the South) to less than 5,000 in and Pacific Northwest rank high on the this the lowest-rate states (Hawaii, Utah, Washington, dimension, while many in the South and Northeast Alaska, and Oregon) (Exhibit 16). rank near the bottom (Exhibit 15). The five top- • Among 33 states that collect all-payer hospital ranked states are Utah, Oregon, Idaho, Hawaii, and data, admission rates for pediatric asthma varied New Mexico. Each of these states has relatively low from 55 per 100,000 children in Vermont to 314 rates of potentially preventable hospitalizations as well as relatively low annual Medicare spending per 100,000 in South Carolina—nearly six times per enrollee and average private health insurance higher. Likewise, the proportion of asthmatic premiums. Notably, of these five, only Hawaii is adults that used emergency care in a year varied a top-ranked state across all State Scorecard di- among 36 states, from 9 percent in Iowa to 29 mensions. Although comparisons are hampered percent in Mississippi. by missing data for some states, the top-ranked The extent of patient “churning” in and out of quartile of states generally performs well across hospitals also varies by state. Among Medicare most indicators of hospital use and costs (see beneficiaries, there is a twofold variation across Appendix Exhibits A7 and A8).12 states in rates of hospital readmission within 30 Interstate variations for many of the indicators in this dimension are among the widest of all State days. There is a threefold variation in rates of Scorecard indicators, suggesting that there are op- hospital admission and 90-day readmission among portunities to lower costs and improve patient care long-stay nursing home residents (Exhibit 17). by reducing complications that result in emergency • Nearly one of four Medicare patients (24%) room use or hospital readmissions. discharged from the hospital is readmitted within potentially avoidable use of hospitals 30 days in Louisiana and Nevada, compared with only 13 percent in Vermont and Wyoming, the Some hospital admissions, readmissions, and two states with the lowest rates. emergency visits for ambulatory care sensitive • Among long-stay nursing home residents, (ACS) conditions can be averted through effective hospital admissions range from a low of 8 management of chronic conditions such as asthma and diabetes and timely preventive care such as percent in Utah to a high of 25 percent in vaccinations against influenza and pneumonia. Louisiana. Readmission rates within 90 days Access to primary care after normal office hours for nursing home residents range from a low through links to community-based physicians and of less than 7 percent in Oregon to 18 percent care centers can also help to avoid hospitalizations in Mississippi. or emergency care. • Home health patients are admitted to the Among vulnerable populations, including hospital at even higher rates, with similar nursing home residents and home health care interstate variation. 30 If all states could achieve the level of the lowest As discussed in the Healthy Lives section rates of admissions for ACS conditions and lowest below, a better understanding of how the or- rates of hospital readmission, the cumulative effect ganization of care systems, population health would be to reduce these hospitalizations by 30 to policies, and underlying population health risks 47 percent and save Medicare $2 billion to $5 billion interact and affect cost and care outcomes for each year. Potential savings would be still greater if diabetes, asthma, and other conditions could similar reductions extended to all patients insured inform strategic efforts to improve. by private payers and Medicaid. costs of care The variations add up to substantially different outcomes in terms of access to and quality of The costs of health care also vary widely by state. care, as well as costs. Some states, such as Utah, Some factors beyond the direct control of policy- Oregon, Washington, and Idaho, have notably makers, such as prevailing wage rates, certainly low rates of potentially preventable hospital use, affect service costs and insurance premiums. suggesting that underlying health system factors However, other contributing factors are amenable provide efficiencies and quality of care. Indeed, to public and private policies, including the degree if these indicators were included in the quality to which the health system emphasizes primary dimension, each of these states would move up care, population health improvement, and care a quartile—rising 17 to 18 places in rank. coordination, and the extent to which payment 31 methods and incentives support and reward more ranged twofold among states in 2003, from a efficient care systems. low of about $4,500 in Hawaii to a high of just There is substantial variation in total costs of over $8,000 in New Jersey.13 care across states, as measured by Medicare annual • Analyses of the factors contributing to these costs per beneficiary, with no systematic relation- variations find a strong association between ship between the cost and quality of care. Some health care costs and the mix of primary and states with high rankings for access to care and high specialized care services as well as efficient quality have low relative costs. Iowa, Minnesota, hospital use.14 and Wisconsin, for example, have low annual • Moreover, analysis in the National Scorecard Medicare spending per person and are also among of Medicare beneficiary survival rates one year the leading states for access and quality. The wide after hospitalization for heart attack, colorectal variation in Medicare costs, adjusted for the health cancer, or hip fracture finds that cost and health and age mix of beneficiaries, points to opportuni- outcomes vary widely, with little association. ties for net national gains from provision of more Some regions of the country achieve superior efficient care. outcomes at lower costs and some regions have There is a close correlation between high overall both high costs and poor outcomes.15 costs and high rates of potentially preventable • The costs for private health insurance premiums hospital use—indicating that there are opportuni- for adults under 65 also vary across states, but ties to improve care experiences and lower costs by more narrowly than Medicare annual costs per focusing policies on symptoms of inefficiency. enrollee. In 2004, average annual premiums for individual employer-sponsored policies were • Medicare—which has a uniform benefit about $3,700. The cost of such premiums ranged structure and payment methodology across the from 13 percent below average, or $3,200, in the country—provides a good basis for comparing five lowest-cost states (Utah, Hawaii, Arkansas, health costs across states. Medicare fee-for- Georgia, and North Dakota) to 15 percent above service adjusted annual spending per enrollee average, or $4,200, in the highest-cost states 32 (Maine, Massachusetts, the District of Columbia, efficient c are systems Rhode Island, and Alaska). and total costs of care • Notably, the average cost of premiums for private The close association of Medicare costs and hospital group coverage varies much less than do state utilization suggests potential targets for strategies incomes. For example, median incomes in Maine to improve the efficiency of care. As illustrated and West Virginia are below the U.S. average, in Exhibit 18, states with high rates of “churning” yet the states have above-average premiums. in and out of hospitals have higher overall total Conversely, in several states with relatively annual costs of care for beneficiaries than states high median incomes, including Maryland, with lower rates of repeat hospitalizations. The Connecticut, Colorado, and Massachusetts, relationship between more intensive and potentially avoidable use of hospitals and high costs also holds the average cost of health insurance premiums for two other indicators: rates of admission for is near the national average. Available data ambulatory care sensitive conditions and rates of on private insurance premiums do not adjust hospital admission and readmission among nursing for patient cost-sharing or scope of benefits. home residents. The patterns suggest that effective Thus, adjustment for differences in the extent transitions across care settings and links between of coverage could reduce or widen variability in primary care and hospital-based providers could premium costs across states.16 improve efficiency of care and reduce costs.17 33 • States with the highest proportion of Medicare These findings suggest that efforts to improve primary patients readmitted to the hospital within 30 care and care transitions could improve population days (Maryland, Texas, Nevada, and Louisiana) health and achieve savings through more efficient are also among the 10 states with the highest use of specialized and expensive resources. per-beneficiary costs (Exhibit 18). • The 30-day readmission rate could be a key indicator of underlying care patterns that Equity increase costs. Average total Medicare costs A per year are 38 percent higher in the five states state’s health system should be judged by with the highest 30-day readmissions, compared how well it performs for its most vulner- with the five states with the lowest rates of able residents. Through programs such as readmissions ($7,200 vs. $5,200). Medicaid and SCHIP, all states devote consider- • Rates of admission and readmission to hospitals able resources to providing care for low-income among nursing home residents are relatively high residents and other vulnerable groups. Policy in Georgia, Kentucky, Texas, Oklahoma, New strategies such as raising eligibility thresholds Jersey, West Virginia, Arkansas, Louisiana, and for public coverage and eliminating barriers to Mississippi. Most of these states have relatively enrollment and retention can contribute substan- higher costs overall, and fall in the bottom half tially to improved access to care for such groups. of the ranking on Medicare costs. Building health system capacity and promoting 34 quality of care through safety net providers can further reduce disparities in access and quality between the insured and uninsured. The State Scorecard assesses equity by comparing gaps in performance among subgroups of patients by income level, insurance coverage, and race/ethnicity. The analysis compares performance levels among each state’s most vulnerable populations to the national average for selected scorecard indicators for which data are available. States ranked at the top of the equity dimension overall tend to have the smallest gaps in performance between national averages and low-income, uninsured, and minority groups (Exhibit 19). Five New England states— Massachusetts, Maine, Vermont, Rhode Island, and New Hampshire—score in the top quartile on this dimension for all three vulnerable population groups. Conversely, 10 of the 13 states in the bottom quartile of the overall equity ranking are also in the bottom quartile for at least two of the three subgroups (race/ethnicity, income level, and insurance coverage). Six states are in the bottom quartile for all three subgroups. The lowest-ranking states on the equity dimension are in the South and West. Yet other states in these regions, including Alaska, Montana, and West Virginia, rank in the top half of the equity rating. This pattern suggests that states facing similar regional circumstances and challenges can still effectively tackle dis- parities in care. There are wide equity gaps in State Scorecard measures for vulnerable popula- tions, with the extent of disparities varying across the states. States that perform well in general on overall statewide rankings tend to have smaller equity gaps among vulnerable populations. Some high-performing states provide care to traditionally disadvantaged groups that, by some indicators, is better than the national average. For example, the percent of low-income diabetics receiving basic recommended services was higher in Minnesota (63%) and North Dakota (60%) than the average among all diabetics across 35 the nation (39%). States with large gaps in asthma recommended preventive services, compared care might learn lessons from the ambulatory care with 59 percent of insured and 54 percent of management strategies in these four states. higher-income adults. A similar pattern exists Conversely, in states that rank low on overall among diabetics. On average, 67 percent of performance across all five dimensions, low per- low-income diabetics did not receive basic care formance extends even to high-income, insured, according to guidelines for their condition. and non-minority groups. The extent to which children have a medical The following section examines gaps in terms home also depends on their family’s income of access to and quality of care, focusing on dis- and their insurance status. Top-ranked states on parities by income level and insurance status. The Healthy Lives section, below, examines how equity generally performed well for all children, well state health systems support their residents’ including those in low-income families or without ability to live long and healthy lives and explores health insurance (Exhibit 21). disparities by race or ethnicity. In most states, variation on many indicators is much greater among uninsured than insured income and insurance populations. For instance: In most states, the quality of care varies by • The proportion of insured adults who reported income and insurance, with lower income and lack of insurance linked to lower quality. But such not seeing a doctor because of cost was under gaps are widest in states that perform poorly on 14 percent in all states. Among the uninsured, indicators of quality and access overall. Gaps are the proportion reporting this ranged from a particularly wide in terms of receipt of preventive low of about one of four uninsured residents care (Exhibit 20). On average across the nation, in North Dakota and Hawaii to a high of 52 78 percent of uninsured and 71 percent of low- percent in the five states with the largest gap for income adults age 50 and older did not receive this indicator. 36 • Across the nation, on average only 14 percent lacking medical homes were 14 percentage points of adults with insurance coverage reported higher among black children and 23 percentage not having a usual source of care. Among the points higher among Hispanic children than white uninsured, proportions without a usual source children (Exhibit 22). Minority adults, too, are at of care ranged from 38 percent in the states with great risk of missing recommended preventive care. the smallest disparities to 70 percent in the states In some states, as many 75 percent to 80 percent of with the largest disparities. black and Hispanic adults age 50 and over did not access and quality: receive all preventive care recommended for this race and ethnicity age group, including cancer screening. The gaps were generally widest in states with the highest The State Scorecard compares health care access uninsured rates. and quality experiences by racial and ethnic Some states ranked low on measures of equitable groups, focusing on states with substantial care for racial/ethnic minorities as a result of large minority populations. Because minorities often shortfalls for selected minority groups that comprise have lower incomes and are more likely to be relatively small shares of their total populations. For uninsured than whites, the disparities observed example, Minnesota’s scores were often low for a among minorities also reflect differences related group that included Asian Americans and Native to income and insurance status. Americans. For these states, improvement efforts Across states, equity gaps vary by minority focused on these groups could substantially reduce group. Hispanics tend to have the highest uninsured health disparities. rates and are the least likely to report a regular This analysis of racial and ethnic disparities source of care among U.S. race/ethnic population focuses on subgroups for which there were sufficient groups. Both black and Hispanic children are at high data for comparisons. As a result, small states with risk of lacking a medical home: rates of children relatively homogeneous populations, such as Maine, 37 Vermont, and Wyoming, often had few subgroups Healthy Lives for ranking. A However, the absence of race/ethnicity data for n overarching goal of the health care some states appears to have little impact on equity system is to contribute to long and healthy rankings. Overall, the rankings for racial and ethnic lives. This can be accomplished through disparities closely follow rankings observed in the public health initiatives, preventive care, care for sickness or injury, management of chronic income and insurance analyses. States in which conditions, and compassionate care at the end of low-income and uninsured groups fared better life. Following the National Scorecard, the State also tended to have the smallest gaps for minority Scorecard assesses how well states support their subgroups. As a result, the equity rankings remain residents’ healthy lives through indicators of similar regardless of whether racial and ethnic mortality amenable to health care and health- disparities are considered. related limitations faced by adults. The analysis Using the national average is only one possible found a wide range of health outcomes across benchmark with which to assess equity. In separate states on multiple indicators. Improving health analyses, we also assessed equity by comparing outcomes is a challenge for health care and public health systems, as states grapple with underlying experiences among low-income, uninsured, and population risks such as rising rates of obesity or racial/ethnic minorities to experiences among high levels of poverty that put individuals’ health their counterparts—higher-income, insured, and and quality of life at risk. white populations within each state. With a few No indicators are currently available across states exceptions, this alternative method yielded results that measure the quality of life from conditions similar to the equity rankings using the national amenable to health care, rates of chronic disease average as the benchmark.18 under control, or the ability to participate in work 38 or community life as a result of timely, appropriate related behaviors, poverty, and environmental and care for potentially disabling conditions. Yet, rates workplace hazards. Education levels and cultural of chronic disease have been rising among adults beliefs influence health outcomes and patients’ and children across the United States, necessitating interactions with the health system. While the public health and health care system responses. pathways through which individuals achieve Notably, three conditions—heart disease, diabetes, optimal health are complex, measures of health and cancer—account for most of the variation outcomes provide targets for improvement.19 among states in rates of mortality amenable to care. Overall, states in the Upper Midwest, Mountain States are increasingly looking to policy initiatives region, and California had the highest average to reverse rising rates of obesity, reduce smoking, rankings on the health outcome measures included and promote earlier detection of breast and colon in the scorecard (Exhibit 23). New Hampshire also cancer. Actions taken now to address risks to ranked among the top quartile of states. population health and provide timely, effective potentially preventable mor tality health care are instrumental to improving health outcomes in the future. Among the measures in this dimension, mortality Health system performance is one of many forces amenable to health care represents the best that shape health status and longevity. Whether overall summary indicator of health outcome people live long and healthy lives depends on variations among states. This measure includes many factors, including family history, health- age-standardized death rates before age 75 from 39 conditions for which timely and effective medical • The gaps translate into thousands of lives. If all care can potentially delay or prevent mortality.20 states improved to levels achieved by the best Internationally, the United States performs poorly state (Minnesota, with 70.2 deaths per 100,000), on aggregate mortality amenable to health care, about 90,000 fewer premature deaths would ranking 15th out of 19 nations (including 18 occur each year. European countries) as of 1998.21 Wide differences exist between mortality rates New analyses prepared for the State Scorecard for conditions amenable to care for black and white reveal startlingly wide variation in potentially populations (Exhibit 25). In at least half the states, preventable death rates among states (Exhibit 24). rates of age-standardized mortality amenable to health care among blacks were twice the rates • There is a twofold range across states in the rate among whites (median rates across states were of deaths amenable to health care. In the leading, 184 deaths per 100,000 blacks, compared with lowest-rate states (Minnesota, Utah, Vermont, 89 deaths per 100,000 whites). The gap between Wyoming, and Alaska), death rates were half the black and white populations narrows but remains rates in the District of Columbia and the states substantial in the five states with the smallest equity at the bottom of the distribution (Tennessee, gap (Hawaii, Oregon, New Mexico, Washington, Arkansas, Louisiana, and Mississippi). Average and Massachusetts): 123 deaths per 100,000 blacks, death rates were 74.1 per 100,000 persons in the compared with 84 per 100,000 whites. top five states, compared with 141.7 per 100,000 In the District of Columbia, reflecting its persons in the bottom five states. population mix, the very high black mortality rate • States in the Northwest, Upper Midwest, and pulls up the average rate (see Appendix Exhibit New England generally had the lowest rates of A11). As a result, it has the highest average amenable mortality amenable to health care and states in death rate in the country in addition to the widest the South had the highest. white and black mortality disparity. After the 40 District of Columbia, southern states and some and 24.6 per 100,000 in the District of Columbia states in the Midwest with large black populations (see Appendix Exhibit A10). States can promote have the greatest gaps in mortality amenable to lower mortality for these conditions by ensuring health care, with more than 100 additional deaths access to early detection services and providing per 100,000 black residents in excess of the overall timely, effective, well-coordinated treatment. national average rate. For the most part, measures in the health Notably, potentially preventable mortality rates outcomes dimension, including disability rates, for whites also varied significantly across states, follow similar geographic patterns. States in the ranging from a low of 67.6 per 100,000 population West and Upper Midwest tend to rank highest (Minnesota) to a high of 118.3 (West Virginia). In (i.e., have better health outcomes), while the general, white rates were highest in states with southern states fare the worst. There is also a high overall rates. pattern of worse health outcomes among some northern industrial states, including Michigan, regional patterns Ohio, and Pennsylvania. Like the measure of potentially preventable The clustering of poor health outcomes in mortality overall, rates of mortality from breast states with high poverty rates, such as Louisiana, cancer and colon cancer are related to health system Mississippi, and West Virginia, places a heavy performance and follow geographic patterns. Age- burden of illness on their populations and raises standardized rates of death from breast cancer in serious challenges for care systems and public the state with the lowest rate (Hawaii, with 16.2 health policies. For example, there is more than per 100,000 women) are half that of the rate in twofold variation in infant mortality (from a low of Louisiana (29.7 per 100,000) and the District of 4.3 deaths per 1,000 live births in Maine to a high of Columbia (34.1 per 100,000). Variation among 10 per 1,000 live births in Louisiana and Mississippi colon cancer mortality rates (age-standardized) is and 11 per 1,000 in the District of Columbia). Infant nearly as wide, with a low of 15.3 per 100,000 people mortality rates tend to be highest among African in Utah to a high of 23.9 per 100,000 in Kentucky American families across states. 41 Population characteristics, including rates of • Leading states consistently outperform lagging poverty and chronic disease and risk factors such states across multiple indicators and dimensions. as smoking and obesity, contribute to variations in The patterns indicate that policies and health state health outcomes. States vary widely in terms system variations make a difference. of the percent of poor and low-income residents, as • Better access to care is closely associated with well as the incidence of cancer and other population better quality of care across states. health risk factors (see Appendix Exhibit A12). • Higher-quality care is not systematically Heart disease, diabetes, asthma, and cancer rates are associated with higher costs. Cost variations point particularly high among low-income populations to significant opportunities to reduce costs as well and in impoverished geographic communities. As a as improve access to and quality of care. result, states with high rates of poverty and income • All states have substantial room to improve. inequality tend to also have higher rates of mortality h e a l t h c a r e a c c e s s , q u a l i t y, from conditions amenable to health care. cost, and efficienc y var y widely Yet, while underlying poverty levels and across the united states. demographics matter, strategic state policies— including public health initiatives—can make a Currently, where one lives in the United States difference. Rhode Island, for example, reduced affects access to care, the quality of care received, infant mortality rates among low-income families and the cost of that care. There is often a two- to by providing a combination of timely access to threefold or greater range in performance across care and coverage of family planning services, plus key indicators of access, quality, and potentially counseling and a public health approach to support avoidable hospital use. These wide variations in healthy births.22 New statewide initiatives, such State Scorecard indicators mirror those found in as one under way in Arkansas, are tackling health other studies, indicating that where one lives affects risks related to obesity and sedentary lifestyles with the amount and kind of health care one receives. programs to provide healthy foods in schools and Improving state health system performance to regular exercise.23 levels achieved by the leading states could bring Failing to provide access to appropriate care higher-quality care to millions of Americans and carries particularly high risks among poor help to prevent thousands of premature deaths. As populations. For example, infant mortality rates discussed in the Impact section, below, there are in Mississippi—already among the highest in the also billions of dollars at stake from potential gains nation—jumped in 2005, following cutbacks to through more efficient use of hospitals, stronger the Medicaid program and community health primary and preventive care, and more effective clinics. Infant mortality rates were highest in management of chronic disease. communities that had little access to community leading states consistently clinics or Medicaid coverage for prenatal counseling outper form lagging states on or family counseling.24 multiple indicators and dimensions. Thirteen states—Hawaii, Iowa, New Hampshire, Cross-Cutting Findings Vermont, Maine, Rhode Island, Connecticut, T Massachusetts, Wisconsin, South Dakota, he State Scorecard indicates that, by Minnesota, Nebraska, and North Dakota—emerge aiming higher, we can do much better as at the top of the overall performance ranking a nation and respond to the increasingly (Exhibit 2). These states generally ranked high on urgent need for action. Overall, five cross-cutting multiple indicators in each of the five dimensions findings emerge. assessed by the State Scorecard. As illustrated in • There is wide variation among states that attests to Exhibit 2, states in the top quartile of one dimension the potential for the country to do much better. were often in the top quartile or top half of the 42 distribution in all dimensions. Many of these states better access is associated with have been leaders in reforming and improving their better quality across states. health systems and have among the lowest rates of Across the country, the same states consistently uninsured residents in the nation. rank low or high on indicators of both health Conversely, the 13 states at the bottom of the care access and quality (Exhibit 3). Four of the overall performance ranking—California, Tennessee, five leading states in terms of access to care— Alabama, Georgia, Florida, West Virginia, Kentucky, Massachusetts, Iowa, Rhode Island, and Maine— Louisiana, Nevada, Arkansas, Texas, Mississippi, and also rank among the top five states in terms of Oklahoma—tend to lag well behind their peers on quality. Moreover, states with low performance multiple indicators across multiple dimensions and on quality indicators tend to have high rates of have high rates of uninsured residents. The results uninsured. This pattern across states points to pull these states down to the bottom quartile of the the importance of affordable access to care as national distribution. an important first step for obtaining essential Health system performance often varies health care, and a prerequisite for care that is by region. Across all dimensions, states in the effective, well coordinated, and patient-centered. Upper Midwest and Northeast often rank in the In states where more people are insured, adults highest quartile, with those in the lowest quartile and children are more likely to have a medical concentrated in the South. Within regions, some home and receive recommended preventive and states perform relatively well compared with chronic care. Identifying care systems, as well neighboring states and national leaders, with rates as state policies that support superior access varying by dimension and indicator. Colorado, and quality and lower costs, will be critical to Montana, Utah, and Washington, for example, improving systemwide performance. do better than other western states, while North The proportion of uninsured children has Carolina and Virginia outrank other southern states. declined following federal and state action to These findings suggest that regional forces alone do expand coverage to low-income children. Yet, not determine performance and that benchmarking the proportion of uninsured working-age adults within regions as well as across states could provide across the nation is high and rising, jeopardizing insights as well as opportunities for collaboration the health of millions of working adults and putting and improvement. states and the nation at risk as we lose access and financial security for the nation’s workforce. A Understanding how particular policies and healthy economy and society require a healthy, health system attributes shape these patterns of productive workforce. health care access, quality, hospital use, and costs could inform national and state policy. Leading there are oppor tunities to reduce states may provide models for other states. For health care costs, as well as to example, in 1974, Hawaii became the first state to i m p r o v e a c c e s s t o c a r e a n d q u a l i t y. enact legislation requiring employers to provide Annual health care costs vary widely across health insurance to full-time workers; it now states, with no systematic relationship between ranks first on access to care. For the past decade, costs, levels of insurance coverage, or residents’ Rhode Island has provided incentive payments to ability to pay as measured by state median or Medicaid managed care plans that reach quality average incomes. Moreover, there is no systematic targets; it now ranks first on quality. Maine, relationship between costs and measures of health Massachusetts, and Vermont lead on equity, and care quality. In the State Scorecard analysis, states are recognized for innovation and leadership with the highest medical care costs during the year on expanding health insurance coverage and also tend to have the highest rates of potentially benchmarking for quality. preventable hospital use, including high rates of 43 hospital readmission within 30 days and high Impact of Improved Performance rates of admission for complications of diabetes, T asthma, and other chronic conditions. High rates here are many ways to improve perfor- of readmission and admission for chronic disease mance, involving stakeholders at all levels provide evidence of access and quality problems of the health care system. This section il- (i.e, not getting the right care), as well as missed lustrates the potential gains in terms of healthy opportunities to prevent the onset of disease lives, access, and dollars if all states were able to complications. Reducing the use of high-cost, meet the levels of performance achieved by top specialized care by preventing complications has the states for selected indicators. It concludes with a potential to provide net gains for all states—better discussion of policy implications for federal and care outcomes at lower costs. state governments. all states have room to improve. aiming higher: impac t of improving per formance Despite clear differences in performance among the states, the State Scorecard points to substantial Exhibit 26 shows the estimated impact if all states room for improvement in every state. No single were to improve their performance to the rate state or group of states performs at the top of the of the best-performing state for 11 key scorecard range on all indicators or dimensions. The five to indicators.26 If all states could approach the low 10 top-ranked states, for instance, each had some levels of mortality from conditions amenable to health care achieved by the top state in 2002, indicators in the bottom quartile of performance nearly 90,000 fewer deaths before the age of 75 rankings (see Appendix Exhibit A1). Moreover, on would occur annually. There also could be po- some measures of quality, even the top rate falls tentially fewer disease complications and activity well below recommended care and levels known to limitations through improved access and timely be achievable by top-performing delivery systems delivery of care. that provide accessible, well-organized, patient- The nation would cover 22 million more adults centered care. and children if all states’ coverage rates reached These findings indicate the need to improve those of the top states, reducing the numbers of performance in all states while narrowing the uninsured by half. If adults age 50 and older or variation in performance across states. Among diabetics in all states receive preventive care at hospitals and managed care plans across the nation, the rates achieved in the top states, almost nine this goal has been achieved for selected quality million older adults would receive recommended indicators, such as heart attack treatment in the preventive care, including cancer screenings, and hospital.25 There are gaps in performance between almost four million diabetics would receive basic even the top-ranked states and levels known to recommended care. In addition, 33 million adults be achievable. It is therefore crucial to identify and children would have a usual source to provide successful strategies and emulate exemplary primary care and help coordinate care. results achieved at the local level by organizations, The Medicare program could potentially save $2 providers, and communities. billion to $5 billion a year by reducing potentially The State Scorecard findings point to opportunities preventable hospitalizations for chronically ill to improve health system performance by learning from state and regional variations. The following Exhibit 26 analysis of potential gains if states achieved section explores the potential gains in terms of the rate of the top state on the 11 indicators is also healthy lives, enhanced access, and reduced costs— available for each state on the Commonwealth gains in overall greater value—if all states were able Fund Web site at www.commonwealthfund.org. The table is available for online viewing and also to raise their performance to levels achieved by the to download along with state-specific profiles. top states on key indicators. 44 EXHIBIT 26 National Cumulative Impact if All States Achieved Top State Rates If all states improved their performance to the Indicator level of the best-performing state for this indicator, then: Insured Adults 17,207,746 more adults (ages 18–64) would be covered by health insurance (public or private), and therefore would be more likely to receive health care when needed. Insured Children 4,391,891 more children (ages 0–17) would be covered by health insurance (public or private), and therefore would be more likely to receive health care when needed. Adult Preventive Care 8,587,664 more adults (age 50 and older) would receive recommended preventive care, such as colon cancer screenings, mammograms, pap smears, and flu shots at appropriate ages. Diabetes Care 3,611,284 more adults (age 18 and older) with diabetes would receive three recommended services (eye exam, foot exam, and hemoglobin A1c test) to help prevent or delay disease complications. Childhood Vaccinations 756,942 more children (ages 19–35 months) would be up-to-date on all recommended doses of five key vaccines. Adults with a Usual Source of Care 22,071,293 more adults (age 18 and older) would have a usual source of care to help ensure that care is coordinated and accessible when needed. Children with a Medical Home 10,858,812 more children (ages 0–17) would have a medical home to help ensure that care is coordinated and accessible when needed. Preventable Hospital Admissions 981,775 fewer hospitalizations for ambulatory care sensitive conditions would occur among Medicare beneficiaries (age 65 and older) and $5.0 billion dollars would be saved from the reduction in hospitalizations. Hospital Readmissions 197,798 fewer hospital readmissions would occur among Medicare beneficiaries (age 65 and older) and $2.3 billion dollars would be saved from the reduction in readmissions. Hospitalization of Nursing Home 125,024 fewer long-stay nursing home residents would be hospitalized and Residents $1.2 billion dollars would be saved from the reduction in hospitalizations. Mortality Amenable to Health Care 88,780 fewer premature deaths (before age 75) might occur from causes that are potentially treatable or preventable with timely and appropriate health care. SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 Medicare patients or by reducing the number of across states over the course of several years, the readmissions by improving transition care. These numbers add up to substantial gains in value for savings would be even greater if these improvements the nation. extended to all patients. Over $1 billion dollars could potentially be saved by providing the standard of care for frail nursing home residents reached in the Moving Forward: The Need for best-performing state. Savings would be contingent Action to Improve Performance on identification of effective interventions, and T some savings might be offset by the costs of the he overall picture that emerges from the intervention. More generally, the nation would State Scorecard is that there is potential save $22 billion to $38 billion per year if higher- for improvement on all key dimensions of cost states achieved access, care, and efficiency performance. Our national values emphasize that improvements sufficient to bring costs down to the we are one nation, yet where adults and children national median or rates achieved by the lowest- live affects their access to care, care quality, care cost quartile of states. experiences, and the affordability of care. The These examples illustrate only a few of the many view across states reveals startlingly wide gaps important opportunities for improvement. Because between leading and lagging states on multiple some indicators would affect the same individuals, indicators. Gaps between actual and achievable some of these numbers cannot be combined. Yet, levels of performance represent illnesses that could 45 have been prevented or better managed, as well as and small employers—reaching beyond narrow dollars that could have been saved or reinvested boundaries that divide public and private to improve population health. The variation in insurance coverage to connect the workforce multiple dimensions provides compelling evidence with sources of continuous, affordable insurance. of the need for coherent, concerted action to aim Maine and New York, for example, have both for improved health system performance across succeeded in lowering the percent of adults all key dimensions. Benchmark levels achieved by uninsured over the past five years through top-performing states are within the reach of all creative strategies that offer publicly sponsored states. Moreover, initiatives by the top performers options with premium assistance and in the case or by models of excellence within states set the of New York, provide reinsurance for small, low- wage firms. The combination of public coverage pace for change. expansions to low-income workers and families The State Scorecard points to the need for action and innovative public–private group health in the following key areas: insurance options offers a potential foundation • Universal coverage: Moving toward universal to build on for the future.27 coverage is critical for improving quality and Federal action as well as state initiatives will be delivering more cost-effective care, as well as essential for substantial progress nationwide. The ensuring access to care. contrasting insurance coverage trends between • Better information to assess performance and adults and children over the last five years are identify benchmarks: It takes information a testament to the potential of, and need for, to guide and drive change. We need more federal action to stimulate and support state sophisticated information systems and better efforts. Federal support of Medicaid expansion information on practices and policies that for children, followed by creation of the State Children’s Health Insurance Program, sparked underlie high or varying performance. broad expansion across states in children’s • Analyses to determine key factors that contribute coverage —reversing the declines that followed to improved outcomes and performance: States the erosion of private coverage in the 1990’s. and public and private delivery systems can use such information to develop evidence-based wide variations point to strategies to improve. opportunities to learn • National leadership and collaboration across It is important to understand the key factors public and private sectors: Working together that contribute to improved outcomes and per- toward shared goals is essential for coherent, formance. States and public and private delivery strategic, and effective improvement efforts. systems can then develop evidence-based strate- universal coverage with gies to improve. States can look to each other as meaningful access: foundation well as to models of excellence within their own for quality and efficient care borders for evidence of effective policies. On many scorecard indicators, a few states outper- Universal coverage that provides meaningful form all states or nearby states that have similar access to essential care and financial protection economic and demographic conditions, providing is the critical foundation upon which to improve important examples of achievable targets. Un- quality and enable more cost-effective care. States derstanding how policy variations and features that have achieved the highest rates of coverage of underlying care delivery systems—including for adults and children consistently have higher primary care, specialized care, hospital care, rates of preventive care, care for chronic disease, and long-term care, health insurance coverage, and continuity of care. and provider payment incentives—contribute to Access variations across states also indicate systemic performance variations could inform the need to expand coverage to low-wage workers efforts to improve. 46 Benchmarks can help policymakers and other to assess, compare, and improve care. Informa- stakeholders set priorities and focus approaches to tion systems have the capacity to improve health achieve higher performance. All states have room outcomes and safety, reduce duplication, and focus for improvement. The State Scorecard suggests that care on patients and outcomes. This capacity is strategic approaches focusing on key performance vital for improving outcomes for the 10 percent of gaps can yield significant improvement. patients with multiple chronic conditions or serious information systems and illnesses that account for two-thirds of all health better information are care expenditures each year. Accelerating the pace critical for improvement of adoption and spread of electronic information technology with the capacity for exchange will Comparative information on health outcomes enable public and private policy leaders to identify and the costs and quality of care are essential gaps, set targets, and develop payment systems to move toward more efficient, higher-value to reward and support higher-value, accountable care systems. The State Scorecard highlights the care systems.28 need for better sources of data and collaborative It takes information to guide and drive change. efforts at the national and state levels to develop information systems capable of supporting more State as well as federal government actions are appropriate, integrated, and coordinated care. needed to move forward. Performance data are increasingly becoming national leadership and public available, thanks to federal, state, private, and and private collaborative public–private efforts, including the federal improvement initiatives Agency for Healthcare Research and Quality and the Centers for Medicare and Medicaid Services States play key roles in the health system as regula- initiatives to develop hospital quality and safety tors, insurers, and sources of financing for care. They indicators. Yet, major gaps remain. are also one group of several key stakeholders with a vital interest in more accessible, higher-quality, and • There are no indicators of patient safety, such as more efficient health systems. Currently, in some hospital infection rates or adverse drug events, states, collaborative efforts involving both public across all states. and private leaders are beginning to build a founda- • Cross-state or regional indicators of potential tion for improved performance. For example: overuse or duplication of services also are not available. Nor are there insurance administrative • The state of Wisconsin is contributing data and costs by state. funding to the Wisconsin Health Information • Only 33 states collect and report multi-payer Organization, a coalition of employer groups, hospital data, and of these states only a handful health plans, and health care providers that is can identify rates of readmissions. creating a repository of health insurance data • The Medicare program is often the only viable with the goal of reporting on the cost and, data source for national and regional analyses of eventually, the quality of care in the state.29 care patterns and patient experiences. Although • Community Care of North Carolina is a Medicare provides a critical view on geographic collaboration among the state, counties, hospitals, patterns of health system performance, and physicians to create accountable, community- comparable information for the under-65 based systems of care that increase access to population is typically not available. primary care and enhance care management Moreover, states and the nation lack advanced for Medicaid-insured patients.30 Also, the Fund’s electronic information systems with the capacity Assuring Better Child Health and Development for exchange across sites of care and the ability to initiative is integrating developmental screening provide doctors and other providers with tools into well-child care visits.31 47 • Value-based purchasing collaborations led by or involving state governments in Massachusetts, Minnesota, Washington, and Wisconsin are seeking to improve the quality and efficiency of care by establishing uniform quality measures, promoting transparency, and adopting incentives for improved performance.32 • In another approach, hospitals and clinicians in northern New England states (Maine, New Hampshire, and Vermont) have collaborated for many years to benchmark heart surgery outcomes and share best practices, achieving results comparable to those of public reporting initiatives.33 And Massachusetts Health Quality Partners—a coalition of physicians, hospitals, health plans, government agencies, purchasers and consumers—is providing comparative data on clinical quality and patient ratings to stimulate and inform statewide improvement in the quality of health care.34 there is an urgent need for ac tion that takes a whole - population perspec tive The aging of the U.S. population, technological advances, and rising rates of chronic disease place upward pressure on health costs, which are already rising faster than incomes and straining family, business, state, and federal budgets. Access to care is deteriorating amid startling evidence of variable quality and inefficient care. Leadership at the state and national level is urgently needed to bring together all stakeholders—including private employers, insurers, health care providers, house- holds, and federal and state governments—for concerted action. Moving forward is a matter of great urgency: all states across the nation have much to gain from aiming higher. All states can do better, and all should continually ask, “Why not the best?” 48 Notes 1 Commonwealth Fund Commission on a High Performance 8 The medical home indicator is scored to reflect the American Health System, Why Not the Best? Results from a National Academy of Pediatrics definition of a medical home. See Scorecard on U.S. Health System Performance (New York: The Medical Home Initiatives for Children with Special Needs Commonwealth Fund, Sept. 2006); and C. Schoen, K. Davis, S. Project Advisory Committee, American Academy of Pediatrics, How, and S. C. Schoenbaum, “U.S. Health System Performance: “The Medical Home,” Pediatrics, July 2002 110(1 Pt. 1):184–86. A National Scorecard,” Health Affairs Web Exclusive (Sept. 20, See also S. J. Blumberg, L. Olson, M. R. Frankel et al, Design 2006):w457–w475. and Operation of the National Survey of Children’s Health, 2003 (Atlanta, Ga.: National Center for Health Statistics, 2005). 2 S. F. Jencks, T. Cuerdon, D. R. Burwen et al., “Quality of Medical Care Delivered to Medicare Beneficiaries: A Profile at State and 9 S. Leatherman and D. McCarthy, Quality of Health Care in the National Levels,” Journal of the American Medical Association, United States: A Chartbook (New York: The Commonwealth Oct. 4, 2000 284(13):1670–76; and S. F. Jencks, E. D. Huff, and T. Fund, Apr. 2002). Cuerdon, “Change in the Quality of Care Delivered to Medicare 1 0K. Davis, C. Schoen, S. Guterman, T. Shih, S. C. Schoenbaum, and Beneficiaries, 1998–1999 to 2000–2001,” Journal of the American I. Weinbaum, Slowing the Growth of U.S. Health Care Expenditures: Medical Association, Jan. 15, 2003 289(3):305–12. What Are the Options? (New York: The Commonwealth Fund, 3 Henry J. Kaiser Family Foundation/Health Research and Jan. 2007); C. Angrisano, D. Farrell, B. Kocher et al., Accounting Educational Trust, Employer Health Benefits 2006 Annual for the Cost of Health Care in the United States (San Francisco: Survey. McKinsey Global Institute, Jan. 2007); and E. S. Fisher, D. O. Staiger, J. P. W. Bynum et al., “Creating Accountable Care 4 These data are the most recent state data currently available. Organizations: The Extended Hospital Medical Staff,” Health The U.S. Census department recently announced it will Affairs Web Exclusive (Dec. 5, 2006):w44–w57. be reissuing insurance data and decreasing the national uninsured count by about 1.8 million. The department noted 1 1 S. Leatherman, D. Berwick, D. Iles et al., “The Business Case for the trends remain up. Adjusted state data and trends are not Quality: Case Studies and an Analysis,” Health Affairs, Mar./Apr. yet available. 2003 22(2):17–30. 5 D. C. Ross, L. Cox, and C. Marks, Resuming the Path to Health 1 2 Although some indicators were missing for several states, Coverage for Children and Parents: A 50-State Update on Eligibility state efficiency rankings remain basically the same if ranked Rules, Enrollment and Renewal Procedures, and Cost-Sharing on only data available in all states. Appendix tables indicate Practices in Medicaid and SCHIP in 2006 (Washington D.C.: Kaiser which states do not currently report to the federal–state all- Commission on Medicaid and the Uninsured, Jan. 2007). payer hospital data system known as Healthcare Cost and Utilization Project (HCUP) or are missing questions on the 6 The hospital quality composite is based on an “opportunities” annual Behavioral Risk Factor Surveillance System (BRFSS) model, which measures the percentage of appropriate care that population surveys. is delivered to patients in aggregate. In contrast, ambulatory care composites are based on an “all-or-nothing” model 1 3 Medicare annual reimbursement rates come from the that measures the percentage of patients who received all Dartmouth Atlas. The state level costs were indirectly adjusted recommended services. Hospital performance has improved for sex, race and age, and were further adjusted for illness since these data were compiled for the State Scorecard. and regional differences in price. See indicator definition in appendix. 7 For an example of how one institution achieved high performance across all three conditions, see S. Crute, “Case 1 4 E. S. Fisher, D. E. Wennberg, T. A. Stukel et al., “The Implications Study: Achieving High-Quality Care at Reid Hospital & Health of Regional Variations in Medicare Spending. Part 1: The Care Services,” Quality Matters, The Commonwealth Fund, Content, Quality, and Accessibility of Care,” Annals of Internal Dec. 2005. Medicine, Feb. 18, 2003 138(4):273–87; and K. Baicker and A. Chandra, “Medicare Spending, the Physician Workforce, and Beneficiaries’ Quality of Care,” Health Affairs Web Exclusive (Apr. 7, 2004):w4-184–w4-197. 5 For cost and quality variations by hospital regions, see Why 1 Not the Best? Results from a National Scorecard on U.S. Health System Performance, p. 25. 49 1 6For analysis of variation in the value of insurance—the scope of 7J. Rosenthal and C. Pernice, Dirigo Health Reform Act: Addressing 2 benefits and protection they offer—see J. Gabel, R. McDevitt, Health Care Costs, Quality, and Access in Maine (New York: The L. Gandolfo et al., “Generosity and Adjusted Premiums in Job- Commonwealth Fund, June 2004); and J. Rosenthal and C. Based Insurance: Hawaii Is Up, Wyoming Is Down,” Health Pernice, Designing Maine’s DirigoChoice Benefit Plan (New York: Affairs, May/June 2006 25(3):832–43. The Commonwealth Fund, Dec. 2004). 1 7 For an illustration of an intervention that reduced hospital 8Medicare Payment Advisory Commission, Report to Congress, 2 readmissions through more effective care transitions, see E. Assessing Alternatives to the Sustainable Growth Rate System, Coleman, C. Parry, S. Chalmers et al., “The Care Transitions Chapter 3: Using Medicare’s Physician and Other Payment Systems Intervention: Results of a Randomized Controlled Trial,” to Improve Value (Washington, D.C.: MedPAC, Mar. 2007). Archives of Internal Medicine, Sept. 25, 2006 166(17):1822–28. Also see D. McCarthy and C. Beck, “Using a Local Care Coordination 2 9S. Silow-Carroll and F. Pervez, “Wisconsin: Private and Network to Improve Patient Transitions,” Quality Matters, The Public Sectors Partner to Promote Transparency,” States in Commonwealth Fund, May/June 2007. Action: A Quarterly Look at Innovations in Health Policy (The Commonwealth Fund, vol. 5, July 2006). 1 8Results from the multiple equity analyses are available from the authors. 3 0“Improving Access to Primary Care: Community Care of North Carolina” (New York: The Commonwealth Fund, Aug. 2004); 1 9To assess the relative effects of health system and health L. A. Dobson, “Improving Medicaid Quality and Controlling outcomes on state rankings, we examined the effect of Costs by Building Community Systems of Care” (Raleigh, N.C.: excluding the healthy lives dimension from overall state Community Care of North Carolina, 2004). rankings. Although 14 states changed quartiles, all states in the top quartile remained in the top quartile (results are available 1 H. Pelletier and M. Abrams, ABCD: Lessons from a Four-State 3 from the authors). Consortium (New York: The Commonwealth Fund, Dec. 2003); and “Providing Developmental Services in Primary Care: The 2 0Appendix B1 describes mortality amenable death rates and North Carolina ABCD Project” (New York: The Commonwealth age ranges included in the analysis. Fund, Aug. 2004). 2 1 E. Nolte and M. McKee, “Measuring the Health of Nations: 3 2S. Silow-Carroll and T. Alteras, “Value-Driven Health Care Analysis of Mortality Amenable to Health Care,” British Medical Involving States and Public-Private Coalitions: Ahead of the Journal, Nov. 15, 2003, 327 (7424):1129–33. See also Schoen, Davis, Curve?” (New York: The Commonwealth Fund, forthcoming). How, Schoenbaum, “U.S. Health System Performance,” 2006. 3 3G. T. O’Connor, S. K. Plume, E. M. Olmstead et al., “A Regional 2S. Silow-Carroll, Building Quality into RIte Care: How Rhode Island 2 Intervention to Improve the Hospital Mortality Associated Is Improving Health Care for Its Low-Income Populations (New with Coronary Artery Bypass Graft Surgery. The Northern New York: The Commonwealth Fund, Jan. 2003). England Cardiovascular Disease Study Group,” Journal of the American Medical Association, Mar. 20, 1996 275(11):841–46. See 2 3For a description of the Arkansas Initiative, see the Healthy also D. McCarthy and S. Leatherman, “Improving Outcomes Arkansas Web site at www.arkansas.gov/ha/home.html. of Heart Bypass Surgery Through Regional Collaboration,” 4E. Eckholm, “In Turnabout, Infant Deaths Climb in South,” New 2 Performance Snapshots (New York: The Commonwealth Fund, York Times, Apr. 22, 2007. Dec. 2006). 5S. C. Schoenbaum, D. McCarthy, and C. Schoen, The Agency 2 3 4See www.mhqp.org for a description of Massachusetts Health for Healthcare Research and Quality’s 2006 National Healthcare Quality Partners. Quality Report (New York: The Commonwealth Fund, Mar. 2007); and S. C. Schoenbaum and A. L. Holmgren, The National Committee for Quality Assurance’s The State of Health Care Quality 2006 (New York: The Commonwealth Fund, Nov. 2006). 2 6See the State Performance Impact Calculator Methodology on the Commonwealth Fund Web site (www.commonwealthfund.org). 50 Appendices E x hibit A 1 Summary of Indicator Rankings by State E x hibit A 2 Access: Dimension and Indicator Ranking E x hibit A 3 Access: Dimension Ranking and Performance on Indicators E x hibit A 4 Quality: Dimension and Indicator Ranking on Getting the Right Care, Coordinated Care, and Patient-Centered Care E x hibit A 5 Quality: Dimension Ranking and Performance on Indicators E x hibit A 6 Hospital Quality Indicator Composite Percent and Rank: Hospitalized Patients Who Received Recommended Care for Acute Myocardial Infarction, Congestive Heart Failure, and Pneumonia E x hibit A 7 Avoidable Hospital Use and Costs: Dimension and Indicator Ranking E x hibit A 8 Avoidable Hospital Use and Costs: Dimension Ranking and Performance on Indicators E x hibit A 9 Healthy Lives: Dimension and Indicator Ranking E x hibit A 1 0 Healthy Lives: Dimension Ranking and Performance on Indicators E x hibit A 1 1 Mortality Amenable to Health Care by Race E x hibit A 1 2 State Demographics: Income and Health Status E x hibit A 1 3 State Demographics: Race and Ethnic Groups A ppendi x B 1 State Scorecard Indicator Descriptions and Data Sources A ppendi x B 2 Complete References for Data Sources 51 EXHIBIT A1 Summary of Indicator Rankings by State Overall No. of main Top 5 Top 2nd 3rd Bottom Bottom Rank* State indicators States Quartile Quartile Quartile Quartile 5 States 41 Alabama 30 0 4 9 9 8 0 26 Alaska 29 3 8 2 9 10 5 26 Arizona 32 3 8 7 7 10 5 48 Arkansas 30 1 2 7 2 19 10 39 California 32 3 6 5 7 14 6 22 Colorado 31 0 11 8 8 4 2 7 Connecticut 31 8 14 10 4 3 0 14 Delaware 31 3 9 7 10 5 0 32 District of Columbia 27 3 7 3 7 10 8 43 Florida 31 1 4 6 8 13 4 42 Georgia 32 1 2 7 11 12 1 1 Hawaii 29 11 16 6 3 4 2 30 Idaho 31 5 11 6 6 8 5 36 Illinois 30 1 2 10 11 7 1 38 Indiana 31 0 1 12 10 8 0 2 Iowa 32 11 17 11 3 1 0 20 Kansas 31 0 6 12 12 1 1 45 Kentucky 32 0 1 6 8 17 5 46 Louisiana 31 1 4 3 2 22 15 5 Maine 31 7 18 9 2 2 1 19 Maryland 32 2 7 7 13 5 2 8 Massachusetts 32 8 17 6 5 4 2 16 Michigan 30 1 5 14 4 7 0 11 Minnesota 32 6 15 10 5 2 0 50 Mississippi 31 0 2 3 6 20 16 37 Missouri 32 0 0 12 15 5 0 17 Montana 31 5 14 5 4 8 1 12 Nebraska 31 6 12 13 5 1 0 46 Nevada 31 0 2 6 4 19 11 3 New Hampshire 31 6 16 9 5 1 0 26 New Jersey 32 2 8 5 10 9 5 35 New Mexico 31 3 9 6 8 8 6 22 New York 32 0 5 7 12 8 3 30 North Carolina 32 0 3 13 11 5 1 13 North Dakota 30 8 12 6 9 3 1 24 Ohio 32 1 5 6 19 2 1 50 Oklahoma 31 0 1 5 6 19 9 34 Oregon 30 6 12 5 5 8 4 15 Pennsylvania 32 3 6 8 10 8 1 6 Rhode Island 32 9 19 6 4 3 1 33 South Carolina 31 0 3 8 14 6 2 10 South Dakota 31 6 15 10 3 3 0 40 Tennessee 31 0 0 13 11 7 3 49 Texas 32 0 0 8 7 17 7 24 Utah 32 9 12 4 6 10 5 3 Vermont 32 8 18 8 3 3 0 29 Virginia 32 1 2 13 12 5 1 17 Washington 32 5 11 11 4 6 1 44 West Virginia 31 0 4 6 8 13 5 9 Wisconsin 32 2 13 16 1 2 0 21 Wyoming 30 4 7 8 9 6 2 * Final rank for overall health system performance across five dimensions SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 52 53 ACCESS EXHIBIT A3 Access: Dimension Ranking and Performance on Indicators Indicator Performance Percent adults Percent Percent Percent adults without time when Dimension adults under children visited doctor in could not see doctor State Rank age 65 insured insured past two years because of cost United States 79.5 89.0 83.3 86.6 Alabama 31 79.8 94.0 81.4 83.4 Alaska 36 77.0 91.1 81.3 86.6 Arizona 33 76.0 84.5 85.7 87.3 Arkansas 42 75.6 91.5 80.5 83.8 California 44 75.5 87.0 76.7 87.7 Colorado 35 79.9 85.9 81.8 87.2 Connecticut 7 85.1 92.2 88.4 90.0 Delaware 19 83.3 88.3 87.9 90.3 District of Columbia 13 83.3 92.8 91.5 86.7 Florida 40 73.1 83.4 86.0 85.0 Georgia 37 76.6 88.7 83.6 84.2 Hawaii 1 87.2 94.7 88.9 96.6 Idaho 43 79.9 89.7 75.2 85.2 Illinois 24 81.9 89.3 84.9 88.4 Indiana 30 81.4 90.8 81.2 87.1 Iowa 3 87.8 94.6 84.1 91.6 Kansas 17 85.2 93.3 83.1 88.1 Kentucky 29 81.6 92.4 82.4 82.3 Louisiana 33 74.7 91.8 85.5 82.8 Maine 5 86.3 93.1 86.3 89.5 Maryland 21 81.5 90.6 88.9 88.3 Massachusetts 2 85.4 94.8 90.3 92.3 Michigan 10 84.1 94.4 85.3 88.1 Minnesota 9 89.0 93.7 82.8 89.0 Mississippi 48 77.7 87.5 80.0 80.1 Missouri 22 83.5 92.1 83.4 87.6 Montana 46 76.3 85.0 78.4 86.8 Nebraska 13 84.2 94.1 81.7 90.2 Nevada 47 77.5 84.8 77.0 85.3 New Hampshire 6 86.0 93.9 85.6 89.4 New Jersey 25 81.1 89.4 88.3 86.9 New Mexico 50 73.9 82.5 80.2 85.0 New York 11 82.1 92.5 88.8 87.8 North Carolina 32 79.6 88.9 86.0 83.1 North Dakota 18 85.3 90.7 82.3 93.3 Ohio 15 84.4 92.0 85.2 89.3 Oklahoma 49 74.5 86.1 81.7 82.0 Oregon 45 78.0 89.6 78.4 81.8 Pennsylvania 15 85.8 90.7 85.5 89.2 Rhode Island 4 85.2 92.5 89.7 90.9 South Carolina 28 77.6 90.8 86.9 84.8 South Dakota 19 83.7 91.6 82.3 91.2 Tennessee 26 81.3 90.6 85.9 85.8 Texas 51 69.6 79.8 78.2 80.9 Utah 38 80.7 88.5 76.6 87.1 Vermont 8 84.5 94.9 84.2 89.2 Virginia 23 81.7 91.7 84.0 87.5 Washington 27 82.0 92.1 81.6 85.9 West Virginia 38 75.9 92.0 81.7 80.7 Wisconsin 11 86.5 93.8 78.0 91.6 Wyoming 40 81.0 89.3 73.9 86.3 DATA: See Appendices B1 and B2 for data source of each indicator. SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 54 55 QUALIT Y EXHIBIT A5 Quality: Dimension Ranking and Performance on Indicators Indicator Performance Percent Percent children Percent Percent adult Percent children with emotional, Percent hospitalized Percent surgical adults age diabetics children ages with medical behavioral, or patients received patients received 50+ received received 19–35 months and dental developmental recommended care appropriate timing recommended recommended received five preventive problems received for AMI, CHF, and of antibiotics to preventive care preventive care vaccines care visits mental health care pneumonia prevent infections Dimension State Rank Getting the Right Care United States 39.7 39.4 80.8 58.8 58.7 84.1 69.5 Alabama 20 35.7 42.0 83.3 59.2 67.0 82.5 71.0 Alaska 49 38.5 41.3 75.4 54.5 52.2 82.9 57.5 Arizona 47 39.4 33.8 79.2 51.9 55.0 83.4 67.0 Arkansas 40 32.7 31.8 67.8 49.0 47.7 79.1 70.5 California 50 37.4 36.6 77.9 53.2 54.0 79.4 60.0 Colorado 30 41.2 50.2 83.4 57.7 56.9 86.1 69.5 Connecticut 4 47.3 42.4 86.1 71.6 74.1 87.5 90.0 Delaware 15 46.3 54.4 84.2 63.2 56.7 82.7 73.5 District of Columbia 25 45.6 * 73.5 65.7 66.1 * 65.5 Florida 45 40.9 37.1 79.3 54.2 54.7 80.3 68.0 Georgia 37 41.4 40.6 84.7 57.9 60.8 79.5 64.5 Hawaii 18 36.6 65.4 80.1 63.7 66.1 79.9 57.5 Idaho 39 32.6 33.1 78.1 45.7 56.9 85.2 72.5 Illinois 29 35.7 * 83.5 60.6 63.0 82.9 66.0 Indiana 28 36.3 39.8 78.1 61.2 66.1 84.5 62.0 Iowa 5 42.1 48.9 84.9 61.6 67.6 87.5 71.5 Kansas 19 39.7 43.2 83.8 60.7 61.3 84.0 65.5 Kentucky 38 35.0 35.6 79.7 60.5 62.5 81.7 62.0 Louisiana 41 37.2 38.6 76.0 51.3 44.2 80.6 59.0 Maine 2 46.8 45.7 83.3 66.4 67.6 85.3 74.5 Maryland 17 49.2 47.5 82.3 65.5 58.9 83.4 69.5 Massachusetts 3 46.7 48.9 93.5 74.9 67.6 85.8 75.5 Michigan 11 42.8 * 82.7 61.0 63.8 86.0 77.5 Minnesota 12 50.1 58.9 85.2 55.0 64.6 86.2 65.5 Mississippi 44 33.0 28.7 83.6 47.2 50.1 79.2 60.5 Missouri 33 38.5 42.6 79.3 56.1 60.2 83.5 72.5 Montana 13 41.1 47.8 79.6 48.9 68.4 86.0 79.5 Nebraska 9 37.3 46.6 89.1 58.5 72.8 87.8 71.5 Nevada 51 34.3 31.3 66.7 46.8 53.2 79.8 50.0 New Hampshire 6 48.6 51.0 82.8 71.8 63.5 86.6 69.0 New Jersey 16 42.5 42.0 78.2 68.3 58.7 87.7 75.0 New Mexico 41 38.7 50.3 78.4 55.3 58.3 79.0 71.5 New York 30 41.9 37.4 81.6 68.6 57.1 83.2 69.0 North Carolina 22 45.7 46.3 85.2 59.3 63.6 83.4 73.0 North Dakota 20 38.8 61.3 85.0 49.0 66.1 83.2 80.0 Ohio 23 38.1 39.2 84.1 61.2 61.2 84.9 64.5 Oklahoma 43 34.2 36.8 75.7 49.2 48.2 84.6 79.0 Oregon 36 40.0 * 72.9 52.2 62.7 83.9 75.0 Pennsylvania 14 38.4 40.6 83.2 66.6 75.8 81.6 62.5 Rhode Island 1 48.6 50.7 83.1 73.9 67.5 88.4 85.0 South Carolina 27 41.7 40.1 78.5 56.8 59.8 82.9 73.5 South Dakota 10 39.5 52.7 86.9 49.1 71.0 85.8 79.5 Tennessee 26 40.0 45.4 82.9 58.5 61.9 82.3 68.0 Texas 46 34.9 34.5 78.4 54.4 43.4 79.9 62.0 Utah 48 37.5 40.1 74.1 51.8 59.2 83.7 60.0 Vermont 7 44.4 47.2 81.5 70.7 70.0 85.3 73.5 Virginia 24 45.1 45.0 85.8 60.8 61.8 83.0 64.0 Washington 34 42.0 48.5 77.8 60.5 56.4 82.2 74.5 West Virginia 32 37.5 41.8 74.9 63.7 63.3 83.0 66.5 Wisconsin 8 43.8 50.5 82.2 61.3 66.8 85.9 71.5 Wyoming 35 37.3 40.0 78.6 56.9 77.2 80.3 58.5 * Indicates data value is missing. AMI = Acute myocardial infarction, CHF = Congestive heart failure DATA: See Appendices B1 and B2 for data source of each indicator. SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 56 QUALIT Y EXHIBIT A5 (continued) Quality: Dimension Ranking and Performance on Indicators (continued) Indicator Performance Percent Percent heart failure Medicare patients Percent Medicare Percent high-risk Percent nursing Percent adults Percent children patients given experienced good patients giving nursing home home residents with a usual with a medical instructions at communication best rating for residents with were physically source of care home discharge with provider care received pressure sores restrained Dimension State Rank Coordinated Care Patient-Centered Care United States 79.3 46.1 48.0 * * 13.4 7.4 Alabama 20 80.1 49.0 49 69.3 71.9 11.7 4.9 Alaska 49 70.5 37.7 37 67.6 65.4 13.3 5.5 Arizona 47 74.1 36.2 37 63.1 64.3 11.0 9.1 Arkansas 40 81.1 40.8 56 69.5 71.2 12.7 15.9 California 50 71.1 37.5 43 66.6 67.9 13.6 15.4 Colorado 30 79.7 45.8 33 65.9 62.4 9.7 6.4 Connecticut 4 86.5 59.1 61 68.6 71.1 13.2 7.3 Delaware 15 89.4 51.7 46 66.9 68.7 14.6 2.6 District of Columbia 25 77.7 45.2 53 71.0 67.5 19.3 2.5 Florida 45 75.4 43.0 49 65.1 67.0 14.2 9.4 Georgia 37 78.3 43.1 44 68.2 70.6 15.1 10.2 Hawaii 18 81.8 45.3 32 71.8 74.3 9.0 3.5 Idaho 39 73.4 37.9 50 67.2 70.3 8.3 6.2 Illinois 29 83.2 48.2 54 69.3 69.4 16.4 4.7 Indiana 28 84.4 51.0 51 68.9 70.5 14.4 5.8 Iowa 5 84.5 52.1 51 68.5 70.5 8.9 2.5 Kansas 19 84.2 49.8 31 68.3 71.5 12.2 3.6 Kentucky 38 82.8 50.5 41 68.7 68.5 13.7 6.9 Louisiana 41 77.6 39.2 54 72.4 71.8 18.1 14.2 Maine 2 88.9 56.6 62 73.4 73.4 10.5 4.7 Maryland 17 84.1 55.0 59 68.2 67.7 14.1 6.6 Massachusetts 3 87.1 60.3 51 71.6 71.8 13.3 6.7 Michigan 11 83.9 48.4 58 68.7 71.3 12.7 6.6 Minnesota 12 74.3 44.1 50 69.2 70.7 9.0 4.5 Mississippi 44 76.8 33.8 45 70.4 71.6 12.2 11.9 Missouri 33 82.8 47.7 45 68.4 69.1 13.5 7.1 Montana 13 74.9 40.9 33 72.2 74.4 7.8 3.0 Nebraska 9 83.3 49.0 44 71.2 71.2 8.1 1.9 Nevada 51 66.3 34.5 22 66.0 65.9 13.2 11.3 New Hampshire 6 87.4 61.0 53 68.5 69.8 11.1 3.3 New Jersey 16 83.0 52.7 67 69.1 68.3 18.4 5.2 New Mexico 41 76.4 39.0 14 64.4 61.2 11.5 8.2 New York 30 82.9 54.2 43 67.4 67.3 14.5 4.8 North Carolina 22 78.2 46.5 52 69.0 69.5 14.0 9.9 North Dakota 20 76.4 41.7 47 67.4 67.2 7.6 2.6 Ohio 23 84.1 52.3 61 68.6 70.0 13.3 7.1 Oklahoma 43 78.0 41.5 43 68.7 70.2 16.4 12.7 Oregon 36 75.7 43.4 31 67.7 69.0 10.7 9.2 Pennsylvania 14 88.6 54.0 47 70.3 72.5 13.7 4.7 Rhode Island 1 85.6 60.4 67 73.2 74.1 15.3 4.0 South Carolina 27 81.8 44.5 52 71.0 71.7 13.3 9.8 South Dakota 10 82.5 38.8 61 70.3 72.2 12.1 4.8 Tennessee 26 82.7 49.7 50 69.7 70.7 13.1 10.5 Texas 46 72.3 39.9 46 69.5 70.4 12.3 7.9 Utah 48 74.1 43.7 42 64.3 65.2 12.4 12.3 Vermont 7 85.5 57.8 51 74.9 71.2 15.7 3.6 Virginia 24 81.0 47.6 47 70.1 69.8 15.8 4.5 Washington 34 77.8 48.5 34 66.4 65.8 12.2 3.8 West Virginia 32 77.1 54.3 55 68.6 66.6 14.9 4.6 Wisconsin 8 83.7 51.2 59 70.0 70.1 10.6 3.2 Wyoming 35 74.9 40.5 36 71.6 70.9 11.6 6.2 * Indicates data value is missing. AMI=Acute myocardial infarction, CHF=Congestive heart failure DATA: See Appendices B1 and B2 for data source of each indicator. SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 57 QUALIT Y EXHIBIT A6 Hospital Quality Indicator Composite Percent and Rank: Hospitalized Patients Who Received Recommended Care for Acute Myocardial Infarction, Congestive Heart Failure, and Pneumonia Percent Rank State Composite AMI CHF Pneumonia Composite AMI CHF Pneumonia Alabama 82.5 90.2 82.8 74.5 35 42 34 23 Alaska 82.9 92.8 84.4 71.4 33 27 24 33 Arizona 83.4 91.4 85.2 73.7 24 37 17 26 Arkansas 79.1 88.3 76.4 72.6 49 49 49 32 California 79.4 91.9 82.0 64.2 47 34 36 50 Colorado 86.1 96.1 87.2 74.9 8 5 10 17 Connecticut 87.5 94.7 89.9 77.8 5 14 3 11 Delaware 82.7 93.4 84.8 70.0 34 21 19 41 District of Columbia * * * * * * * * Florida 80.3 89.3 83.6 68.1 41 46 28 46 Georgia 79.5 89.3 81.3 68.0 46 47 39 47 Hawaii 79.9 87.9 78.5 73.4 44 50 46 29 Idaho 85.2 94.8 82.9 78.0 16 12 31 9 Illinois 82.9 92.4 85.6 70.8 31 31 16 38 Indiana 84.5 92.8 82.9 77.8 19 26 33 12 Iowa 87.5 94.7 86.2 81.6 4 13 12 1 Kansas 84.0 93.3 80.7 78.0 20 23 41 8 Kentucky 81.7 91.2 79.2 74.6 38 39 45 20 Louisiana 80.6 90.2 80.4 71.2 40 43 43 37 Maine 85.3 96.0 89.4 70.5 15 7 5 40 Maryland 83.4 91.8 87.0 71.3 26 35 11 36 Massachusetts 85.8 97.1 90.4 69.8 13 1 2 42 Michigan 86.0 94.2 88.7 74.9 10 17 7 18 Minnesota 86.2 95.8 86.1 76.5 7 9 13 14 Mississippi 79.2 89.7 77.4 70.6 48 45 48 39 Missouri 83.5 92.2 83.6 74.8 23 33 27 19 Montana 86.0 96.1 82.9 78.9 9 4 32 7 Nebraska 87.8 95.2 87.8 80.4 2 10 8 4 Nevada 79.8 88.6 84.7 66.2 45 48 20 49 New Hampshire 86.6 95.8 89.7 74.3 6 8 4 24 New Jersey 87.7 93.3 89.0 80.6 3 22 6 2 New Mexico 79.0 92.5 78.3 66.3 50 30 47 48 New York 83.2 92.6 85.7 71.4 27 29 15 34 North Carolina 83.4 93.0 84.4 72.7 25 25 23 31 North Dakota 83.2 95.2 79.9 74.5 28 11 44 22 Ohio 84.9 93.9 87.4 73.4 17 19 9 28 Oklahoma 84.6 92.6 80.6 80.4 18 28 42 3 Oregon 83.9 93.7 84.6 73.3 21 20 21 30 Pennsylvania 81.6 91.8 84.2 68.8 39 36 25 45 Rhode Island 88.4 96.1 91.2 77.8 1 6 1 10 South Carolina 82.9 91.1 84.0 73.5 32 40 26 27 South Dakota 85.8 96.3 82.0 79.2 12 3 37 6 Tennessee 82.3 90.9 81.4 74.6 36 41 38 20 Texas 79.9 89.8 80.7 69.3 43 44 40 43 Utah 83.7 92.3 84.9 74.0 22 32 18 25 Vermont 85.3 96.8 83.2 76.1 14 2 30 15 Virginia 83.0 93.2 84.5 71.3 29 24 22 35 Washington 82.2 94.0 83.5 69.0 37 18 29 44 West Virginia 83.0 91.3 82.4 75.2 30 38 35 16 Wisconsin 85.9 94.5 85.9 77.3 11 15 14 13 Wyoming 80.3 94.3 66.1 80.4 42 16 50 5 * Indicates data value is missing. AMI = Acute myocardial infarction, CHF = Congestive heart failure DATA: 2004 CMS Hospital Compare data. See Appendix B1 for description of indicator. SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 58 59 AV O I D A B L E H O S P I TA L U S E A N D C O S T S EXHIBIT A8 Avoidable Hospital Use and Costs: Dimension Ranking Indicator Performance and Performance Percent Hospital Percent Medicare long-stay Percent nursing Percent on Indicators admissions for asthmatics admissions for Medicare nursing home home residents home health pediatric asthma with emergency ACS conditions 30-day hospital residents with readmission patients per 100,000 room or ugent per 100,000 readmission with hospital within three with hospital children care visit beneficiaries rates admission months admission Dimension Avoidable Hospital Use State Rank United States 187.6 17.4 7,712 18.4 * * 28 Alabama 41 * * 9,432 18.2 18.8 14.3 33.7 Alaska 19 * 13.1 4,726 15.7 * * 24.2 Arizona 8 141.9 15.5 5,565 18.0 12.3 7.2 20.0 Arkansas 44 * * 9,429 19.9 20.9 17.2 35.5 California 18 154.4 16.2 6,383 18.2 16.2 7.5 21.9 Colorado 15 174.9 * 5,729 17.5 11.3 9.6 22.5 Connecticut 25 * 16.6 6,647 16.6 12.6 9.4 30.4 Delaware 31 * 18.6 6,851 18.2 14.6 12.7 26.4 District of Columbia 47 * 25.8 8,101 20.4 * * 27.3 Florida 26 238.5 * 6,680 17.4 19.7 10.5 21.2 Georgia 32 184.2 19.8 8,531 18.1 20.1 14.0 28.8 Hawaii 4 160.7 13.1 4,069 14.5 * * 24.7 Idaho 3 * 11.0 5,591 14.8 9.2 8.9 22.8 Illinois 40 179.5 * 8,480 20.3 20.0 12.3 28.0 Indiana 33 * 20.3 8,113 17.6 16.9 12.6 30.3 Iowa 13 93.8 9.1 6,199 14.0 16.3 11.8 31.4 Kansas 26 162.8 * 7,328 18.9 14.2 13.1 27.2 Kentucky 45 273.3 19.2 10,452 19.0 21.2 14.2 36.1 Louisiana 51 * 17.4 11,368 23.8 24.9 17.3 46.4 Maine 21 111.5 * 6,798 17.5 8.7 11.0 27.1 Maryland 34 176.7 14.3 8,031 20.6 18.3 13.7 22.6 Massachusetts 35 154.4 13.7 7,830 19.8 16.0 10.1 29.0 Michigan 38 * 19.4 7,278 19.4 16.6 11.6 25.8 Minnesota 10 125.3 10.2 5,588 15.0 13.7 9.6 26.9 Mississippi 49 * 29.4 11,537 17.7 23.9 17.5 40.0 Missouri 30 220.7 18.7 8,084 17.8 19.4 12.9 26.6 Montana 7 * 15.1 6,468 14.9 10.1 9.1 22.9 Nebraska 14 91.0 * 6,492 14.3 14.0 10.6 24.8 Nevada 24 141.7 * 5,594 23.5 14.9 11.0 24.6 New Hampshire 20 * 12.8 6,246 17.1 8.9 9.9 29.8 New Jersey 46 225.6 18.7 8,526 18.3 23.2 16.0 26.5 New Mexico 5 * 13.6 5,811 15.3 8.5 9.5 24.3 New York 39 303.9 21.2 7,767 17.9 16.5 8.5 30.5 North Carolina 22 196.1 27.1 7,680 15.9 15.6 12.5 27.4 North Dakota 9 * * 6,662 16.2 10.4 12.3 23.7 Ohio 37 177.3 15.1 8,689 18.6 17.7 12.0 29.3 Oklahoma 50 * 18.8 9,392 20.5 21.5 15.4 37.1 Oregon 2 75.2 * 5,116 14.1 9.1 6.7 20.2 Pennsylvania 36 244.3 15.1 8,541 20.1 17.3 12.1 26.0 Rhode Island 23 212.4 16.2 8,025 15.5 14.9 9.5 26.4 South Carolina 26 314.2 * 7,766 16.0 15.5 12.3 28.8 South Dakota 17 * 12.5 7,225 20.2 16.3 11.3 22.5 Tennessee 42 221.6 * 9,764 18.2 18.7 13.2 34.7 Texas 48 210.4 15.5 8,794 20.6 21.3 14.9 34.5 Utah 1 91.8 11.9 4,432 15.8 8.3 7.9 18.3 Vermont 11 54.9 12.7 5,932 13.2 9.5 9.0 30.2 Virginia 29 187.2 21.7 7,328 16.7 16.1 12.8 27.5 Washington 6 149.2 11.9 4,706 16.5 10.8 8.4 20.9 West Virginia 42 197.8 * 10,424 17.0 20.7 16.1 34.9 Wisconsin 16 118.0 13.8 6,145 16.3 12.3 10.1 26.7 Wyoming 12 * * 6,016 13.3 13.4 10.3 25.6 * Indicates data value is missing. ACS = Ambulatory care sensitive DATA: See Appendices B1 and B2 for data source of each indicator. SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 60 AV O I D A B L E H O S P I TA L U S E A N D C O S T S EXHIBIT A8 (continued) Avoidable Hospital Use and Costs: Dimension Ranking Indicator Performance and Performance on Indicators (continued) Total single health insurance Total Medicare premium per reimbursements enrolled employee per enrollee Dimension Annual Costs State Rank United States $3,705 $6,611 Alabama 41 3,414 6,492 Alaska 19 4,379 6,431 Arizona 8 3,438 6,077 Arkansas 44 3,250 5,845 California 18 3,534 7,424 Colorado 15 3,684 6,114 Connecticut 25 3,864 7,384 Delaware 31 3,830 6,637 District of Columbia 47 4,218 6,312 Florida 26 3,807 7,225 Georgia 32 3,335 5,979 Hawaii 4 3,119 4,530 Idaho 3 3,429 5,126 Illinois 40 3,768 6,625 Indiana 33 3,586 5,851 Iowa 13 3,561 4,888 Kansas 26 3,711 6,070 Kentucky 45 3,542 6,384 Louisiana 51 3,485 7,716 Maine 21 4,116 5,581 Maryland 34 3,721 7,305 Massachusetts 35 4,141 7,804 Michigan 38 3,918 6,841 Minnesota 10 3,809 5,287 Mississippi 49 3,607 6,525 Missouri 30 3,559 5,990 Montana 7 3,680 5,178 Nebraska 14 3,725 5,370 Nevada 24 3,874 7,109 New Hampshire 20 4,084 5,842 New Jersey 46 3,882 8,076 New Mexico 5 3,401 5,120 New York 39 3,858 7,663 North Carolina 22 3,551 5,873 North Dakota 9 3,342 4,766 Ohio 37 3,782 6,470 Oklahoma 50 3,644 6,675 Oregon 2 3,706 4,933 Pennsylvania 36 3,671 6,860 Rhode Island 23 4,368 6,824 South Carolina 26 3,773 5,975 South Dakota 17 3,449 5,024 Tennessee 42 3,634 6,411 Texas 48 3,781 7,192 Utah 1 3,034 5,333 Vermont 11 4,074 5,580 Virginia 29 3,865 5,568 Washington 6 3,608 5,523 West Virginia 42 3,692 6,041 Wisconsin 16 3,927 5,407 Wyoming 12 3,761 5,323 * Indicates data value is missing. ACS = Ambulatory care sensitive DATA: See Appendices B1 and B2 for data source of each indicator. SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 61 Photo: Roger Carr 63 H ealthy L ives EXHIBIT A10 Healthy Lives: Dimension Ranking and Performance on Indicators Indicator Performance Percent adults Mortality under age 65 amenable limited in activities to health Breast cancer Colorectal because of care, deaths Infant mortality, deaths per cancer deaths physical, mental, Dimension per 100,000 deaths per 1,000 100,000 female per 100,000 or emotional State Rank population live births population population problems United States 103.3 7.0 25.6 19.7 14.1 Alabama 38 120.7 9.1 25.5 18.7 17.1 Alaska 4 78.2 5.6 20.5 17.9 16.8 Arizona 9 93.6 6.4 22.6 16.6 14.9 Arkansas 44 132.0 8.4 25.4 20.6 17.2 California 3 92.7 5.4 23.9 17.1 10.9 Colorado 2 79.0 6.0 22.7 18.1 12.6 Connecticut 17 86.7 6.5 25.3 19.3 13.0 Delaware 26 105.2 8.6 23.5 20.4 12.9 District of Columbia 48 160.0 11.0 34.1 24.6 10.8 Florida 25 96.9 7.5 23.7 18.2 17.1 Georgia 35 121.5 9.0 25.2 19.4 15.1 Hawaii 8 87.0 7.4 16.2 17.3 * Idaho 12 82.5 6.1 25.2 15.4 15.8 Illinois 36 112.8 7.4 27.1 22.0 12.5 Indiana 33 107.0 7.8 25.7 21.4 13.9 Iowa 9 86.8 5.3 24.8 20.0 11.9 Kansas 27 91.0 7.2 26.4 20.2 13.8 Kentucky 49 118.2 7.2 27.4 23.9 20.0 Louisiana 50 138.3 10.0 29.7 23.3 15.2 Maine 20 80.3 4.3 24.0 20.9 17.5 Maryland 39 110.8 7.6 29.4 20.8 15.4 Massachusetts 20 86.0 4.8 26.2 21.3 13.7 Michigan 37 109.2 8.1 26.8 19.3 17.5 Minnesota 7 70.2 5.3 22.7 18.5 17.4 Mississippi 51 150.4 10.0 26.6 22.5 19.9 Missouri 45 111.0 8.5 26.1 21.3 17.8 Montana 28 81.2 7.5 27.5 18.3 16.4 Nebraska 23 85.9 7.0 24.2 21.7 13.6 Nevada 31 111.5 6.1 25.9 21.2 14.3 New Hampshire 6 79.9 5.0 24.2 17.5 16.2 New Jersey 28 98.5 5.7 28.3 21.5 12.8 New Mexico 14 89.1 6.1 21.9 18.2 16.6 New York 30 103.6 6.0 26.1 20.5 16.0 North Carolina 34 114.4 8.1 26.4 19.6 14.3 North Dakota 17 86.2 6.3 26.0 20.0 11.4 Ohio 41 111.0 7.9 28.0 21.1 15.5 Oklahoma 47 120.1 8.2 27.0 20.0 18.4 Oregon 19 83.8 5.7 24.8 17.9 19.0 Pennsylvania 39 104.7 7.6 27.9 21.3 15.6 Rhode Island 22 96.6 7.1 23.4 21.1 13.3 South Carolina 43 126.3 9.3 26.9 20.0 15.9 South Dakota 11 80.4 6.7 23.9 19.0 13.7 Tennessee 42 128.0 9.3 25.8 19.9 16.5 Texas 24 111.2 6.3 24.3 18.8 14.5 Utah 1 71.1 5.6 23.9 15.3 14.2 Vermont 14 74.7 4.4 21.4 22.3 16.4 Virginia 32 104.1 7.4 26.9 20.2 15.1 Washington 13 80.3 5.8 23.8 16.9 18.6 West Virginia 45 119.0 8.9 22.8 23.1 22.8 Wisconsin 16 86.4 6.8 24.6 18.4 14.2 Wyoming 5 76.2 6.7 19.4 19.2 14.5 * Indicates data value is missing. DATA: See Appendices B1 and B2 for data source of each indicator. SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 64 H ealthy L ives EXHIBIT A11 Mortality Amenable to Health Care by Race, Deaths per 100,000 Population, 2002 Total White Black 2002 Rank 2002 Rank 2002 Rank United States 103.3 NA 93.6 NA 194.1 NA Alabama 120.7 44 100.6 39 195.5 27 Alaska 78.2 5 70.2 3 * * Arizona 93.6 24 91.5 29 139.9 6 Arkansas 132.0 48 117.7 50 241.5 42 California 92.7 23 90.3 27 183.0 19 Colorado 79.0 6 77.1 8 141.4 7 Connecticut 86.7 18 81.5 15 150.4 8 Delaware 105.2 31 94.6 31 164.6 14 District of Columbia 160.0 51 61.1 1 216.0 35 Florida 96.9 26 86.2 23 183.3 20 Georgia 121.5 45 98.7 37 197.3 29 Hawaii 87.0 20 84.5 19 106.0 1 Idaho 82.5 12 82.5 17 * * Illinois 112.8 39 96.9 36 224.4 38 Indiana 107.0 32 100.8 40 192.0 24 Iowa 86.8 19 85.9 22 163.0 12 Kansas 91.0 22 85.5 21 201.1 32 Kentucky 118.2 41 113.8 47 196.1 28 Louisiana 138.3 49 106.1 44 225.1 39 Maine 80.3 8 80.5 11 * * Maryland 110.8 34 91.2 28 176.7 17 Massachusetts 86.0 15 84.6 20 137.3 5 Michigan 109.2 33 94.0 30 220.3 37 Minnesota 70.2 1 67.6 2 164.9 15 Mississippi 150.4 50 116.5 49 232.4 40 Missouri 111.0 35 101.0 41 210.5 34 Montana 81.2 11 79.0 9 * * Nebraska 85.9 14 81.0 13 193.4 25 Nevada 111.5 38 107.5 45 176.8 18 New Hampshire 79.9 7 80.5 11 * * New Jersey 98.5 27 88.9 26 183.8 22 New Mexico 89.1 21 87.6 24 120.3 3 New York 103.6 28 96.5 35 157.1 11 North Carolina 114.4 40 95.9 33 195.2 26 North Dakota 86.2 16 81.3 14 156.1 10 Ohio 111.0 35 101.5 43 197.5 30 Oklahoma 120.1 43 115.6 48 185.2 23 Oregon 83.8 13 84.0 18 120.0 2 Pennsylvania 104.7 30 96.4 34 199.1 31 Rhode Island 96.6 25 94.6 31 152.7 9 South Carolina 126.3 46 99.2 38 208.1 33 South Dakota 80.4 10 72.8 5 * * Tennessee 128.0 47 111.8 46 237.1 41 Texas 111.2 37 101.2 42 218.4 36 Utah 71.1 2 70.3 4 * * Vermont 74.7 3 74.7 6 * * Virginia 104.1 29 88.5 25 183.4 21 Washington 80.3 8 79.7 10 133.6 4 West Virginia 119.0 42 118.3 51 163.4 13 Wisconsin 86.4 17 82.1 16 166.2 16 Wyoming 76.2 4 76.0 7 * * * Indicates data value is missing. NA = Not applicable DATA: Analysis of 2002 CDC Multiple Cause-of-Death data files using Nolte and McKee methodology, BMJ 2003. SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 65 EXHIBIT A12 State Demographics: Income and Health Status Mortality amenable Percent of populationIndicator Performance to health care, deaths per with incomes less than 200% Cancer incidence rate 100,000 population of federal poverty level Median household income per 100,000 population 2002 Rank 2004-2005 Rank 2004-2005 Rank 2002 Rank United States 103.3 36 $46,071 462.2 Alabama 120.7 44 41 42 37,502 45 437.7 8 Alaska 78.2 5 30 8 56,398 6 425.9 6 Arizona 93.6 24 40 37 45,279 25 391.6 1 Arkansas 132.0 48 43 47 36,406 48 * * California 92.7 23 40 37 51,312 12 441.0 11 Colorado 79.0 6 30 8 51,518 11 440.4 9 Connecticut 86.7 18 27 3 56,889 5 494.6 38 Delaware 105.2 31 31 11 50,445 14 486.8 34 District of Columbia 160.0 51 42 45 44,949 27 482.9 33 Florida 96.9 26 37 32 42,440 36 457.8 17 Georgia 121.5 45 38 34 44,140 30 467.3 24 Hawaii 87.0 20 33 18 58,854 3 408.6 4 Idaho 82.5 12 36 30 45,009 26 454.4 15 Illinois 112.8 39 33 18 48,008 18 482.6 30 Indiana 107.0 32 34 22 43,091 34 462.0 19 Iowa 86.8 19 31 11 45,671 24 469.4 27 Kansas 91.0 22 34 22 42,233 37 440.6 10 Kentucky 118.2 41 41 42 36,750 47 498.2 41 Louisiana 138.3 49 45 50 37,442 46 482.6 30 Maine 80.3 8 35 28 43,317 31 508.9 43 Maryland 110.8 34 29 6 59,762 2 488.0 35 Massachusetts 86.0 15 29 6 54,888 8 505.8 42 Michigan 109.2 33 34 22 44,801 28 488.8 36 Minnesota 70.2 1 24 2 56,098 7 475.3 29 Mississippi 150.4 50 48 51 34,396 51 * * Missouri 111.0 35 35 28 43,266 32 447.8 13 Montana 81.2 11 41 42 36,202 49 462.3 20 Nebraska 85.9 14 31 11 46,587 20 459.9 18 Nevada 111.5 38 36 30 48,496 17 482.6 30 New Hampshire 79.9 7 23 1 57,850 4 495.1 39 New Jersey 98.5 27 27 3 60,246 1 516.5 45 New Mexico 89.1 21 44 49 39,916 42 402.6 2 New York 103.6 28 38 34 46,659 19 469.3 26 North Carolina 114.4 40 38 34 41,820 39 416.9 5 North Dakota 86.2 16 32 16 41,362 40 444.4 12 Ohio 111.0 35 33 18 44,349 29 453.0 14 Oklahoma 120.1 43 40 37 39,292 44 456.7 16 Oregon 83.8 13 37 32 43,262 33 465.9 23 Pennsylvania 104.7 30 34 22 45,941 22 496.2 40 Rhode Island 96.6 25 33 18 49,511 16 514.6 44 South Carolina 126.3 46 40 37 40,107 41 467.8 25 South Dakota 80.4 10 34 22 42,816 35 * * Tennessee 128.0 47 40 37 39,376 43 * * Texas 111.2 37 43 47 42,102 38 427.0 7 Utah 71.1 2 34 22 53,693 9 405.7 3 Vermont 74.7 3 28 5 49,808 15 463.4 21 Virginia 104.1 29 30 8 52,383 10 * * Washington 80.3 8 31 11 51,119 13 491.2 37 West Virginia 119.0 42 42 45 35,467 50 472.1 28 Wisconsin 86.4 17 32 16 45,956 21 465.1 22 Wyoming 76.2 4 31 11 45,817 23 * * * Indicates data value is missing. DATA: Mortality amenable - 2002 CDC Multiple Cause-of-Death data files using Nolte and McKee methodology (Nolte and McKee, BMJ 2003); Income less than 200% of poverty - Kaiser statehealthfacts.org (2005 and 2006 Current Population Survey); Median income - 2005 and 2006 Current Population Survey; Cancer - Kaiser statehealthfacts.org (National Cancer Institute); Overweight, Asthma, Diabetes, and Smoking - Kaiser statehealthfacts.org (2005 BRFSS) SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 66 EXHIBIT A12 (continued) State Demographics: Income and Health Status (continued) Percent of adults Percent of adults Percent of adults Indicator Performance ever been who have who are overweight who have been told told by a doctor that Percent of adults or obese they have asthma they have diabetes who smoke 2005 Rank 2005 Rank 2005 Rank 2005 Rank United States 58.5 12.6 7.3 20.6 Alabama 62.3 44 11.2 9 9.8 48 24.8 45 Alaska 62.4 47 12.5 24 4.4 1 25.0 46 Arizona 53.5 5 12.2 21 7.5 28 20.3 22 Arkansas 61.9 42 11.4 13 8.1 35 23.5 42 California 57.9 18 13.2 33 7.1 21 15.1 2 Colorado 52.3 3 13.3 34 4.8 2 19.8 15 Connecticut 55.2 9 12.4 23 6.5 11 16.5 3 Delaware 60.5 37 12.6 25 8.6 42 20.7 27 District of Columbia 52.1 2 15.3 49 7.1 21 20.1 20 Florida 58.2 20 11.7 17 8.8 43 21.7 33 Georgia 60.5 37 11.5 14 8.3 39 22.2 35 Hawaii 51.4 1 14.1 40 7.3 25 17.0 4 Idaho 58.3 24 11.7 17 6.8 17 17.9 7 Illinois 58.2 20 10.6 3 7.9 33 19.9 17 Indiana 59.5 32 12.7 29 8.3 39 27.2 50 Iowa 60.2 34 11.6 15 6.8 17 20.4 23 Kansas 58.0 19 10.8 4 6.9 19 17.8 6 Kentucky 62.5 49 13.3 34 8.9 44 28.7 51 Louisiana 62.3 44 10.8 4 9.2 47 22.5 37 Maine 56.9 16 15.0 48 7.5 28 20.8 28 Maryland 58.2 20 13.1 31 7.2 24 18.9 11 Massachusetts 52.9 4 14.2 41 6.4 8 18.1 9 Michigan 60.5 37 13.9 39 8.1 35 22.1 34 Minnesota 59.4 30 11.8 20 5.8 5 20.0 18 Mississippi 64.9 51 11.1 7 9.8 48 23.7 44 Missouri 62.4 47 14.2 41 7.7 30 23.4 41 Montana 54.7 8 12.6 25 5.7 4 19.2 12 Nebraska 60.2 34 10.8 4 7.3 25 21.3 30 Nevada 56.2 12 12.6 25 7.1 21 23.1 40 New Hampshire 57.4 17 14.7 46 6.5 11 20.4 23 New Jersey 55.3 10 11.7 17 7.7 30 18.0 8 New Mexico 58.2 20 14.5 44 7.3 25 21.5 32 New York 56.8 15 13.8 38 8.1 35 20.5 25 North Carolina 59.4 30 10.1 1 8.5 41 22.6 39 North Dakota 62.0 43 11.1 7 6.7 15 20.2 21 Ohio 60.3 36 11.3 12 7.7 30 22.4 36 Oklahoma 60.6 40 13.3 34 8.9 44 25.0 46 Oregon 56.5 13 15.3 49 6.7 15 18.5 10 Pennsylvania 59.2 28 12.3 22 8.1 35 23.6 43 Rhode Island 55.8 11 15.3 49 6.4 8 19.7 14 South Carolina 62.3 44 11.2 9 10.3 50 22.5 37 South Dakota 60.7 41 10.5 2 6.4 8 19.8 15 Tennessee 59.3 29 11.6 15 9.1 46 26.7 48 Texas 58.8 25 11.2 9 7.9 33 20.0 18 Utah 54.3 7 13.0 30 5.5 3 11.5 1 Vermont 54.2 6 14.9 47 6.0 6 19.3 13 Virginia 58.9 26 14.2 41 6.9 19 20.6 26 Washington 56.5 13 14.6 45 6.3 7 17.5 5 West Virginia 63.6 50 13.4 37 10.4 51 26.7 48 Wisconsin 59.1 27 13.1 31 6.6 14 20.8 28 Wyoming 59.6 33 12.6 25 6.5 11 21.4 31 * Indicates data value is missing. DATA: Mortality amenable - 2002 CDC Multiple Cause-of-Death data files using Nolte and McKee methodology (Nolte and McKee, BMJ 2003); Income less than 200% of poverty - Kaiser statehealthfacts.org (2005 and 2006 Current Population Survey); Median income - 2005 and 2006 Current Population Survey; Cancer - Kaiser statehealthfacts.org (National Cancer Institute); Overweight, Asthma, Diabetes, and Smoking - Kaiser statehealthfacts.org (2005 BRFSS) SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 67 EXHIBIT A13 State Demographics: Race and Ethnic Groups, U.S. (2005) and States (2004-2005) White Black Hispanic Other United States 67 12 15 6 Alabama 69 26 2 3 Alaska 69 3 5 23 Arizona 58 3 31 7 Arkansas 77 16 4 3 California 44 6 35 14 Colorado 72 4 20 4 Connecticut 77 9 10 4 Delaware 69 20 7 4 District of Columbia 31 56 9 4 Florida 62 15 20 3 Georgia 59 29 8 4 Hawaii 19 2 7 72 Idaho 85 0 10 4 Illinois 68 15 12 6 Indiana 85 9 4 2 Iowa 90 2 5 3 Kansas 83 5 6 5 Kentucky 89 7 1 2 Louisiana 63 32 2 3 Maine 95 1 1 3 Maryland 58 28 7 6 Massachusetts 80 5 8 6 Michigan 78 14 4 4 Minnesota 86 4 4 6 Mississippi 58 37 3 2 Missouri 83 11 3 3 Montana 90 0 2 7 Nebraska 83 4 9 4 Nevada 60 7 22 10 New Hampshire 94 1 1 3 New Jersey 64 13 16 7 New Mexico 44 2 44 11 New York 61 15 16 8 North Carolina 67 21 7 5 North Dakota 89 1 2 9 Ohio 83 12 3 3 Oklahoma 73 7 5 15 Oregon 82 2 9 8 Pennsylvania 83 10 4 3 Rhode Island 80 6 10 4 South Carolina 66 29 3 2 South Dakota 87 1 3 10 Tennessee 77 17 4 2 Texas 48 11 37 4 Utah 84 1 10 5 Vermont 95 1 1 3 Virginia 68 19 7 6 Washington 78 3 7 12 West Virginia 95 3 0 1 Wisconsin 86 6 5 4 Wyoming 89 1 7 3 DATA: Kaiser statehealthfacts.org (2005 and 2006 Current Population Survey) SOURCE: Commonwealth Fund State Scorecard on Health System Performance, 2007 68 Appendix B.1. State Scorecard Indicator Descriptions and Data Sources Complete references for data sources are provided in Appendix B.2. Indicator Description Indicator Description 1 Adults under age 65 insured: Employee Benefits 8 Children with both medical and dental preventive Research Institute (EBRI) analysis of 2005 and 2006 care visits: Percent of children ages 0–17 with one U.S. Census Bureau Current Population Survey (CPS) or more medical and dental preventive care visits March Supplement (U.S. Census Bureau, 2005, 2006). during the past 12 months. Child and Adolescent Health Measurement Initiative (CAHMI) analysis 2 Children insured: EBRI analysis of 2005 and 2006 U.S. of the 2003 National Survey of Children’s Health Census Bureau CPS March Supplement (U.S. Census (CAHMI 2005). Bureau, 2005, 2006). 9 Children with emotional, behavioral, or 3 Adults visited a doctor in past two years: Rutgers developmental problems received mental health Center for State Health Policy (CSHP) analysis of 2000 care: Percent of children ages 1–17 with current Behavioral Risk Factor Surveillance System (BRFSS) emotional, developmental, or behavioral problems (NCCDPHP, BRFSS 2000). requiring treatment or counseling who received 4 Adults with a time in past year when they needed to some type of mental health care during the past 12 see a doctor but could not because of cost: Rutgers months. CAHMI analysis of 2003 National Survey of CSHP analysis of 2002 and 2004 BRFSS (NCCDPHP, Children’s Health (CAHMI 2005). BRFSS 2002, 2004). 2002 data was imputed for one 10 Hospitalized patients received recommended care state. for acute myocardial infarction, congestive heart 5 Adult age 50 and older received recommended failure, and pneumonia: Proportion of cases where a preventive care: Percent of adults age 50 and older hospital provided the recommended process of care who have received: sigmoidoscopy or colonoscopy for patients with acute myocardial infarction (AMI), in the last ten years or a fecal occult blood test in congestive heart failure (CHF), and pneumonia for 10 the last two years; a mammogram in the last two indicators. The composite includes 5 clinical services years (women only); a pap smear in the last three for AMI (aspirin within 24 hours before or after years (women only); and a flu shot in the past year arrival at the hospital and at discharge; beta-blocker and a pneumonia vaccine ever (age 65 and older within 24 hours after arrival and at discharge; and only). Rutgers CSHP analysis of 2002 and 2004 BRFSS angiotensin-converting enzyme (ACE) inhibitor (NCCDPHP, BRFSS 2002, 2004). 2002 data were for left ventricular systolic dysfunction), 2 for CHF imputed for one state. (assessment of left ventricular function and the use of an ACE inhibitor for left ventricular dysfunction), 6 Adult diabetics received recommended preventive and 3 for pneumonia (initial antibiotic therapy care: Percent of adults age 18 and older who were received within four hours of hospital arrival, told by a doctor that they had diabetes and have pneumococcal vaccination, and assessment of received: hemoglobin A1c test, dilated eye exam, oxygenation). Analysis of 2004 CMS Hospital and foot exam in the past year. Rutgers CSHP Compare data conducted by A. Jha and A. Epstein at analysis of 2002 and 2004 BRFSS (NCCDPHP, BRFSS the Harvard School of Public Health (DHHS n.d.). 2002, 2004). 2002 data were imputed for six states. 11 Surgical patients received appropriate timing of 7 Children ages 19–35 months received all antibiotics to prevent infections: Proportion of cases recommended doses of five key vaccines: Percent of where a hospital provided prophylactic antibiotics children ages 19 to 35 months who have received within 1 hour prior to surgery and discontinued at least 4 doses of diphtheria-tetanus-acellular within 24 hours after surgery. Data from 2005 CMS pertussis (DTaP), at least 3 doses of polio, at least Hospital Compare (DHHS n.d.), reported in AHRQ 1 dose of measles-mumps-rubella (MMR), at least 2006 National Healthcare Quality Report (AHRQ 3 doses of Haemophilus influenzae B (Hib), and at 2006). least 3 doses of hepatitis B antigens. Data from the 2005 National Immunization Survey (NCHS, NIS 12 Adults with a usual source of care: Percent of adults 2005). age 18 and older who have one (or more) person they think of as their personal doctor or health care provider. Rutgers CSHP analysis of 2002 and 2004 BRFSS (NCCDPHP, BRFSS 2002, 2004). 2002 data were imputed for one state. 13 Children with a medical home: Percent of children ages 0–17 who have at least one preventive medical care visit in the past year; are able to access needed specialist care and services; and have a personal doctor/nurse who usually/always spends enough time and communicates clearly, provides telephone advice and urgent care when needed, and follows up after specialist care. CAHMI analysis of 2003 National Survey of Children’s Health (CAHMI 2005). 69 Appendix B.1. State Scorecard Indicator Descriptions and Data Sources (continued) Complete references for data sources are provided in Appendix B.2. Indicator Description Indicator Description 14 Heart failure patients given written instructions 21 Medicare hospital admissions for ambulatory at discharge: Percent of heart failure patients with sensitive conditions per 100,000 beneficiaries: documentation that they or their caregivers were Hospital admissions of fee-for-service Medicare given written instructions or other educational beneficiaries age 65 and older for one of 11 materials at discharge. Data retrieved from CMS ambulatory care sensitive conditions (AHRQ Hospital Compare database on January 25, 2006 Indicators): short-term diabetes complications, (DHHS n.d.). long-term diabetes complications, lower extremity amputation among patients with diabetes, 15 Medicare fee-for-service patients whose health asthma, chronic obstructive pulmonary disease, provider always listens, explains, shows respect, hypertension, congestive heart failure, angina and spends enough time with them: Data from (without a procedure), dehydration, bacterial 2003 National Consumer Assessment Healthcare pneumonia, and urinary tract infection. Analysis Providers and Systems (CAHPS) Benchmarking of 2003 Medicare Standard Analytical Files (SAF) Database (AHRQ, CAHPS n.d.), reported in AHRQ 5% Inpatient Data conducted by G. Anderson and 2005 National Healthcare Quality Report (AHRQ R. Herbert at Johns Hopkins Bloomberg School of 2005). Public Health (CMS, SAF 2003). 16 Medicare fee-for-service patients giving a best 22 Medicare 30-day hospital readmission rates: Fee-for- rating for health care received: Percent of Medicare service Medicare beneficiaries age 65 and older with fee-for-service patients who reported a doctor’s initial admissions due to one of 31 select conditions visit in the last 12 months and gave a best rating who are readmitted within 30 days following for health care received. Data from 2003 National discharge for the initial admission. Analysis of 2003 CAHPS Benchmarking Database (AHRQ, CAHPS n.d.), Medicare SAF 5% Inpatient Data conducted by G. reported in AHRQ 2005 National Healthcare Quality Anderson and R. Herbert at Johns Hopkins (CMS, Report (AHRQ 2005). SAF 2003). 17 High-risk nursing home residents with pressure 23 Long-stay nursing home residents with a hospital sores: Data from 2004 CMS Minimum Data Set admission: Analysis of 2000 Medicare enrollment (CMS, MDS n.d.), reported in AHRQ 2005 National data and MedPAR file conducted by V. Mor at Brown Healthcare Quality Report (AHRQ 2005). University, under a grant funded by the National 18 Long-stay nursing home residents who were Institute of Aging (#AG20557, State Policies and physically restrained: Data from 2004 CMS Minimum Hospitalizations from Nursing Homes). Data Set (CMS, MDS n.d.), reported in AHRQ 2005 24 Nursing home residents with a hospital readmission National Healthcare Quality Report (AHRQ 2005). within three months: Percent of long-stay 19 Hospital admissions for pediatric asthma per residents hospitalized within three months of 100,000 population: Data from 2002 Healthcare Cost being discharged from a hospital to a nursing and Utilization Project State Inpatient Databases home. Analysis of 2000 Medicare enrollment data (AHRQ, HCUP-SID 2002), reported in AHRQ 2005 and MedPAR file conducted by V. Mor at Brown National Healthcare Quality Report (AHRQ 2005). University, under a grant funded by the National Institute of Aging (#AG20557). 20 Asthmatics with an emergency room or urgent care visit: Percent of adults age 18 and older who were 25 Home health patients with a hospital admission: told by a doctor that they had asthma and had an Percent of acute care hospitalization for home health emergency room or urgent care visit in the past 12 episodes. Data from 2004 Outcome and Assessment months. Rutgers CSHP analysis of 2001, 2002, 2003 Information Set (CMS, OASIS n.d.), reported in AHRQ and 2004 BRFSS (NCCDPHP, BRFSS 2001, 2002, 2003, 2005 National Healthcare Quality Report (AHRQ 2004). 2005). 26 Total single premium per enrolled employee at private-sector establishments that offer health insurance: Data from 2004 Medical Expenditure Panel Survey – Insurance Component (AHRQ, MEPS- IC 2004). 70 Appendix B.1. State Scorecard Indicator Descriptions and Data Sources (continued) Complete references for data sources are provided in Appendix B.2. Indicator Description Indicator Description 27 Total Medicare reimbursements per enrollee: 29 Infant mortality, deaths per 1,000 live births: Data 2003 data from Dartmouth Atlas of Health Care from 2002 National Vital Statistics System (NVSS) (Dartmouth Atlas Project 2003). Total Medicare fee- (NCHS, NVSS n.d.), reported in AHRQ 2005 National for-service reimbursements include payments for Healthcare Quality Report (AHRQ 2005). both Part A and Part B (exclude capitated payments). Reimbursement rates were indirectly adjusted for 30 Breast cancer deaths per 100,000 female population: sex, race, and age, and were further adjusted for Age-adjusted to US 2000 standard population. illness, and regional differences in price. Data from 2002 NVSS (NCHS, NVSS n.d.), reported in AHRQ 2005 National Healthcare Quality Report 28 Mortality amenable to health care: Number of (AHRQ 2005). deaths before age 75 per 100,000 population that resulted from causes considered at least partially 31 Colorectal cancer deaths per 100,000 population: treatable or preventable with timely and appropriate Age-adjusted to US 2000 standard population. medical care (see list), as described in Nolte and Data from 2002 NVSS (NCHS, NVSS n.d.), reported McKee (Nolte and McKee, BMJ 2003). Analysis in AHRQ 2005 National Healthcare Quality Report conducted by K. Hempstead at Rutgers CSHP using (AHRQ 2005). 2002 mortality data from CDC Multiple Cause-of- 32 Adults under age 65 limited in any activities because Death file and U.S. Census Bureau population data of physical, mental, or emotional problems: Rutgers (NCHS, MCD n.d.). CSHP analysis of 2004 BRFSS (NCCDPHP, BRFSS Cause of deaths Age 2004). Intestinal infections 0–14 Tuberculosis 0–74 Other infections (diphtheria, Tetanus, 0–74 septicaemia, poliomyelitis) Whooping cough 0–14 Measles 1–14 Malignant neoplasm of 0–74 colon and rectum Malignant neoplasm of skin 0–74 Malignant neoplasm of breast 0–74 Malignant neoplasm of cervix uteri 0–74 Malignant neoplasm of cervix 0–44 uteri and body of uterus Malignant neoplasm of testis 0–74 Hodgkin’s disease 0–74 Leukaemia 0–44 Diseases of the thyroid 0–74 Diabetes mellitus 0–49 Epilepsy 0–74 Chronic rheumatic heart disease 0–74 Hypertensive disease 0–74 Cerebrovascular disease 0–74 All respiratory diseases (excluding 1–14 pneumonia and influenza) Influenza 0–74 Pneumonia 0–74 Peptic ulcer 0–74 Appendicitis 0–74 Abdominal hernia 0–74 Cholelithiasis and cholecystitis 0–74 Nephritis and nephrosis 0–74 Benign prostatic hyperplasia 0–74 Maternal death All Congenital cardiovascular anomalies 0–74 Perinatal deaths, all causes, All excluding stillbirths Misadventures to patients during All surgical and medical care Ischaemic heart disease: 50% 0–74 of mortality rates included 71 Appendix B.2. Complete References for Data Sources DHHS, Hospital Compare (U.S. Department of Health and Human Services, Hospital Compare Database). (n.d.). Washington, DC: AHRQ (Agency for Healthcare Research and Quality). (2006). National http://www.hospitalcompare.hhs.gov/Hospital/Static/Resources- Healthcare Quality Report, 2006. AHRQ Pub. No. No. 07-0013. DownloadDB.asp?dest=NAV|Home|Resources|DownloadDB#TabTop Rockville, MD: U.S. Department of Health and Human Services. NCCDPHP, BRFSS (National Center for Chronic Disease Prevention AHRQ (Agency for Healthcare Research and Quality). (2005). National and Health Promotion, Behavioral Risk Factor Surveillance System). Healthcare Quality Report, 2005. AHRQ Pub. No. 06-0018. Rockville, (2000, 2001, 2002, 2003, 2004). Atlanta, GA: Centers for Disease MD: U.S. Department of Health and Human Services. Control. http://www.cdc.gov/brfss/index.htm. AHRQ, CAHPS (Agency for Healthcare Research and Quality, NCHS, MCD (National Center for Health Statistics, Multiple Cause-of- Consumer Assessment of Healthcare Providers and Systems). (n.d.). Death Data Files). (n.d.). Hyattsville, MD: Centers for Disease Control Rockville, MD: Center for Quality Improvement and Patient Safety, and Prevention. U.S. Department of Health and Human Services. NCHS, NIS (National Center for Health Statistics, National AHRQ, HCUP-SID (Agency for Healthcare Research and Quality, Immunization Survey). (2005, n.d.). Hyattsville, MD: Centers for Disease Healthcare Cost and Utilization Project-State Inpatient Databases). Control and Prevention. (2001, 2002). Rockville, MD: Center for Delivery, Organization, and Markets, U.S. Department of Health and Human Services. NCHS, NVSS (National Center for Health Statistics, National Vital Statistics System). (n.d.). Hyattsville, MD: Centers for Disease Control AHRQ, MEPS-IC (Agency for Healthcare Research and Quality, Medical and Prevention. Expenditure Panel Survey-Insurance Component). (2004). Washington, D.C.: U.S. Department of Health and Human Services. http://www. Nolte and McKee. (2003). “Measuring the Health of Nations: Analysis meps.ahrq.gov/Data_Pub/IC_TOC.htm. of Mortality Amenable to Health Care.” London, UK: British Medical Journal Volume 327, November 15, 2003. CAHMI (Child and Adolescent Health Measurement Initiative). (2005). National Survey of Children’s Health. Portland, OR: Data Resource U.S. Census Bureau, CPS (Current Population Survey) March Center on Child and Adolescent Health, Oregon Health and Science Supplement. (2005, 2006). Washington, D.C.: U.S. Department of University. www.nschdata.org. Commerce. CMS, MDS (Centers for Medicare and Medicaid Services, Minimum Data Set). (n.d.). Baltimore, MD: U.S. Department of Health and Human Services. CMS, OASIS (Centers for Medicare and Medicaid Services, Outcome and Assessment Information Set). (n.d.). Baltimore, MD: U.S. Department of Health and Human Services. CMS, SAF (Centers for Medicare and Medicaid Services, Standard Analytic File 5% Inpatient Data). (2003). Baltimore, MD: U.S. Department of Health and Human Services. Dartmouth Atlas Project (2003). Dartmouth Atlas of Health Care. Hanover, NH: Center for the Evaluative Clinical Sciences, Dartmouth Medical School. http://www.dartmouthatlas.org/index.shtm. 72 About the Authors Health and directed special projects at the UMASS Labor Relations and Research Center. During Joel C. Cantor, Sc.D., is the director of the the 1980s, she directed the Service Employees Center for State Health Policy and professor of International Union’s research and policy de- Public Policy at Rutgers University. Dr. Cantor’s partment. Earlier, she served as staff to President research focuses on issues of health care regula- Carter’s national health insurance task force. Prior tion, financing, and delivery. His recent work to federal service, she was a research fellow at the includes studies of health insurance market regula- Brookings Institution. She has authored numerous tion, state health system performance, and access publications on health policy and insurance issues, to care for low-income and minority popula- and national/international health system per- tions. Dr. Cantor has published widely on health formance, including the Fund’s 2006 National policy topics, and serves on the editorial board Scorecard on U.S. Health System Performance of the policy journal Inquiry. He is a frequent and co-authored the book Health and the War advisor on health policy matters to New Jersey on Poverty. She holds an undergraduate degree state government and was the 2006 recipient of in economics from Smith College and a graduate Rutgers University President’s Award for Research degree in economics from Boston College. in Service to New Jersey. Dr. Cantor received his doctoral degree in health policy and management Sabrina K. H. How, M.P.A., is research associate from the Johns Hopkins Bloomberg School of for the Fund’s Commission on a High Performance Public Health. Health System. She is co-author of the Commis- sion’s 2006 National Scorecard on U.S. Health System Dina Belloff, M.A., is a senior research analyst Performance. Ms. How also served as program at the Rutgers Center for State Health Policy. She associate for two programs, Health Care in New conducts research and policy analysis on access to York City and Medicare’s Future. Prior to joining care, affordability of care, and health care financing. the Fund, she was a research associate for a manage- Prior to coming to the Center, she worked at the ment consulting firm focused on the health care U.S. General Accounting Office determining the industry. Ms. How holds a B.S. in biology from adequacy of Medicare Part B reimbursement for Cornell University and an M.P.A. in health policy covered prescription drugs. She also worked at and management from New York University. Mathematica Policy Research in Princeton, N.J., where she participated in evaluations of Medicaid Douglas McCarthy, M.B.A., president of Issues expansion programs, prospective payment for Research, Inc., in Durango, Colo., is senior home health care, and social health maintenance research advisor to The Commonwealth Fund. organizations. She received her bachelor’s degree He supports The Commonwealth Fund Com- with highest honors from Rutgers College and a mission on a High Performance Health System’s master’s degree in health policy studies from the Scorecard project and is a contributing editor Johns Hopkins University. to the bimonthly newsletter Quality Matters. Cathy Schoen, M.S., is senior vice president Mr. McCarthy received his bachelor’s degree at The Commonwealth Fund, a member of the with honors from Yale College and a master’s Fund’s executive management team, and research degree in health care management from the director of the Fund’s Commission on a High University of Connecticut. During 1996–97, Performance Health System. Her work includes he was a public policy fellow at the Humphrey strategic oversight of surveys, research, and policy Institute of Public Affairs at the University of initiatives to track health system performance. Pre- Minnesota. He has more than 20 years of ex- viously Ms. Schoen was on the research faculty of perience in public and private sector research, the University of Massachusetts’ School of Public policymaking, and management. 73 Further Reading Publications listed below can be found on The Commonwealth Fund’s Web site at www.commonwealthfund.org. Mirror, Mirror on the Wall: An International Update on the Comparative Performance of American Health Care (May 2007). Karen Davis, Cathy Schoen, Stephen C. Schoenbaum, Michelle M. Doty, Alyssa L. Holmgren, Jennifer L. Kriss, and Katherine K. Shea. An Analysis of Leading Congressional Health Care Bills, 2005-2007: Part I, Insurance Coverage (Mar. 2007). Sara R. Collins, Karen Davis, and Jennifer L. Kriss. The Agency for Healthcare Research and Quality’s National Healthcare Quality Report, 2006 (Mar. 2007). Stephen C. Schoenbaum, Douglas McCarthy, and Cathy Schoen. State Strategies to Expand Health Insurance Coverage: Trends and Lessons for Policymakers (Jan. 2007). Alice Burton, Isabel Friedenzohn, and Enrique Martinez-Vidal. Slowing the Growth of U.S. Health Care Expenditures: What Are the Options? (Jan. 2007). Karen Davis, Cathy Schoen, Stuart Guterman, Tony Shih, Stephen C. Schoenbaum, and Ilana Weinbaum. The State Children’s Health Insurance Program: Past, Present, and Future (Jan. 2007). Jeanne M. Lambrew. The National Committee for Quality Assurance’s The State of Health Care Quality 2006 (Nov. 2006). Stephen C. Schoenbaum and Alyssa L. Holmgren. Why Not the Best? Results from a National Scorecard on U.S. Health System Performance (Sept. 2006). The Commonwealth Fund Commission on a High Performance Health System. “U.S. Health System Performance: A National Scorecard” (Sept. 20, 2006). Cathy Schoen, Karen Davis, Sabrina K. H. How, and Stephen C. Schoenbaum. Health Affairs Web Exclusive. Framework for a High Performance Health System for the United States (Aug. 2006). The Commonwealth Fund Commission on a High Performance Health System. Public Views on Shaping the Future of the U.S. Health System (Aug. 2006). Cathy Schoen, Sabrina K. H. How, Ilana Weinbaum, John E. Craig, Jr., and Karen Davis. Gaps in Health Insurance: An All-American Problem—Findings from the Commonwealth Fund Biennial Health Insurance Survey (Apr. 2006). Sara R. Collins, Karen Davis, Michelle M. Doty, Jennifer L. Kriss, and Alyssa L. Holmgren. 74 O N E E AST 75TH STR E E T N E W YO R K , N Y 10 021-2692 T E L 2 12 . 6 0 6 . 3 8 0 0 FA X 212 . 6 0 6 . 35 0 0 w w w. co m m o nwe a l t hf u n d . o r g