Using Telephone Support to Manage Chronic Disease June 2005 Using Telephone Support to Manage Chronic Disease Prepared for CALIFORNIA HEALTHCARE FOUNDATION By John D. Piette, Ph.D. June 2005 About the Author John D. Piette, Ph.D., is a Department of Veterans Affairs career scientist and an associate professor of internal medicine at the University of Michigan, Ann Arbor. About the Foundation The California HealthCare Foundation, based in Oakland, is an independent philanthropy committed to improving California’s health care delivery and financing systems. Formed in 1996, our goal is to ensure that all Californians have access to affordable, quality health care. For more information, visit us online (www.chcf.org) This report was produced under the direction of CHCF’s Chronic Disease Care Program, which seeks to improve the health of Californians by working to assure those with chronic diseases receive care based on the best scientific knowledge. Visit www.chcf.org/programs/ for more information about CHCF and its programs. ISBN 1-932064-86-9 Copyright © 2005 California HealthCare Foundation Contents 4 Overview 7 I. Background Prevalence of Telephones in the U.S. Use of Interactive Technology in Telephone Care 10 II. Benefits of Telephone Care Clinical Effectiveness Cost-Effectiveness 12 III. Telephone Support in Chronic Illness Care Supporting Administrative Processes Patient Assessment Patient Education and Counseling Facilitating Peer Support 18 IV. Which Patients Can Benefit? Patients with Limited Health Literacy Patients with Multiple Chronic Health Problems Patients with Gaps in Care Value of Registries and Service Targeting 21 V. Linking Telephone Care Services with Usual Processes of Care In-house and Contracted Outside Programs Stand-alone Services 23 VI. Evaluating Telephone Care Programs 25 VII. Conclusion and Recommendations 27 Appendix: Acknowledgments 28 Endnotes Overview TELEPHONE CARE SERVICES CAN ENHANCE THE delivery of care to patients with one or more chronic illness and help them to self-manage their disease. By providing for regular contact with these patients, telephone programs can: monitor patients’ status between visits; deliver patient educa- tion or other counseling; send appointment reminders; and facilitate peer support and referrals for coping with illness. However, the benefits of telephone care depend on a number of variables, including the target population, program struc- ture, computer support, specific goals, and other factors. Research findings have begun to build a body of knowledge that can assist health systems and plans in designing and implementing telephone care programs. The purpose of this report is to inform clinicians and health care managers about the benefits and challenges of telephone care programs, and what is known to date about how to opti- mize effectiveness in a cost-constrained health care environ- ment. A variety of research findings on the clinical effective- ness as well as the cost-effectiveness of telephone care pro- grams are cited. Although the picture is not yet complete, there is some evidence that telephone-based patient education can improve chronic disease outcomes and help patients become more effective advocates for their own care. In addition, automated telephone reminders increase the likeli- hood that patients will keep appointments and take their medication; such support of administrative processes has great potential in outpatient settings. Health systems and health plans are looking to telephone care as a way to fill service gaps caused by funding cuts and reductions in staff resources. However, these services can be labor-intensive and therefore expensive. Research findings on the cost-effectiveness of such programs are less conclusive than those focused more on clinical effectiveness. In fact, short-term cost-effectiveness is often an unrealistic goal, since effective telephone care can lead to greater use of services in the near term. 4 | CALIFORNIA HEALTHCARE FOUNDATION To help health systems and health plans in their Health systems and plans must decide whether decision-making about telephone care, this to provide telephone care services in-house or report offers a number of observations that have contract with an outside vendor. Challenges emerged from the published findings and the with outsourced services include coordinating author’s experience in this field. These include patient care with additional providers, sharing the following: information across organizational boundaries, and monitoring program success. Advantages I Telephone counseling should be clearly may include access to sophisticated technology structured and based on established platforms for providing services. A key to suc- behavior-change principles. cess with the vendor option is contracting I Programs should be designed with specific based on explicit measurable targets. goals in mind, and should not try to accomplish too much at one time. Over the coming decade, as clinicians and health care systems establish more effective ways I Services should target patients who can of implementing and evaluating telephone care most benefit from them. Patients with programs, chronically ill patients may well bene- limited health literacy, multiple chronic fit from greater access to education, treatment, illnesses, or gaps in their care may be the and improved outcomes. At the same time, best candidates for telephone care programs. health systems and health plans may benefit from I Programs that draw from patient registries, more cost-effective ways to organize and deliver electronic medical records, or claims data- care to their chronically ill patients. bases may be most effective in identifying patients most in need of assistance. About This Report I The most effective programs are closely This report is intended to inform clinicians linked with outpatient care and clinician and health care managers about telephone care follow-up. services and programs for patients with chronic I Regular screening and assessment tools can illnesses. Specifically, the report addresses: be useful to help telephone care providers I How telephone care services can contribute determine whether they should intensify, to improved patient care; reduce, or discontinue services. I Characteristics of effective programs; I Telephone peer-support programs deserve serious consideration by both health care I Patients most likely to benefit from tele- systems and researchers. phone care; I Evaluation should incorporate process I Integrating telephone care services into measures such as number and type of systems of care; and patient contacts as well as changes in I How to evaluate programs and identify patients’ health and resource use; these areas for improvement. combined findings can provide an assess- ment of a program’s impact and point to ideas for improvement. Using Telephone Support to Manage Chronic Disease | 5 The report is relevant to “traditional” telephone care services (delivered “live” by nurse care man- agers or other clinician counselors), as well as those that use automated technology to augment such programs. More “high tech” telemedicine formats, such as video consultation with special- ists and multimedia communication between patients and primary care providers, are beyond the scope of this report. However, it should be noted that such services could prove valuable, especially for patients in vulnerable socio-eco- nomic groups and those living in rural areas or prison populations1, 2—and may even be cost- effective.3, 4 Comprehensive information on telemedicine can be found in reviews by the Cochrane Collaboration5 and by McBride and Rimer.6 The report draws on the author’s experience in evaluating telephone care and the use of interac- tive technology in chronic illness care, as well as a systematic search of Medline to identify ran- domized trials and reviews focused on the use of telephone care to manage chronic disease or promote health behavior change; and semi- structured interviews with health care systems and vendors specializing in telephone care delivery. The interviewees represent some of the largest and most experienced providers of tele- phone care services in the United States. 6 | CALIFORNIA HEALTHCARE FOUNDATION I. Background CHRONIC DISEASES CAN PRESENT ALMOST overwhelming difficulties for both patients and clinicians. Typically, chronically ill patients must monitor themselves for early signs of acute exacerbations; comply with medication reg- imens; make difficult changes in their health behaviors (e.g., diet and physical activity levels); and negotiate the often frus- trating processes involved in receiving and paying for health care. For those with more than one chronic condition—as many as 21 percent of all Americans and 62 percent of older adults—coordination of services and medicine management is even more complex. Managing a chronic disease is particularly difficult for patients in vulnerable socio-economic groups, who often receive care from safety-net health care systems with limited resources. Although effective chronic disease management usually requires frequent outpatient visits, these patients face multi- ple barriers to getting these services. Many have limited health literacy or English proficiency, which complicates communication with clinicians and makes it difficult to complete eligibility applications. Such patients may also have limited transportation and inflexible work schedules that make attendance at frequent visits difficult or impossible. Long waiting times for appointments and extended stays in clinic waiting rooms make face-to-face clinical encounters both more frustrating and less effective.7 In addition, mental health problems, which are common among people with certain chronic illnesses, may limit patients’ ability to meet their day-to-day self-care demands. Physicians and their staffs face equally difficult challenges in organizing effective and affordable care for their chronically ill patients. Managing multiple chronic diseases for a single patient requires complex scheduling, medicine regimens, and monitoring tasks—in addition to the counseling and patient education that is crucial to effective self-care. Time and cost burdens for managing a whole population of such patients can be a major problem for providers, particularly in a time of financial constraints. Some chronically ill patients may need weekly or even daily support for their self-care, demands that strain even the most effective clinic-based care. In fact, health care providers often are unaware of chronically ill patients’ self-management goals8 or financial pressures.9, 10 Using Telephone Support to Manage Chronic Disease | 7 Telephone care services can assist both chroni- less likely to have their blood pressure checked.12, 13 cally ill patients and their caregivers by address- It is important for health care systems to create ing some of these challenges to effective care alternatives to meet the needs of these people. management. By allowing clinicians and patients to communicate without a formal Use of Interactive Technology office visit, telephone care can address disease in Telephone Care management problems in a more timely way and enable communication when patients are Telephone care services delivered by nurses can in their homes or workplaces. be labor-intensive and therefore costly. While clinicians in outpatient settings devote much of However in order to be effective, telephone their working day to patient care—as opposed to support services must be carefully organized. administrative functions—the reverse is often Otherwise they can easily become costly add-ons true for telephone care providers. Researchers in that deliver no true benefits. two studies found that telephone care nurses averaged less than 15 minutes per patient per Prevalence of Telephones month actually counseling patients.14, 15 Moreover, in the United States effective behavior-change efforts may require even more frequent and extended conversations Telephone care is widely accessible because the with patients than is typical in outpatient set- overwhelming majority of Americans have a tings. Given the nursing shortages and financial phone. Less than 3 percent of U.S. households pressures in safety-net health care systems, are “phoneless” and the phenomenon is even less administrators may consider telephone-based common among older adults, the population behavioral interventions unaffordable. To be with the highest prevalence of chronic illness. economically feasible, programs must have As a group, phoneless people share the same rates computer support capable of increasing their of common chronic diseases found in the popu- efficiency. lation as a whole, as well as similar blood pressure and cholesterol levels. Interactive Voice Response (IVR) systems, which use hardware and software available in most Nevertheless, an important minority are beyond voicemail systems, allow patients to respond the reach of telephone support. Census data indi- to queries for clinical information and select cate that households below the poverty level are appropriate health education messages using nearly five times as likely to be without a tele- touch-tone or voice recognition technologies. As phone as higher-income households (Table 1). a component of a telephone care program, IVR African-Americans and Native Americans are may allow clinicians to communicate with large more likely to lack a phone than Caucasians. numbers of patients at relatively low cost. Unlike One study found that 66 percent of Americans systems that require the patient to use a com- without telephones have less than a high school puter (such as text messaging, email, or Web- education.11 People without a telephone are also based communication), IVR requires only that more likely than other Americans to report fair patients have either a standard household tele- or poor health status (38 percent versus 16 per- phone or cell phone. Therefore, this report gives cent); they are less likely to have had their choles- special emphasis to the use of IVR as a tool for terol checked in the prior year (21 percent versus extending the reach of telephone care providers. 56 percent); less likely to be physically active; and 8 | CALIFORNIA HEALTHCARE FOUNDATION Table 1: Availability of Telephone Service Among U.S. Households With Telephone Without Telephone Number of households 86,503,689 1,762,641 Percent of all households 98% 2% Race White 98% 2% Black 95% 5% Asian 99% 1% American Indian/Alaskan Native 88% 12% Native Hawaiian/Pacific Islander 96% 4% Two or more races 96% 4% Ethnicity White, Non-Hispanic/Latino 98% 2% Hispanic/Latino 95% 5% Age Group 15 to 24 94% 6% 25 to 34 97% 3% 35 to 44 97% 3% 45 to 54 98% 2% 55 to 64 98% 2% 65 to 74 99% 1% 75+ 99% 1% Poverty Level Below poverty level 90% 10% At or above poverty level 98% 2% Source: Data compiled from 2000 Census data (www.census.gov). Using Telephone Support to Manage Chronic Disease | 9 II. Benefits of Telephone Care HEALTH CARE SYSTEMS TYPICALLY ARE MOTIVATED to provide telephone care for their chronically ill patients because they want to improve treatment effectiveness and reduce costs. As discussed below, there is some evidence that telephone care improves clinical outcomes, but little evidence that it decreases the overall cost of health care, particularly in the short term. Clinical Effectiveness Although the study findings are not uniformly positive, there is evidence that telephone care programs can enhance both the processes and outcomes of chronic disease care. For example, telephone care can improve diabetes patients’ glycemic control and symptom burden,16, 17 and improve other key outcomes for patients with asthma,18 heart failure,19 and chronic pain.20 Many studies have found that telephone care programs improve self- management behaviors, including the proper use of medication and self-monitoring. Two recent trials highlight some of the design features that can make telephone care programs especially effective. One of these studies focuses on depression, which is an ideal disease target for telephone care, because: (1) it is a common chronic disease; (2) it has clear guidelines for disease management; and (3) costly recurrences often result from patients’ difficulty adhering to clinician follow-up and self-management goals. In this study, investigators evaluated the impact of telephone counseling on patients who were beginning treatment with antidepressants. They found that cognitive-behavioral therapy (CBT) for depressed patients can be delivered effectively via telephone.21 Intervention patients received their first tele- phone contact by care managers soon after initiating antide- pressant therapy. This was followed by a structured CBT counseling program. Each telephone contact included a brief assessment of symptoms and medication adherence, as well as carefully scripted counseling on strategies for enhancing com- pliance. Each patient received a detailed self-management workbook that reinforced messages delivered during the calls. After six months, patients who received the telephone care were substantially more likely to have improvements in their depressive symptoms, and be more satisfied with their treat- 10 | CALIFORNIA HEALTHCARE FOUNDATION ment, than patients who did not receive the goal of telephone care is to increase access among intervention. patients who have difficulty using clinic-based services, programs may increase resource use and Like depressed patients, those with heart failure cost, at least in the short term. For example, tele- often experience problems with self-management phone care support for diabetes patients may and preventable exacerbations, and therefore may increase patients’ use of recommended services benefit from telephone care supports. In a recent such as retinal exams, cholesterol tests, and home study,22 intervention patients received specialized glucose monitoring supplies. electronic home scales to monitor their weight and report changes to telephone care nurses. In one influential study, telephone follow-up was These patients completed daily symptom assess- substituted for face-to-face outpatient visits ments tailored to their unique needs by their among chronically ill patients treated in Veterans cardiologist. Cardiac nurses monitored patients’ Affairs health care systems. The program led to weight and symptom reports and re-contacted significant decreases in costly acute care use, patients by phone within 24 hours if they identi- outpatient visits, and medication use.24 However, fied a health or behavioral problem. The nurses when the intervention was replicated among consulted with patients’ cardiologists by phone Medicare patients, there were no cost-savings as needed to resolve problems. After six months, and “telephone appointments became simply there were 60 percent fewer deaths among an additional service.”25 patients receiving the telephone care relative to Telephone care for asthma patients can be cost- those receiving usual care (8 percent mortality effective when delivered in conjunction with versus 18 percent). In addition, there were fewer other services.26 For patients with heart failure, hospitalizations and emergency department results of randomized trials have been variable,27 visits, although those findings were not statisti- with some studies showing cost-savings28 and cally significant. others showing either no benefit29 or inconclusive These two rigorous multi-site randomized trials findings.30,31 A study of arthritis patients found demonstrate that telephone care can improve that telephone care had little positive impact on outcomes, if it is carefully structured and has treatment costs, but that, overall, the service was strong links to patients’ usual outpatient care. cost-neutral.32 However, a comparative study by the Cochrane General conclusions about the cost implications Collaboration concluded that the evidence for of telephone care programs are difficult to make, the effectiveness of telephone care is mixed, and since the research studies evaluated a wide range that the low quality of most telephone care of interventions and the participating patients studies makes it difficult to discern consistent had a variety of clinical and socio-demographic findings.23 Growing interest in telemedicine has characteristics. For example, for patients with increased both the quality and number of these diabetes or other long-term illnesses, the impact trials. As more is known about what works of telephone care on disease severity may take best, programs are likely to be less varied and years to realize, and no long-term trials to identify their clinical effectiveness easier to measure and such effects have been conducted. Given that compare. telephone care services are often poorly reim- bursed, health care systems and health plans Cost-Effectiveness may have difficulty investing near-term dollars Research findings on the cost-effectiveness of to achieve uncertain long-term gains. telephone care are mixed. Because an important Using Telephone Support to Manage Chronic Disease | 11 III. Telephone Support in Chronic Illness Care LIKE ANY CLINICAL SERVICE, TELEPHONE CARE programs are most effective when they are designed with spe- cific goals in mind. For chronically ill patients, telephone care may be effectively used to: I Assist patients with administrative tasks (e.g., follow-up visit reminders); I Monitor patients to identify health and behavioral problems; I Deliver patient education or other disease management counseling; or I Facilitate informal support (such as peer support) for coping with illness. Regardless of the specific goal, the most effective telephone care providers keep the process on track by making sure that each contact has specific, explicit, and realistic goals. This focus is especially important when patients have multiple chronic illnesses or a variety of psychosocial challenges. Telephone care providers should help patients understand the focus of each call and the limits to the services that the clinician can provide. Providers need to be familiar with other services available to patients and procedures for making referrals. Supporting Administrative Processes Patients managing one or more chronic illnesses, especially if other problems are present such as limited education or English proficiency, often miss their scheduled appointments. No-show rates are often highest among those with the greatest need for clinical care.33 In one study,34 more than a third of diabetes patients who lacked health insurance or had Medicaid coverage went without some prescription drugs in the prior year due to cost concerns—even though nearly all of those patients were eligible for first-dollar medication coverage through drug cost assistance programs. Interactive Voice Response (IVR) systems are ideally suited to place brief, outgoing messages focused on administrative tasks such as reminder calls. In one study,35 registry-based IVR reminder calls led to increased vaccination rates among low- income patients and were just as effective as “live” follow-up calls. A seminal study conducted in a public health care system 12 | CALIFORNIA HEALTHCARE FOUNDATION found that reminder calls delivered via IVR to between face-to-face visits. Although most tuberculosis patients increased visit attendance patient monitoring occurs during outpatient rates.36 The calls were effective for patients with a visits, few health care organizations have the variety of primary languages including Mandarin, information systems needed to trigger a com- Vietnamese, Tagalog, and Spanish. Low-tech prehensive assessment when patients seek care alternatives such as mailed reminders and “live” through different entry points (e.g., an emer- telephone reminders may also improve atten- gency department). As a result, clinicians often dance rates, but IVR reminders are cost-effective miss opportunities to prevent health crises, and even in the context of these more labor-intensive educational efforts lack the timeliness they need alternatives.37 Other studies have found that to be effective. automated reminders can assist patients in taking One recent study40 found that telephone assess- their medications as prescribed.38 ments increased the proportion of asthma patients who received appropriate monitoring, Patient Assessment compared to face-to-face consultation in outpa- Telephone assessments may be an ideal way to tient clinics (74 percent versus 48 percent). In monitor the status of patients in order to identi- addition, telephone consultations were ten min- fy health or self-care problems before they result utes shorter on average than clinic-based assess- in acute crises. Interactive monitoring tools, ments. Even though there were no differences in such as IVR, electronic scales,39 or electronic patients’ asthma-related quality of life associated blood pressure cuffs can be a useful component with the telephone care program, it increased of many telephone care programs, improving patients’ access and resulted in outcomes compa- the information base available to clinicians rable to face-to-face care. Figure 1: Prevalence of Psychiatric Diagnoses Identified Using an IVR-Delivered Assessment (IVR PRIME), a Face-to-Face Interview (FF PRIME), and a Standard Clinical Interview (SCID) 60– IVR PRIME 50– FF-PRIME SCID 40– 30– 20– 10– 0– Any Mood Disorder Dysthymia Alcohol Abuse Bulimia Nervosa Source: Reprinted with permission from: Kobak KA, Taylor LV, Dottl SL, et al. A computer-administered telephone interview to identify mental disorders. JAMA 1997;278:905-910. Note: PRIME = Primary Care Evaluation of Mental Disorders; IVR = interactive voice response telephone administration; FF = face-to- face; SCID = Structured Clinical Interview for DSM-IV Mental Disorders. Using Telephone Support to Manage Chronic Disease | 13 IVR assessments may be an effective way of Patient Education and Counseling extending the reach of telephone care man- Patients with chronic illnesses often need large agers—allowing them to monitor the status of amounts of health education, and those needs large numbers of patients and focus their atten- may change over the disease course. Unfortu- tion on those with the greatest need for “live” nately, safety-net and other providers frequently counseling or follow-up. Low-income patients do not communicate effectively with patients,46 are able and willing to complete regular IVR and many clinicians are unaware of their patients’ assessments over an extended period of time.41 In self-management goals.47 Patients often remember fact, patients often see IVR assessment calls as an little of what they are told during outpatient integral component of their disease manage- encounters,48 and health information conveyed ment—especially when IVR-reported problems during acute illness episodes may be even more are soon followed up by clinicians. IVR-based difficult for patients to process.49 Patients with screening for mental health problems can provide language barriers or low health literacy may lack comparable data to that obtained during face-to- even basic information about their disease and face clinical encounters (Figure 1),42 and patients’ self-care.50, 51 reports about their physical and mental function- ing are similar whether obtained via IVR or Telephone care can help overcome these barriers “live” telephone interviewers.43 Importantly, these by providing patients with important health studies and others have found that more patients information at a time and pace that increases with psychiatric symptoms are identified using comprehension and retention. There is some IVR assessments than when patients have to evidence that telephone-based patient education report this sensitive information directly to can improve chronic disease outcomes. In one another person. study,52 patients discharged from an academic general medicine service received a follow-up Health care systems must carefully plan for how call by a pharmacist two days after discharge to they will use the information gathered through review the patient’s medications and reinforce IVR assessments. Asking general questions about educational messages. More patients receiving the patients’ status may obligate providers to sched- follow-up calls were satisfied with their discharge ule in-person follow-ups for vague or self-limiting medication instructions compared to patients health problems. Screening and “case-finding” without telephone follow-up (86 percent versus with feedback to providers has little impact on 61 percent). Pharmacists identified and resolved patient outcomes when providers have limited medication-related problems in 19 percent of ability to change practice patterns, or treatment counseled patients and referred 15 percent to changes are not tightly linked with health out- their inpatient team. Most important, only 10 comes.44,45 Providers often lack the resources percent of patients from the phone call group required to effectively follow up on serious, but returned to the emergency department within chronic patient needs, such as dysthymia (mild 30 days, compared to 24 percent of patients chronic depression) or barriers to self-manage- who were not called. ment. In designing telephone care assessments, it is important to balance the repercussions of Telephone care is well-suited for patients seeing patients for erroneously identified “prob- attempting difficult behavioral changes related lems” versus missing potential patient needs due to smoking, diet, or physical activity. Although to assessment protocols that are not sufficiently some studies of telephone-based smoking cessa- sensitive. tion counseling have shown little benefit,53,54 one 14 | CALIFORNIA HEALTHCARE FOUNDATION large study found that telephone counseling ses- that parents calling a telephone triage line were sions to callers of a statewide smokers’ helpline equally compliant with instructions about self- increased quit attempts and overall quit rates care and seeking urgent care regardless of (Figure 2).55 whether telephone counseling was provided by nurses or pediatricians. (However, parents were Many telephone care programs focus on self- somewhat less likely to follow instructions to seek management regimens (e.g., diet, glucose non-urgent outpatient care when nurses provided self-monitoring, and medication adherence). the advice.) Other programs focus on self-empowerment. This approach helps patients become effective Patient education programs that incorporate self-advocates in making decisions with their cli- structured behavior change strategies are more nicians and receiving services such as laboratory effective than those that use free-flowing encoun- monitoring and appropriate medications. It can ters. There are several well-established models for be a potent tool to improve the process and out- motivating behavior change including cognitive- comes of chronic illness care.56, 57 behavioral therapy,59 problem-solving therapy,60 and motivational interviewing.61 These tech- Some research has been done to see whether niques have been successfully used to support patient responsiveness to telephone advice is behavior change related to chronic illnesses. related to the counselor’s professional back- ground. In a recent study,58 investigators found Figure 2: Rates of Abstinence Among California Smokers Who Did or Did Not Call a Telephone Cessation Help Line 25%– Did Not Call (n=1,309) 20%– Did Call (n=1,973) Percent Abstinence 15%– 10%– 5%– 0– I Month 3 Months 6 Months I2 Months Length of Abstinence Source: Reprinted with permission from: Zhu SH, Anderson CM, Tedeschi GJ, Rosbrook B, Johnson CE, Byrd M, et al. Evidence of real-world effectiveness of a telephone quitline for smokers. New England Journal of Medicine 2002;347(14):1106-9. Using Telephone Support to Manage Chronic Disease | 15 One useful technique is to ask patients to restate Facilitating Peer Support instructions in their own words, so that the clini- Many chronically ill patients lack effective social cian can assess the effectiveness of his or her own support, and therefore are at greater risk for poor explanations. This technique, the “Interactive self-care and health outcomes.66 One solution to Communication Loop” (see Figure 3) checks for this problem is telephone-based peer support, lapses in recall and understanding. It also can which has been shown to help not only those uncover health beliefs, reinforce messages, and who receive the support, but those who give it. activate patients by opening a dialogue about their self-care goals and values.62, 63 Such enhance- Individuals who provide social support to others ments in recall and comprehension improve experience less depression,67 heightened self- subsequent adherence.64 A recent study of face- esteem and self-efficacy,68 and improved quality to-face clinical encounters found that physicians of life.69 Providing support to others can lead to rarely use this communication tool, although it improved health behaviors on the part of the is strongly associated with improved clinical out- helper,70 decreased mortality risk,71 and improved comes among diabetes patients.65 health outcomes.72 Peer support between individ- uals living with the same illness can be especially Figure 3: The Interactive Communication Loop in Clinician-Patient Counseling Clinician assesses New concept, health patient recall and information, advice, or comprehension L change in management L L Clinician explains new concepts Clinician clarifies and Patient recalls and tailors explanation comprehends L L L Clinician reassesses Adherence patient recall and comprehension Source: Reprinted with permission from Schillinger D, Grumbach K, Piette JD, et al. Closing the loop: Missed opportunities in communicating with diabetes patients who have health literacy problems. Archives of Internal Medicine 2003;163:83-90. 16 | CALIFORNIA HEALTHCARE FOUNDATION effective in reducing problematic health behav- In the face of growing numbers of chronically ill iors73 and mental health symptoms.74 However, patients and significant resource constraints, tele- most chronic disease peer-support models require phone peer-support programs, such as those facil- patients to attend frequent outpatient visits. itated by IVR, may be a promising approach for Given the constraints on safety-net providers and both health care systems and researchers. their patients, these services often are not feasible. Telephone-based peer-helper interventions can be a satisfactory substitute for face-to-face peer interaction,75 and many people prefer the relative anonymity and increased privacy of talking on the telephone.76, 77 Some studies suggest that tele- phone-based peer-support interventions may lead to improvements in chronic disease outcomes.78, 79 However, patients may be reluctant to share their telephone numbers and pay the cost of telephone calls. Even willing participants sometimes lack the initiative or organization to ensure that con- tacts are made regularly. From a health system perspective, telephone peer-support initiatives can be difficult to monitor, and few if any have been designed to interface with standard outpa- tient nursing care. Researchers at the University of Michigan recent- ly conducted a peer-support pilot program for elderly diabetes patients, facilitated by IVR tech- nology. In this system, patients did not need to share phone numbers, and calls could be blocked during certain hours or at the request of either partner. The IVR system generated automatic reminder calls to participants who had not con- tacted each other in a given week. More than 80 percent of patients in the pilot study spoke to their partner regularly and found the IVR system easy to use. Nearly all participants said they would be more satisfied with their health care if IVR-facilitated peer-support services were available. Participants also found positive rein- forcement for their own behavioral goals by supporting their partners’ efforts to manage their self-care. Using Telephone Support to Manage Chronic Disease | 17 IV. Which Patients Can Benefit? TELEPHONE CARE IS NOT EQUALLY BENEFICIAL for everyone. Many patients, including some in safety-net health care systems, already have the resources they need to manage their illness effectively; they may receive little additional benefit from telephone care. At the other extreme, some patients may not benefit from even the most creative telephone care program; this category includes some patients with serious psychiatric disorders, and those with unstable residences or inconsistent telephone access. In selecting patient populations for telephone care, providers often target those with the poorest health status (e.g., diabetes patients with the worst glycemic control or heart failure patients with the most acute exacerbations). However, it should be noted that telephone care probably offers the greatest benefit to the large number of patients who simply need reminders, monitoring, self-management information, and coaching. Telephone care must be responsive to patients’ changing needs over time. For example, patients may benefit from additional self-management education soon after a new diagnosis, an acute episode, or a significant change in treatment (e.g., after adding insulin to a diabetic’s medication regimen). Clinicians and health care systems should develop triage protocols such as screening and assessment tools to determine whether to inten- sify, reduce, or even discontinue services. Drawing updated information from clinical registries is particularly effective in identifying when patients need additional assistance. Patients with Limited Health Literacy Functional health literacy (FHL) consists of skills such as basic reading and numerical tasks that are critical in the health care environment.80 Poor FHL is common among patients with low educational attainment, those from racial/ethnic minority groups, older patients, and individuals whose primary language is not English. As many as one-third of all Medicare recipients and most patients treated in public health care settings have poor FHL. One study found that Medicaid patients with low FHL had annual health care costs that were more than four times those of other patients.81 Diabetes patients with low FHL 18 | CALIFORNIA HEALTHCARE FOUNDATION are more than twice as likely to have retinopathy as on the limited attention that clinicians can (a serious diabetes-related complication) com- devote to self-management education during pared to patients with adequate FHL; and they outpatient encounters. are almost three times as likely to have cere- Telephone care services can be valuable for such brovascular disease.82 Patients with low FHL are patients, although even telephone care providers more likely to report that they do not understand cannot address all patient problems at the same their providers’ explanations of their health con- time. To address complex and multiple needs, dition or instructions on how to manage their telephone care vendors often use algorithms to care.83 identify priorities and set patient-specific man- Telephone care providers targeting Medicaid agement goals. patients and other socio-economically vulnerable populations report that they face many of the Patients with Gaps in Care challenges associated with serving low-FHL pop- ulations. To meet these patients’ needs, providers Although outpatient disease management may link with social services, and provide addi- protocols are established for almost all common tional support such as purchasing telephones chronic illnesses, many patients fail to receive for patients or using videos rather than written recommended standards of care. Telephone care material to reinforce self-management education. programs that draw from patient registries, elec- Nevertheless, large telephone care providers with tronic medical records, or claims databases can Medicaid programs report that such services are be effective in targeting patients with significant feasible and can be managed within budget gaps in their treatment. For example, telephone constraints. care providers for diabetes patients may be most effective in preventing cardiovascular complica- Patients with limited FHL may be ideally suited tions if they ensure that patients receive appro- for telephone care support and should be a high- priate blood pressure and lipid monitoring, priority target population when telephone care as well as aggressive medication management resources are limited. Simple screening tools when blood pressure or cholesterol levels are are available to help clinicians and health care unacceptably high. systems identify low-FHL patients.84, 85 Telephone counselors who are closely allied with patients’ primary care providers are well-suited to Patients with Multiple Chronic flag patients and schedule appropriate follow-up, Health Problems or even make changes in medication regimens. As many as 62 percent of Medicare patients have Counselors who work less directly with physi- multiple chronic illnesses. These patients can be cians may still be effective in coaching patients overwhelmed by their self-care needs, resulting in to seek appropriate care, serve as their own health negative consequences to their health. For exam- advocates, and monitor their own treatment ple, patients with diabetes and depression often quality. have poorer self-management and glycemic con- trol than those with diabetes alone, and a recent Value of Registries and study found that chronic pain was a common Service Targeting risk factor for poor diabetes self-care.86 Multiple chronic conditions serve as competing demands To provide effective telephone care services, on patients’ time and emotional resources, as well providers need some mechanism for identifying Using Telephone Support to Manage Chronic Disease | 19 the population they hope to serve. Disease reg- istries are one of the hallmarks of effective care management; patient identification on the basis of diagnoses is an important first step. However, some of the populations described above require more detailed data collection (e.g., surveys to identify patients with health literacy deficits). Other populations may require statistical analyses in order to identify the factors that predict poor outcomes, gaps in services, or preventable health care costs. Private telephone care vendors often develop sophisticated analytic techniques in order to identify their population targets and monitor their success in achieving program goals. Health plans and health systems also can develop these tools, but smaller providers may lack the infra- structure required to target telephone care pro- grams effectively. 20 | CALIFORNIA HEALTHCARE FOUNDATION V. Linking Telephone Care Services with Usual Processes of Care TELEPHONE CARE SERVICES THAT ARE TIGHTLY linked to clinic-based care are most effective. For example in the depression87 and heart failure management88 programs, telephone care providers regularly reported problems to the patients’ clinicians and worked closely with them to make changes in patients’ disease management plans. Similarly, in two studies of IVR-supported telephone nursing care for diabetes patients, close linkages were forged between care managers and patients’ regular providers.89, 90, 91 After 12 months, patients showed improvements in glycemic control, symptom burden, self-management behaviors, and use of guideline-recommended diabetes services. Nurse care managers (rather than the IVR support system) served as the primary source of patient counseling and the interface between patients and their primary care team. Integrating telephone support as part of the role of clinic-based nurses or other allied health staff may be an effective approach, since it minimizes the need to transfer patient records across physical locations, minimizes the number of clinicians involved in patients’ care, and increases the likelihood that patients’ physicians will seriously consider recommendations of tele- phone care providers. Although few studies have directly com- pared similar telephone care protocols delivered by clinicians that varied in their level and type of training, the study by Lee and colleagues92 indicates that recommendations made by telephone care nurses may be just as effective as those of physicians. At a baseline, telephone care programs should address the most glaring gaps in patients’ knowledge about how to work with their health system in managing their disease. This may include helping patients: I Understand their health coverage; I Know how to apply for assistance programs; and I Know basic administrative information, such as the name of their primary care provider, how to schedule appoint- ments with appropriate clinicians, and how to get health questions answered between outpatient visits. Using Telephone Support to Manage Chronic Disease | 21 In-house and Contracted Outside diabetes patients had little impact on patients’ Programs health status, largely because they worked sepa- Health systems and plans must decide whether rately from patients’ usual care providers, and to provide telephone care services in-house or their recommendations were often ignored by contract with an outside vendor. Some of the those clinicians.94 challenges with outsourced telephone care Some telephone care programs are delivered by services include coordinating patient care with employer groups or health plans who have less additional providers, sharing information contact with clinicians or patients’ clinical across organizational boundaries, and monitor- records. One way to increase these programs’ ing program success. On the other hand, several effectiveness is to coach patients to be better large vendors offer validated statistical algorithms advocates for quality care, rather than attempting or “analytics” for targeting telephone care based to influence outpatient clinicians directly. For on a health system’s goals (e.g., increasing guide- such services to be effective, providers need to line adherence or decreasing preventable admis- have access to enough updated and quality sions). Many large vendors also use proprietary health information to credibly counsel patients software to structure interactions with patients regarding service gaps and priorities for seeking and ensure that the program is targeted and follow-up care. efficient. Regardless of the location of the services, the most effective programs have structured com- puter supports to ensure that: all necessary assessments are conducted; findings are well documented; and the communication process is monitored over time by trained clinicians. Stand-alone Services Telephone care programs with weak linkages to patients’ usual care tend to be ineffective. In one recent study, investigators evaluated a call-in IVR counseling program designed to help patients increase their physical activity levels.93 Although the IVR calls used tailored, recorded messages based on sound health behavior change theory, one in four patients never called the toll-free number to receive behavior-change messages, and less than half were using the system after three months. Not surprisingly, the service had no significant impact on patients’ behavior. “Live” telephone care service providers may have better results, although they also may have limited impact if they are not integrated into patients’ overall care team. In a recent random- ized trial, telephone care managers for high-risk 22 | CALIFORNIA HEALTHCARE FOUNDATION VI. Evaluating Telephone Care Programs EVALUATIONS SHOULD REFLECT A PROGRAM’S primary aims. Trying to achieve both health outcome improve- ments and short-term cost-savings simultaneously may be unrealistic. Some health changes may take years to realize, and most evaluation plans lack the resources to detect these long- term benefits. Evaluating more direct outcomes of telephone care programs— as well as changes in patients’ health and resource use—can provide concrete measures of a program’s impact and point to areas for improvement (Table 2). For example, documenting the content of care sessions helps program managers determine whether some patients’ urgent needs divert too much attention from other patients or care requirements. The program’s rela- tionship with other clinical services should also be monitored. Patients may not benefit from telephone care if they cannot access recommended follow-up services due to limited system capacity or ability to pay. The impact on staff should be monitored. Programs need to be structured with appropriate caseloads and realistic expectations for the frequency and nature of patient contacts. If programs are not appropriately designed, telephone care providers can become frustrated, burned out, and less aggressive in address- ing care management problems. The traditional “gold standard” for evaluating clinical services, including telephone care, has been randomized controlled trials. However, health care systems often have not imple- mented services that have proven effective in a research context or have found disappointing results in real-world settings. Consequently, policymakers and clinicians have begun to evaluate telephone care and other programs using alternative frameworks that take a broader range of program characteristics into account, such as RE-AIM (Reach, Efficacy, Adoption, Implementation, and Maintenance). Using the RE-AIM framework, health system managers may find that telephone care programs are worthwhile, even when they have only a modest impact on patient outcomes.95 When health care systems outsource telephone care services, explicit outcome-based contracting is essential to ensure that a program is successful. Health plans may structure service agree- ments so that vendors are at some financial risk for achieving Using Telephone Support to Manage Chronic Disease | 23 defined program goals. In designing these con- tion reports, and how effectiveness will be deter- tracts, careful attention should be given to the mined in the context of other changes such as database for monitoring program success, includ- temporal trends. ing data quality, timeliness, the form of evalua- Table 2: Questions to Ask When Evaluating a Telephone Care Program The Patients • How many patients are enrolled? What is the average caseload for telephone care providers? • Are the characteristics of enrollees what was intended? • How many patients refuse to enroll? How are refusers different from enrollees? Telephone Care Process • How many days does it take for patients to be contacted by telephone care providers after enrollment? What proportion of patients are not contacted for more than a month? • What is the proportion of missed telephone care contacts? How does contact success rate vary across patient types and service providers? • How long are telephone contacts? Is the content of the conversations what was intended or do other press- ing patient needs take precedence during the calls? • How many telephone care contacts lead to communication between telephone care providers and other cli- nicians? How many lead to a request for in-person follow-up? • How many patients drop out of the program? What are the reasons for drop-out? How many patients are lost to follow-up? Patients’ Self-Care • What is the impact of telephone care on patients’ self-management behaviors, such as self-monitoring and medication adherence? • What is the program’s impact on lifestyle behaviors such as smoking, diet, and physical activity? Coordination with Usual Care • To what extent do telephone care providers draw on patient’s medical record as the basis for determining the content of telephone calls? • How often and what types of information from the calls are available to other service providers in standard outpatient records? Patients’ Health Status, Service Use, and Other Outcomes • What impact does telephone care have on key disease-specific measures of patients’ health? Which patients benefit the most and which do not benefit at all? • How do telephone care services affect use of: urgent care, general medicine and specialty outpatient care, inpatient care, and guideline-recommended disease management services? • How satisfied are patients with the telephone care service? How does receiving telephone care affect their satisfaction with health services more generally? Clinician Satisfaction • Do providers “burn out” when delivering telephone care? What is the turnover rate? • How confident are telephone care providers that they can determine patients’ health status over the phone? How comfortable are they in deciding who needed in-person follow-up? • How many hours per week do clinicians feel is a reasonable maximum for telephone care providers? What do they feel is a reasonable caseload? 24 | CALIFORNIA HEALTHCARE FOUNDATION VII. Conclusion and Recommendations TELEPHONE CARE REPRESENTS A BROAD PLATFORM for patient communication, and its benefits depend on how a health care system chooses to structure and support its program. Telephone care services can improve chronic disease management and health outcomes—if the program is well- designed, targets the right patient population, focuses on spe- cific goals, and closely links services to its regular outpatient care. Table 3 provides ideas for organizations to consider in designing an effective program. A number of conclusions and recommendations have emerged from the author’s research and experience, including the following: I Telephone counseling should be clearly structured and based on established behavior-change principles. I Programs should be designed with specific goals in mind, not aimed at multiple objectives. I Services should target patients who can most benefit from them. Patients with limited health literacy, multiple chronic illnesses, or gaps in their care may be the best candidates for telephone care programs. I Programs that draw from patient registries, electronic medical records, or claims databases may be most effective in identifying patients most in need of assistance. I The most effective programs are closely linked with regular outpatient care and clinician follow-up. I Regular screening and assessment tools can be useful to help telephone care providers determine whether they should intensify, reduce, or discontinue services. I Telephone peer-support programs deserve serious consid- eration by both health care systems and researchers. I Evaluation should incorporate concrete measures such as number and type of patient contacts as well as changes in patients’ health and resource use; these combined findings can provide an assessment of a program’s impact and point to ideas for improvement. As financial pressures on traditional health care services increase, especially within safety-net health care systems, tele- phone care services have potential to fill service gaps. Carefully Using Telephone Support to Manage Chronic Disease | 25 designed telephone care can also provide addi- care programs. Ultimately these efforts should tional, unique types of disease management sup- improve chronically ill patients’ treatment access port not available through traditional practice and outcomes. At the same time, new research models. Over the coming decade, clinicians and findings will help providers in their efforts to health care systems will establish more effective make high-quality chronic illness care more means of implementing and evaluating telephone cost-effective. Table 3: Dimensions of Telephone Care Programs and Considerations for Program Design Dimension Comments Which patients should receive Best target populations include: telephone care? G Patients with complex self-care regimens; G Patients with a disease associated with high rates of preventable adverse events; G Patients with limited health literacy; G Patients with multiple chronic illnesses; and G Patients with gaps in recommended care. Registries and other data sources that can identify patients experiencing gaps in care are helpful. How should telephone counselors G There is little evidence that more highly trained clinicians be trained? (physicians or nurse practitioners) provide more effective telephone care services. G Explicit counseling techniques such as motivational inter- viewing or cognitive behavioral therapy are helpful. Where should telephone counselors G Phone counselors who also have face-to-face contact be located? with patients in clinic may be especially effective. G Clinic-based staff may have greater influence with patients’ physicians. G When counselors are employed by a separate organiza- tion,they may be most effective if they focus on “ empowering” patients to be more active in their own medical management, rather than communicating directly with patients’ usual care providers. When should patients receive G Telephone care may be most valuable in the weeks and telephone care? months following a change in patient’s status (e.g., posthospitalization), self-management regimen, or diagnosis, as well as when patients experience significant gaps in care. How should telephone care G Evaluation should reflect program goals. Programs programs be evaluated? designed to increase access to outpatient care may not decrease overall resource use in the short term. G Programs should first be evaluated in terms of the telephone care process, then their effect on other care processes, and then on outcomes. Evaluating the impact on counselors is also important. 26 | CALIFORNIA HEALTHCARE FOUNDATION Acknowledgments Joanne A. Kimata, R.N., M.P.H., provided invaluable assistance in organizing the interviews with health system representatives and vendors of telephone care services, and in editing drafts of this report. Michele Heisler, M.D., M.P.A., contributed to the section on telephone-based peer-support services. The following individuals generously gave their time for interviews: American Healthways, Nashville, TN: Dr. Henriette Coetzer, Senior Director of Medical Integrity, and Heath Shackleford, Manager of Public Relations Blue Shield of California, San Francisco: Martin Detels, Senior Program Manager Eliza Corporation, Beverly, MA: Lucas Merrow, President and CEO, and Pamela Rosenberg, Vice President of Marketing and Business Development Health Dialog, Boston, MA: Dr. Patrick Mattingly, Chief Medical Officer, and Patricia Cmielewski, Vice President of Marketing LifeMasters Supported SelfCare, Inc., Irvine, CA: Christobel E. Selecky, Executive Chairman Using Telephone Support to Manage Chronic Disease | 27 Endnotes 1. Edwards MA, Patel AC. Telemedicine in the state of 14. Weinberger M, Kirkman MS, Samsa GP, Shortliffe EA, Maine: A model for growth driven by rural needs. Landsman PB, Cowper PA, et al. A nurse-coordinated Telemedicine Journal and E Health 2003;9(1):25-39. intervention for primary care patients with non-insulin- 2. Brodey BB, Claypoole KH, Motto J, Arias RG, Goss R. dependent diabetes mellitus: Impact on glycemic control Satisfaction of forensic psychiatric patients with remote and health-related quality of life. J Gen Intern Med 1995 telepsychiatric evaluation. Psychiatr Serv 2000; 51(10): Feb;10(2):59-66. 1305-7. 15. DeBusk RF, Miller NH, Superko HR, Dennis CA, 3. Aoki N, Dunn K, Fukui T, Beck JR, Schull WJ, Li HK. Thomas RJ, Lew HT, et al. A case-management system Cost-effectiveness analysis of telemedicine to evaluate dia- for coronary risk factor modification after acute myocar- betic retinopathy in a prison population. Diabetes Care dial infarction. Ann Intern Med 1994;120(721-729). 2004;27(5):1095-101. 16. See note 14. 4. McCue MJ, Hampton CL, Malloy W, Fisk KJ, Dixon L, 17. Aubert RE, Herman WH, Waters J, Moore W, Sutton Neece A. Financial analysis of telecardiology used in a D, Peterson BL, et al. Nurse case management to correctional setting. Telemedicine Journal and E Health improve glycemic control in diabetic patients in a health 2000;6(4):385-91. maintenance organization: A randomized controlled 5. Currell R, Urquhart C, Wainwright P, Lewis R. trial. Ann Intern Med 1998;129(8):605-12. Telemedicine versus face to face patient care: Effects on 18. George MR, O’Doud LC, Martin I, Lindell KO, professional practice and health care outcomes. The Whitney F, Ramondo T, Walsh L, et al. A comprehen- Cochrane Database of Systematic Reviews 2004;4. sive educational program improves clinical outcome 6. McBride CM, Rimer BK. Using the telephone to measures in inner-city patients with asthma. Arch Intern improve health behavior and health service delivery. Med 1999;159(15):1710-6. Patient Educ Couns 1999;37:3-18. 19. Riegel B, Carlson B, Kopp Z, LePetri B, Glaser D, 7. Smedley BD, Stith AY, Nelson AR, Eds. Unequal Unger A. Effect of a standardized nurse case-manage- Treatment: Confronting Racial and Ethnic Disparities in ment telephone intervention on resource use in patients Health Care: A Report of the Institute of Medicine. with chronic heart failure. Arch Intern Med 2002; 162(6): Washington DC: National Academy Press, 2002. 705-12. 8. Heisler M, Vijan S, Anderson RM, Ubel PA, Bernstein 20. Weinberger M , Tierney WM, Booher P, Katz BP. Can SJ, Hofer TP. When do patients and their physicians the provision of information to patients with osteoarthri- agree on diabetes treatment goals and strategies, and what tis improve functional status? A randomized, controlled difference does it make? J Gen Intern Med 2003 trial. Arthritis & Rheumatism 1998;32:1577-83. Nov;18(11):893-902. 21. Simon GE, Ludman EJ, Tutty S, Operskalski B, Von 9. Alexander GC, Casalino LP, Meltzer DO. Patient-physi- Korff M. Telephone psychotherapy and telephone care cian communication about out-of-pocket costs. JAMA management for primary care patients starting antide- 2003;290(7):953-8. pressant treatment. JAMA 2004;292(8):935-42. 10. Piette JD, Heisler M, Wagner TH. Cost-related medica- 22. Goldberg LR, Piette JD, Walsh MN, Frank TA, Jaski tion under-use: Do patients with chronic illnesses tell BE, Smith AL, et al. Randomized trial of a daily elec- their doctors? Arch Intern Med 2004;164(16):1749-55. tronic home monitoring system in patients with advanced heart failure: The Weight Monitoring in 11. Ford ES. Characteristics of survey participants with and Heart Failure (WHARF) trial. Am Heart J 2003 Oct; without a telephone: Findings from the Third National 146(4):705-12. Health and Nutrition Examination Survey. J Clin Epidemiol 1998;51 (1):55-60. 23. See note 5. 12. Ibid. 24. Wasson J, Gaudette C, Whaley F, Sauvigne A, Baribeau P, Welch HG. Telephone care as a substitute for routine 13. Anderson JE, Nelson DE, Wilson RW. Telephone clinic follow-up. JAMA 1992 Apr; 267(13):1788-93. coverage and measurement of health risk indicators: Data from the National Health Interview Survey. Am J 25. Welch HG, Johnson DJ, Edson R. Telephone care as an Public Health 1998;88:1392-5. adjunct to routine medical follow-up: A negative ran- domized trial. Eff Clin Pract 2000;3(3):123-30. 28 | CALIFORNIA HEALTHCARE FOUNDATION 26.Pinnock H, Bawden R, Proctor S, Wolfe S, Scullion J, 39. See note 22. Price, et al. Accessibility, acceptability, and effectiveness 40. See note 26. in primary care of routine telephone review of asthma: A pragmatic, randomised controlled trial. Br Med J 41. Piette JD, McPhee SJ, Weinberger M, Mah CA, 2003;326:477-9. Kraemer FB. Use of automated telephone disease man- agement calls in an ethnically diverse sample of low- 27. McAlister FA , Lawson FM, Teo KK, Armstrong PW. A income patients with diabetes. Diabetes Care 1999 Aug; systematic review of randomized trials of disease manage- 22(8):1302-9. ment programs in heart failure. American Medical Journal 2001;110(5):378-84. 42. Kobak KA, Taylor LH, Dottl SL, Greist JH, Jefferson JW, Burroughs D, et al. A computer-administered tele- 28. See note 19. phone interview to identify mental disorders. JAMA 29. DeBusk RF, Miller NH, Parker KM, Bandura A, 1997;278(11):905-10. Kraemer HC, Cher DJ, et al. Care management for low- 43. Millard RW, Carver JR. Cross-sectional comparison of risk patients with heart failure: A randomized, controlled live and interactive voice recognition administration of trial. Ann Intern Med 2004;141(8):606-13. the SF-12 Health Status Survey. Am J Manag Care 30. Laramee AS, Levinsky SK, Sargent J, Ross R, Callas P. 1999;5(153-159). Case management in a heterogeneous congestive heart 44. Reiber GE, Au D, McDonell M, Fihn SD. Diabetes failure population: A randomized controlled trial. Arch quality improvement in Department of Veterans Affairs Intern Med 2003;163(7):809-17. ambulatory care clinics: A group randomized clinical 31. Jerant AF, Azari R, Nesbitt TS. Reducing the cost of fre- trial. Diabetes Care 2004;27(Suppl 2):B61-B68. quent hospital admissions for congestive heart failure: A 45. Rubenstein LV, Calkins DR, Young RT, Cleary PD, Fink randomized trial of a home telecare intervention. Med A, Kosecoff J, et al. Improving patient function: A ran- Care 2001;39(11):1234-45. domized trial of functional disability screening. Ann 32. Weinberger M, Tierney WM, Cowper PA, Katz BP, Intern Med 1989;111(10):836-42. Booher PA. Cost-effectiveness of increased telephone 46.Schillinger D, Piette J, Grumbach K, Wang F, Wilson C, contact for patients with osteoarthritis: A randomized Daher C, et al. Closing the loop: Physician communica- controlled trial. Arthritis Rheum 1993;36(2):243-6. tion with diabetic patients who have low health literacy. 33. Karter AJ, Parker MM, Moffet HH, Ahmed AT, Ferrara Arch Intern Med 2003 Jan;163(1):83-90. A, Liu JY, et al. Missed appointments and poor glycemic 47. See note 8. control: An opportunity to identify high-risk diabetic patients. Med Care 2004;42(2):110-5. 48. Rost K, Carter W, Inui T. Introduction of information during the initial medical visit: Consequences for patient 34. Piette JD, Wagner TH, Potter MB, Schillinger D. follow-through with physician recommendations for Health insurance status, medication self-restriction due medication. Soc Sci Med 1989;28(4):315-21. to cost, and outcomes among diabetes patients in three systems of care. Med Care 2004;42(2):102-9. 49. Crane JA. Patient comprehension of doctor-patient communication on discharge from the emergency 35. LeBaron CW, Starnes DM, Rask KJ. The impact of department. J Emerg Med 1997;15(1):1-7. reminder-call interventions on low vaccination coverage in an inner-city population. Archives of Pediatric and 50. Williams MV, Baker DW, Honig EL, Lee TM, Nowlan Adolescent Medicine 2004;158(3):255-61. A. Inadequate literacy is a barrier to asthma knowledge and self-care. Chest 1998;114:1008-15. 36.Tanke ED, Leirer VO. Automated telephone reminders in tuberculosis care. Med Care 1994;32:380-9. 51. Ad Hoc Committee on Health Literacy for the Council on Scientific Affairs AMA. Health literacy: Report of the 37. Lieu TA, Capra AM, Makol J, Black SB, Shinefield HR. Council on Scientific Affairs. JAMA 1999;281(6):552-7. Effectiveness and cost-effectiveness of letters, automated telephone messages, or both for underimmunized chil- 52. Dudas V, Bookwalter T, Kerr KM, Pantilat SZ. The dren in a health maintenance organization. Pediatrics impact of follow-up telephone calls to patients after hos- 1998;101(4):E3. pitalization. Am J of Med, 2001;111(9B):26S-30S. 38. Leirer VO, Morrow DG, Tanke ED, Pariante GM. 53. Lando HA, Pirie PL, Roski J, McGovern PG, Schmid Elders’ nonadherence: Its assessment and medication LA. Promoting abstinence among relapsed chronic reminding by voicemail. Gerontologist 1991;(514-520). smokers: The effect of telephone support. Am J Public Health 1996;86(12):1786-90. Using Telephone Support to Manage Chronic Disease | 29 54. Reid RD, Pipe A, Dafoe WA. Is telephone counselling 68. Wheeler JA, Gorey KM, Greenblatt B. The beneficial a useful addition to physician advice and nicotine effects of volunteering for older volunteers and the peo- replacement therapy in helping patients stop smoking? ple they serve: A meta-analysis. Int J Aging Hum Dev A randomized controlled trial. Can Med Assoc J 1998;47(1):69-79. 1999;160(11):1577-81. 69. West DA, Kellner R, Moore-West M. The effects of 55. Zhu SH, Anderson CM, Tedeschi GJ, Rosbrook B, loneliness: A review of the literature. Compr Psychiatry Johnson CE, Byrd M, et al. Evidence of real-world effec- 1986 Jul-1986 Aug;27(4):351-63. tiveness of a telephone quitline for smokers. N Engl J 70. Schwartz CE, Sendor M. Helping others helps oneself: Med 2002;347(14):1106-9. Response shift effects in peer support. Soc Sci Med 1999 56.Anderson RM, Arnold MS, Funnell MM, Fitzerald JT, Jun;48(11):1563-75. Butler PM, Feste CC. Patient empowerment: Results of 71. Brown SL, Nesse RM, Vinokur AD, Smith DM. a randomized controlled trial. Diabetes Care 1995;18 Providing social support may be more beneficial than (7):943-9. receiving it: Results from a prospective study of mortal- 57. Greenfield S, Kaplan SH, Ware JE Jr, Yano EM, Frank ity. Psychol Sci 2003 Aug;14(4):320-7. HJ. Patients’ participation in medical care: Effects on 72. Van Willigen M. Differential benefits of volunteering blood sugar control and quality of life in diabetes. J Gen across the life course. J Gerontol B Psychol Sci Soc Sci Intern Med 1988 Sep-1988 Oct;3(5):448-57. 2000 Sep;55(5):S308-18. 58. Lee TJ, Baraff LJ, Wall SP, Guzy J, Johnson D, Woo H. 73. Keyserling TC, Samuel-Hodge CD, Ammerman AS, Parental compliance with after-hours telephone triage Ainsworth BE, Henriquez-Roldan CF, Elasy TA, et al. A advice: Nurse advice service versus on-call pediatricians. randomized trial of an intervention to improve self-care Clin Pediatr (Phila) 2003;42(7):613-9. behaviors of African-American women with type 2 dia- 59. Simon GE, VonKorff M, Rutter C, Wagner E. betes: Impact on physical activity. Diabetes Care 2002 Randomised trial of monitoring, feedback, and manage- Sep;25(9):1576-83. ment of care by telephone to improve treatment of 74. Arnstein P, Vidal M, Wells-Federman C, Morgan B, depression in primary care [see comments]. BMJBr Med Caudill M. From chronic pain patient to peer: Benefits J 2000 Feb;320(7234):550-4. and risks of volunteering. Pain Manag Nurs 2002 60. Unutzer J, Katon W, Callahan CM, Williams JW Jr, Sep;3(3):94-103. Hunkeler E, Harpole L, et al. Collaborative care man- 75. Rudy RR, Rosenfeld LB, Galassi JP, Parker J, Schanberg agement of late-life depression in the primary care set- R. Participants’ perceptions of a peer-helper, telephone- ting: A randomized controlled trial. JAMA 2002 based social support intervention for melanoma patients. Dec;288(22):2836-45. Health Commun 2001;13(3):285-305. 61. Resnicow K, DiIorio C, Soet JE, Ernst D, Borrelli B, 76. Colon Y. Telephone support groups: A nontraditional Hecht J. Motivational interviewing in health promotion: approach to reaching underserved cancer patients. It sounds like something is changing. Health Psychol Cancer Pract 1996 May-1996 Jun;4(3):156-9. 2002;21(5):444-51. 77. Kappen DM, Branscombe, NR. The effects of reasons 62. Bertakis KD. The communication of information from given for ineligibility on perceived gender discrimination physician to patient: A method for increasing patient and feelings of injustice. 40. 2001;40(Pt 2):295-313.Br J retention and satisfaction. J Fam Pract 1977;5:217-22. Soc Psychol 2001 Jun;40(Pt 2):295-313. 63. Schwartzberg J. Health Literacy Introductory Kit. 78. Stewart MJ, Hart G, Mann K, Jackson S, Langille L, American Medical Association, Chicago, IL 2000. Reidy M. Telephone support group intervention for 64. Ley P. Communicating with Patients: Improving persons with hemophilia and HIV/AIDS and family Communication, Satisfaction, and Compliance. New York, caregivers. Int J Nurs Stud 2001 Apr;38(2):209-25. NY: Croom Helm, 1988. 79. See note 75. 65. See note 46. 80. See note 51. 66.House JS, Landis KR, Umberson D. Social relationships 81. Weiss BD, Blanchard JS, McGee DL, Hart G, Warren and health. Science 1988;241:540-5. B, Burgoon M, et al. Illiteracy among Medicaid recipi- 67. Musick MA, Wilson J. Volunteering and depression: The ents and its relationship to health care costs. J Health role of psychological and social resources in different age Care Poor Underserved 1994;5(2):99-111. groups. Soc Sci Med 2003 Jan;56(2):259-69. 82. Schillinger D, Grumbach K, Piette J, Wang F, Osmond D, Daher C, et al. Association of health literacy with diabetes outcomes. JAMA 2002 Jul;288(4):475-82. 30 | CALIFORNIA HEALTHCARE FOUNDATION 83. Schillinger D, Bindman A, Stewart A, Wang F, Piette JD. Functional health literacy and the quality of physi- cian-patient communication among diabetes patients. Patient Educ Couns 2004;52(315-323). 84. Davis TC, Michielutte R, Askov EN, Williams MV, Weiss BD. Practical assessment of adult literacy in health care. Health Educ Behav 1998;25(5):613-24. 85. Chew LD, Bradley KA, Boyko EJ. Brief questions to identify patients with inadequate health literacy. Fam Med 2004;36(8):588-94. 86.Krein SL, Heisler M, Piette JD, Makki F, Kerr EA. The effect of chronic pain on diabetes patients’ self-manage- ment. Diabetes Care 2005;28(1):65-70. 87. See note 21. 88. See note 22. 89. Piette JD, Weinberger M, Kraemer FB, McPhee SJ. Impact of automated calls with nurse follow-up on dia- betes treatment outcomes in a Department of Veterans Affairs Health Care System: Arandomized controlled trial. Diabetes Care 2001 Feb;24(2):202-8. 90. Piette JD, Weinberger M, McPhee SJ. The effect of automated calls with telephone nurse follow-up on patient-centered outcomes of diabetes care: A random- ized, controlled trial. Med Care 2000 Feb;38(2):218-30. 91. Piette JD, Weinberger M, McPhee SJ, Mah CA, Kraemer FB, Crapo LM. Do automated calls with nurse follow-up improve self-care and glycemic control among vulnerable patients with diabetes? Am J Med 2000 Jan;108(1):20-7. 92. See note 58. 93. Pinto BM, Friedman R, Marcus BH, Kelley H, Tennstedt S, Gillman MW. Effects of a computer-based, telephone-counseling system on physical activity. Am J Prev Med 2002;23(2):113-20. 94. Krein SL, Klamerus ML, Vijan S, Lee JL, Fitzgerald JT, Pawlow A, et al. Case management for patients with poorly controlled diabetes: A randomized controlled trial. Am J Med 2004;116(11):732-9. 95. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact of health promotion interventions: The RE-AIM framework. Am J Public Health 1999;89: 1322-7. Using Telephone Support to Manage Chronic Disease | 31