United States General Accounting Office GAO Report to the Chairman, Subcommittee on Oversight and Investigations, Committee on Energy and Commerce, House of Representatives September 2003 MEDICARE Most Beneficiaries Receive Some but Not All Recommended Preventive Services GAO-03-958 September 2003 MEDICARE Most Beneficiaries Receive Some but Not Highlights of GAO-03-958, a report to the All Recommended Preventive Services Chairman, Subcommittee on Oversight and Investigations, Committee on Energy and Commerce, House of Representatives Medicare, the federal health Most Medicare beneficiaries receive some preventive services through their program insuring almost 35 million visits to physicians, but relatively few receive the full range of preventive beneficiaries age 65 and older, services available. Survey data showed, for example, that in 2000 about 30 covers certain preventive services, percent of beneficiaries did not receive a flu shot, and 37 percent had never such as flu shots and been vaccinated against pneumonia. Moreover, many Medicare beneficiaries mammograms. Most beneficiaries receive care through Medicare’s are apparently unaware that they may have conditions that preventive fee-for-service program, under services are meant to detect. For example, in a 1999–2000 nationally which they generally receive these representative survey during which people received physical examinations, services as part of visits to the nearly one-third of those age 65 and older who were found to have high doctor for specific illnesses or cholesterol measurements said they had not previously been told by a conditions. Other beneficiaries physician or other health professional that they had high cholesterol. receive services under Medicare’s Projected nationally, this percentage could represent 2.1 million people. managed care program, called Estimated Number of Medicare Beneficiaries Age 65 and Older Who Were Aware or Unaware Medicare + Choice. GAO was asked That They Might Have High Blood Pressure or High Cholesterol, 1999–2000 to determine (1) the extent to which beneficiaries received recommended preventive services through existing visits, (2) whether approaches used by Medicare + Choice plans provide insight for improving delivery of preventive care services for fee-for-service beneficiaries, and (3) what the Centers for Medicare & Medicaid Services (CMS) is doing to explore suggested options for delivering preventive care to fee-for-service beneficiaries. GAO’s work included analyzing data from four national health Note: About one-third of Medicare beneficiaries examined and found to have high cholesterol or surveys and reviewing five elevated blood pressure measurements were previously unaware that they might have the condition, representing millions nationwide. Medicare + Choice plans considered to have innovative No clear “best practice” approach to delivering preventive care stands out approaches to delivering among the innovative Medicare + Choice plans GAO studied. All five plans preventive services. GAO also identify health risks, provide feedback on risks to patients or their interviewed Department of Health physicians, and follow up to reduce those risks. But their follow-up and Human Services (HHS) and programs, approaches, and priorities differ, and little is known about the CMS officials and reviewed effectiveness of these efforts for the Medicare-age population. documents on CMS demonstrations related to preventive services. CMS has begun the development work to design a project evaluating the use of individual assessments of health risks, followed by counseling and other services, as a way to improve preventive care delivery. Another suggested approach—adding a routine physical examination benefit to Medicare’s fee- www.gao.gov/cgi-bin/getrpt?GAO-03-958. for-service program—could provide more opportunities, but at increased To view the full report, including the scope cost and without guarantee that preventive services would actually be and methodology, click on the link above. provided to Medicare beneficiaries. For more information, contact Janet Heinrich on 202-512-7250. HHS generally concurred with the findings of this report. Contents Letter 1 Results in Brief 3 Background 5 Most Beneficiaries Receive Some Preventive Services, but Not All That Are Recommended 7 Medicare + Choice Plans Reviewed Assess Health Risks Using Varying Approaches 11 New Ways to Improve the Provision of Preventive Services within Medicare’s Fee-for-Service Program Are Promising but Untested 17 Concluding Observations 24 Agency Comments 24 Appendix I Scope and Methodology 26 Appendix II Preventive Services Recommended by the U.S. Preventive Services Task Force or Covered by Medicare 30 Appendix III National Health and Nutrition Examination Survey Methodology and Results 32 Appendix IV Comments from the Department of Health and Human Services 34 Tables Table 1. Feedback Processes Described by Medicare + Choice Plans 14 Table 2: Four National Health Surveys with Preventive Services Data, 1999–2000 26 Table 3: Estimated Proportion of Fee-for-Service Physician Visits Made by People Age 65 and Older, by Major Reason for the Visits, 2000 27 Page i GAO-03-958 Medicare Preventive Services Table 4: Estimated Proportion of Fee-for-Service Physician Visits in Which Diet Counseling Services Were Provided or Ordered, by Major Reason for the Visits, 2000 28 Table 5: Estimated Proportion of Fee-for-Service Physician Visits in Which Blood Pressure Measurements Were Provided or Ordered, by Major Reason for the Visits, 2000 28 Table 6: Medicare + Choice Plans Included in GAO’s Study 29 Table 7: NHANES Data GAO Used to Determine if Participants Had Measures of Specific Health Conditions 32 Table 8: People Age 65 and Older in the United States Found to Have Measures of Specific Health Conditions, NHANES 1999–2000 33 Table 9: People Age 65 and Older in the United States Found to Have Measures of Specific Health Conditions and Who Reported They Had Not Previously Been Told They Might Have the Condition, NHANES 1999–2000 33 Figures Figure 1: Major Reasons for Physician Visits by Medicare Beneficiaries in the Fee-for-Service Program, 2000 8 Figure 2: Estimated Number of Medicare Beneficiaries Age 65 and Older Who Were Aware and Unaware That They Might Have High Blood Pressure or High Cholesterol, 1999–2000 11 Abbreviations AMA American Medical Association ACE Inhibitor Angiotensin-converting enzyme inhibitor BRFSS Behavior Risk Factor Surveillance Survey CDC Centers for Disease Control and Prevention CMS Centers for Medicare & Medicaid Services HHS Department of Health and Human Services NHANES National Health and Nutrition Examination Survey Td Tetanus-diphtheria Page ii GAO-03-958 Medicare Preventive Services This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Page iii GAO-03-958 Medicare Preventive Services United States General Accounting Office Washington, DC 20548 September 8, 2003 The Honorable Jim Greenwood Chairman Subcommittee on Oversight and Investigations Committee on Energy and Commerce House of Representatives Dear Mr. Chairman: Medicare, the federal government’s health insurance program that covers almost 35 million people age 65 and older, was created largely to help pay beneficiaries’ health care costs once they become ill or injured.1 For the most part, the federal government pays physicians and other health care providers to treat Medicare beneficiaries for illnesses and health conditions. In addition, the Congress has broadened Medicare coverage to include specific preventive services, aimed at either (1) keeping an illness or condition from developing or (2) keeping it from becoming more serious through early detection and subsequent management. Immunization against influenza (a “flu shot”) is an example of the first type of preventive service; a mammogram to detect breast cancer is an example of the second. Overall preventive care depends heavily on identifying health risks associated with the onset or progression of disease and taking steps to reduce or mitigate these risks. We previously reported to you that Medicare beneficiaries’ use of covered preventive services has increased over time but varies widely from service to service.2 In response, you asked us to follow up on several issues. One issue is the success of providing preventive services through a Medicare service delivery system based primarily on treating existing illnesses and health conditions. Under Medicare’s fee-for-service program, which enrolls about 84 percent of Medicare beneficiaries, no specific provision exists for 1 We focused our work on the people covered by Medicare who are 65 and older—about 86 percent of the entire Medicare population. Besides this age group, Medicare also covers about 5.8 million disabled persons younger than age 65. Throughout this report, except where otherwise noted, we use the term “Medicare beneficiaries” to refer only to those beneficiaries age 65 and older. 2 U.S. General Accounting Office, Medicare: Beneficiary Use of Clinical Preventive Services, GAO-02-422 (Washington, D.C.: April 2002). Page 1 GAO-03-958 Medicare Preventive Services a routine annual physical or checkup that could be a vehicle for delivering preventive services.3 Unless beneficiaries in the fee-for-service program have supplemental insurance that covers such a checkup, they may have to depend on receiving preventive services during their visits for specific illnesses or conditions, or during other visits for those specific preventive services that Medicare does cover. A second issue is what can be learned about the effectiveness of preventive service approaches put in place by plans that contract with Medicare to offer health care on a managed care basis.4 These plans, which enroll about 14 percent of all Medicare beneficiaries under an option known as Medicare + Choice, generally offer a benefit for periodic checkups.5 Some of these Medicare + Choice plans are regarded as particularly innovative in assessing risk, providing screening services, and conducting prevention programs. This report addresses the following questions: • Do Medicare beneficiaries receive recommended preventive services through existing physician visits? • What approaches for preventive care have been taken by selected Medicare + Choice plans, and what is known about their effectiveness for the Medicare beneficiaries they serve? • What delivery options for identifying and reducing health risks have been suggested for Medicare fee-for-service beneficiaries, and are any of these options being explored by the Centers for Medicare & Medicaid Services (CMS), the agency administering the program? Because no single source contained all the information we needed to assess the extent to which Medicare beneficiaries receive preventive services through existing physician visits, we analyzed data from four 3 “Fee-for-service” is the Medicare arrangement sometimes referred to as the original Medicare plan. Under this option, Medicare pays a health care practitioner for each visit or procedure received by a patient, and a beneficiary can visit any hospital, physician, or health care provider who accepts Medicare patients. Medicare pays a set percentage of the expenses, and the beneficiary is responsible for certain deductibles and coinsurance payments—the portion of the bill that Medicare does not pay. 4 These are health care options (like health maintenance organizations) in some areas of the country. In most programs, the beneficiary can go only to doctors, specialists, or hospitals on the program’s list. Programs must cover all Medicare part A and part B health care but can also cover extras, like prescription drugs and periodic checkups. 5 Besides the 84 percent of Medicare beneficiaries in fee-for-service and the 14 percent in Medicare + Choice (2002 data), a small percentage of Medicare beneficiaries receive services through such arrangements as prepaid group practice plans or Medicare demonstrations. Page 2 GAO-03-958 Medicare Preventive Services nationally representative health surveys. The Centers for Disease Control and Prevention’s (CDC) Behavioral Risk Factor Surveillance System asks a range of health questions over the telephone, including if respondents received a “routine checkup” within the past year. CMS’s Medicare Current Beneficiary Survey collects self-reported data, including whether respondents have received influenza or pneumonia immunizations. CDC’s National Health and Nutrition Examination Survey (NHANES) collects data on health conditions by means of both comprehensive health examinations and interviews, where patients self-report information, including whether a physician or other health professional has ever told them that they have a given health condition. Unlike the other surveys, which take a sample of the population, CDC’s National Ambulatory Medical Care Survey samples physician practices, collecting detailed information about office visits, including the major reason for the visit and which preventive services were ordered or provided. In addition, this survey captured information that allowed us to assess whether visits by Medicare beneficiaries were on a fee-for-service basis. Unless otherwise noted, however, the data we report generally included beneficiaries from both systems. To describe the approaches of selected Medicare + Choice plans in delivering preventive services, we assessed literature and interviewed national experts to identify plans that were considered innovative in preventive care. We then obtained information from five such plans: AvMed Health Plans, Group Health Cooperative, Highmark Blue Cross and Blue Shield, Kaiser Permanente, and Oxford Health Plans. Collectively, an estimated 1.2 million Medicare beneficiaries in 15 states plus the District of Columbia receive their health care under these plans. To determine suggested options for identifying and reducing health risks and what CMS is doing to assess them, we reviewed the results of past related research demonstrations and congressionally mandated studies and interviewed Department of Health and Human Services (HHS) and CMS officials and other experts. (App. I further describes our scope and methodology.) We conducted our work from October 2002 through August 2003 in accordance with generally accepted government auditing standards. Most Medicare beneficiaries receive some but not all recommended Results in Brief preventive services, although they typically visit a physician several times during a year. Our analysis of year 2000 data shows that nearly 9 in 10 Medicare beneficiaries visited a physician at least once that year, with a beneficiary making an average of six visits or more within the year. Preventive services are delivered during all types of visits—whether for Page 3 GAO-03-958 Medicare Preventive Services illnesses, health conditions, or nonillness care. Regardless of the reason for a visit, however, many beneficiaries did not receive recommended preventive services. In 2000, for example, about 30 percent of Medicare beneficiaries did not receive an influenza vaccination and 37 percent had never had a pneumonia vaccination, as recommended under current guidelines for people age 65 and older. Moreover, many Medicare beneficiaries may have conditions of potential concern that they are unaware of. For example, among the Medicare beneficiaries who participated in a nationally representative survey and were found through physical examinations to have high cholesterol, about one-third said they had not previously been told by a physician or other health professional that they might have this condition. Projected nationally, this percentage translates into about 2.1 million people age 65 and older. Although they differ from one another in approach and emphasis, the preventive care approaches of the Medicare + Choice plans we reviewed share common elements. In particular, their approaches screen enrollees to identify health risks and then provide a number of follow-up activities designed to reduce those risks. The plans generally use combinations of methods to ascertain needed preventive services, including periodic preventive visits, health risk questionnaires, and periodic assessments of medical claims and pharmacy data. All plans also have follow-up strategies to help beneficiaries obtain needed preventive services, although their strategies and priorities vary. Follow-up interventions include counseling programs to encourage behavioral change, cancer screening for early detection of disease, and programs to coordinate and manage chronic conditions such as diabetes and cardiovascular disease. Although some plans furnished us with data suggesting that their approaches hold promise, few had conducted a systematic evaluation of whether the approaches improved health outcomes or lowered health care costs. Those studies that do show a relationship between greater use of preventive services and improved health outcomes or cost savings are limited in terms of how their findings might be generalized to Medicare beneficiaries. Several options have been suggested for improving the provision of preventive services under Medicare’s fee-for-service program, each with its own advantages and disadvantages. Two options center on adding a new benefit for a nonillness-related examination, specifically either (1) a one-time “welcome-to-Medicare” examination for new beneficiaries or (2) a periodic examination benefit for all beneficiaries. Coverage of a one- time or periodic wellness examination could be easily administered, and the examination could provide an opportunity for beneficiaries to receive Page 4 GAO-03-958 Medicare Preventive Services some preventive services. Adding such a benefit, however, could increase Medicare costs and still not guarantee that beneficiaries receive the preventive services they need. The results of a past CMS demonstration indicate that offering Medicare beneficiaries packages of broad-based preventive services has not consistently improved health or lowered hospital and other costs. As a result, CMS has recently considered an alternative option that would essentially create a different structure using nonphysician providers to assess health risks and ensure the delivery of preventive services within the fee-for-service program. The agency has started the development work to design a project to examine whether assessments of individual health risks, combined with continued counseling and follow-up services provided by nonphysicians, will improve delivery of preventive services and beneficiary health. CMS also has under way several other demonstration projects related to preventive care in the fee-for-service program, such as a smoking cessation program tailored to Medicare beneficiaries. Results from these demonstration efforts are not expected for several years. HHS reviewed a draft of this report and generally concurred with the findings. Many of the health conditions that people age 65 and older experience are Background preventable and linked to specific health risks. Some health risks are difficult to change, and some, such as a hereditary predisposition for a given disease, cannot be changed. For these, preventive services such as cancer screens can help identify disease in its early stages so that people can be referred to other services that can help manage or treat the disease. Other health risks, such as complications from influenza, can be successfully reduced by targeted preventive services. For example, studies show that immunizations against influenza can prevent thousands of hospitalizations and deaths each year among those age 65 and older. Health risks such as high blood pressure and high cholesterol are also considered health conditions because, if left alone, they can develop into potentially more significant conditions, such as cardiovascular disease, or lead to stroke. The term preventive care covers a wide spectrum of actions aimed at reducing risks for deteriorating health and improving the detection and management of disease. Generally, preventive care is intended for three purposes: Page 5 GAO-03-958 Medicare Preventive Services • To prevent a health condition from occurring at all. Vaccinations and physical activity to reduce the risk of heart disease, for example, qualify as this first type of preventive care (termed primary prevention). • To prevent or slow a condition’s progression to more significant health conditions by detecting a disease in its early stages. Mammograms to detect breast cancer and other screens to detect disease early are examples of this second type of preventive care (termed secondary prevention). • To prevent or slow a condition’s progression to more significant health conditions by minimizing the consequences of a disease. Care coordination and self-management of an existing disease, such as diabetes or asthma, are examples of this third type of preventive care (termed tertiary prevention). Many people associate the idea of preventive care with annual physical examinations, or “routine checkups,” by a family doctor, a practice first proposed by the American Medical Association (AMA) in the early twentieth century. In the early 1980s, however, the AMA determined that appropriate preventive care depends on an individual’s age and particular health risks, not simply on the results of a standard battery of tests.6 To evaluate preventive care for different age and risk groups, HHS in 1984 established a panel of experts called the U.S. Preventive Services Task Force. At present, the task force recommends certain screening, immunization, and counseling services for people age 65 and older (see app. II). Medicare covers some, but not all, of the task force-recommended preventive services (see comparison in app. II). Medicare’s fee-for-service program—which comprises approximately 84 percent of Medicare beneficiaries—does not cover periodic checkups, where clinicians might assess an individual’s health risk and provide needed preventive services. These Medicare beneficiaries may, however, receive some of these services during office visits for other health problems. Under Medicare + Choice, which covers about 14 percent of Medicare beneficiaries, a benefit for periodic checkups generally does exist. 6 The annual physical examination of healthy persons, in which a standard set of tests and procedures is performed, was first proposed by the AMA in 1922. For many years afterward, health professionals recommended routine physicals and comprehensive laboratory testing as effective preventive medicine. But in 1983, the AMA withdrew its support for a standard annual examination. Instead, the organization supported periodic visits in which patients receive preventive services depending upon the individual’s unique combination of age, sex, and health risk. Page 6 GAO-03-958 Medicare Preventive Services Medicare beneficiaries typically visit a physician several times during a Most Beneficiaries year and most receive some preventive services, but most do not receive Receive Some the full range of recommended services. Based on 2000 survey data and U.S. Bureau of the Census estimates of people age 65 and older, we Preventive Services, estimate that beneficiaries visit a physician at least six times a year, on but Not All That Are average, mainly for illnesses or medical conditions.7 About 1 in 10 visits occurred when beneficiaries were well, and most Medicare beneficiaries Recommended reported having what they considered to be a “routine checkup” in the previous year. The purposes of these routine checkups and the specific services that are delivered during these visits, however, remain unknown. Many Medicare beneficiaries did not receive recommended preventive services, such as influenza and pneumonia immunizations. Moreover, another national survey indicated that a substantial share of Medicare beneficiaries who were at risk for a condition that preventive services are meant to identify said that they had not been told by a health professional that they might have that condition. Medicare Beneficiaries In 2000, 88 percent of Medicare beneficiaries reported that they visited a Visit Physicians Often, and physician at least once that year.8 On the basis of data from CDC’s Most Report Receiving National Ambulatory Medical Care Survey, we estimate that, on average, beneficiaries visit physicians at least six times a year.9 Almost 9 in 10 visits Routine Checkups made by beneficiaries in the fee-for-service program were to treat illnesses or health conditions: more than half the visits targeted preexisting (chronic) problems, more than one-fourth targeted illnesses of sudden or recent onset (acute), and about 10 percent of visits took place pre- or postsurgery or to follow up after injuries. Only about 10 percent of visits 7 The surveys and other data sources from which we developed our information generally did not disaggregate the information into beneficiaries receiving care through fee-for- service and beneficiaries receiving care through Medicare + Choice programs. As a result, unless otherwise noted, the data reported include beneficiaries from both groups. 8 CMS’s Medicare Current Beneficiary Survey, 2000. 9 To estimate the average number of physician visits, we used data from the National Ambulatory Medical Care Survey and the U.S. Bureau of the Census. See app. I for a description of our methodology. We believe that the result is a conservative estimate of the average number of physician visits, since the segment of the survey that we analyzed excluded visits made in hospital outpatient and emergency departments or other institutional settings and also excluded physicians in the specialties of anesthesiology, pathology, and radiology. Page 7 GAO-03-958 Medicare Preventive Services dealt with nonillness care when the patient was considered healthy (see fig. 1).10 Figure 1: Major Reasons for Physician Visits by Medicare Beneficiaries in the Fee- for-Service Program, 2000 53% Chronic problem (Routine and flare-up) 26% Acute problem 10% Pre- and postsurgery or injury follow-up 10% Nonillness care Unknown 2% Care for specific conditions Source: CDC's National Ambulatory Medicare Care Survey, 2000. Note: Numbers do not add to 100 percent due to rounding. The survey defined an “acute problem” as a condition or illness of sudden or recent onset, a “chronic problem” as a preexisting long-term or recurring condition or illness, and “nonillness care” as a general health maintenance examination or routine periodic examination of a presumably healthy person. For chronic problems, the survey reported results separately for “routine chronic problems” and for “chronic problem flare-ups.” We combined these results in this figure. The separate results are found in app. I. Even though the majority of visits to physicians are for treating illness or health conditions, most Medicare beneficiaries reported receiving routine checkups. In CDC’s 2000 Behavioral Risk Factor Surveillance System Survey, for example, 93 percent of respondents age 65 and older reported that they had received a “routine checkup” within the previous 2 years. 10 Because Medicare’s fee-for-service program does not cover routine physical examinations but does cover some preventive services, such as immunizations and certain cancer screening tests, it is possible that some of the nonillness visits in 2000 were to obtain such services. In addition, some fee-for-service beneficiaries may be paying for nonillness examinations through other means, such as employer-provided or other supplemental insurance. According to CMS’s Medicare Current Beneficiary Survey, in the year 2000 about 41 percent of Medicare fee-for-service beneficiaries had insurance from former employers to supplement their basic Medicare benefit. Page 8 GAO-03-958 Medicare Preventive Services This survey did not, however, provide information on which specific services were delivered during those checkups. Indeed, as the following section shows, few beneficiaries receive all recommended services, although they receive some preventive services during visits when they are healthy as well as during visits to treat illnesses or health conditions. Despite Frequency of Despite how often Medicare beneficiaries visit physicians, many of them Visits, Many Medicare do not receive a full complement of recommended preventive services, Beneficiaries Do Not including some recommended by the U.S. Preventive Services Task Force and currently covered by Medicare. As we reported earlier, use of specific Receive the Full Range of preventive services varies widely by service.11 Although each preventive Recommended Preventive service we reviewed was delivered to a majority of Medicare beneficiaries, Services relatively few beneficiaries received the full range of preventive services. For example, 91 percent of female Medicare beneficiaries received at least one preventive service, but only 10 percent were screened for cervical, breast, and colon cancer and also immunized against influenza and pneumonia.12 Our analysis of additional data since our previous report shows that many Medicare beneficiaries still do not receive certain recommended preventive services. The task force recommends, for example, that all people age 65 and older receive an annual influenza vaccination and at least one pneumonia vaccination. In CMS’s Medicare Current Beneficiary Survey of 2000, however, about 30 percent of Medicare beneficiaries did not receive an influenza vaccination, and 37 percent had never had a pneumonia vaccination. Survey data showing the services provided during office visits indicate that Medicare beneficiaries do receive some preventive services during visits when they are ill or being treated for a health condition, and services are delivered at comparable rates during all types of visits, whether for nonillness care or for treating acute or chronic conditions. Beneficiaries in the fee-for-service program receive preventive services, such as cholesterol and blood tests, during visits when they are healthy and during visits to treat acute or chronic health conditions. Some tests are typically provided or ordered slightly more often during visits for nonillness care. In 11 GAO-02-422. 12 In January 2003, the U.S. Preventive Services Task Force released new recommendations for the use of pap smears to screen for cervical cancer. The task force now “recommends against screening women 65 and older who have had adequate recent screenings with normal Pap smears and are not otherwise at increased risk for cervical cancer.” Page 9 GAO-03-958 Medicare Preventive Services 2000, for example, blood tests for anemia13 were provided in about 16 percent of visits for nonillness care, compared with 7 percent of visits for chronic problems and 5 percent of visits for acute conditions. Other preventive services were provided at similar rates during the different types of visits. For example, we estimate that blood pressure measurement, a clinical screen for conditions such as hypertension, was done during 56 to 62 percent of visits, depending on the type of visit. Diet counseling services were provided during 13 to 20 percent of visits, depending on the type of visit.14 Many Beneficiaries May Be Many Medicare beneficiaries may not know that they are at risk for health Unaware of Their Risk for conditions that preventive care could detect—strong evidence that they Health Conditions That may not be receiving the full range of recommended preventive services.15 For example, data from CDC’s NHANES for 1999–2000 show that, of Preventive Care Is Meant beneficiaries participating in this nationally representative survey who had to Detect a physical examination and were found to have elevated blood pressure readings at the time of the examination, 32 percent reported that no physician or other health professional had ever told them about the condition. On the basis of this survey, we estimate that, during the period when the survey was conducted, 21 million Medicare beneficiaries may have been at risk for high blood pressure, and an estimated 6.6 million of them may have been unaware of this risk. Similarly, 32 percent of those found in the 1999–2000 survey to have a high cholesterol level reported 13 Anemia is a condition in which the blood is deficient in red blood cells, hemoglobin, or total volume. The hematocrit/hemoglobin test is used to test for anemia and to measure the concentration of packed red blood cells and hemoglobin in the blood. Hemoglobin is an iron-containing respiratory pigment in red blood cells that helps transport oxygen from the lungs to the body tissues. 14 Specifically, blood pressure measurements were provided at 56 percent of visits for acute problems, 59 percent of visits for chronic problems, and 62 percent of nonillness visits. Diet counseling services were provided at 13 percent of visits for acute problems, 20 percent of visits for chronic problems, and 18 percent of nonillness visits. For both blood pressure measurement and diet counseling service estimates, the differences in these percentages were not statistically significant at the 95 percent confidence level. See app. I for a discussion of the methodology and specific results. Source: CDC’s National Ambulatory Medical Care Survey, 2000. 15 The source of data for this statement was CDC’s National Health and Nutrition Examination Survey of 1999-2000. This survey oversampled—that is, included a larger number of persons age 60 and older in the sample, providing for a sample size that enabled us to focus our analysis specifically on the Medicare-age population for selected conditions. App. III contains a description of this survey and the specific results of our analyses. Page 10 GAO-03-958 Medicare Preventive Services that no one had told them that they had high cholesterol. Projected nationally, this percentage translates into 2.1 million Medicare beneficiaries (see fig. 2). Figure 2: Estimated Number of Medicare Beneficiaries Age 65 and Older Who Were Aware and Unaware That They Might Have High Blood Pressure or High Cholesterol, 1999–2000 14.4 High blood pressure 6.6 5.0 High cholesterol 2.1 0 2 4 6 8 10 12 14 16 Estimated number of Medicare beneficiaries (in millions) Told by physician or health professional Not told by physician or health professional Source: CDC’s National Health and Nutrition Examination survey. Note: CDC’s NHANES measured blood pressure three or four times during its 1-day physical examination. For our analysis, we calculated the average of the blood pressure measurements and applied CDC’s definition of high blood pressure: that is, a patient’s having an average systolic blood pressure equal to or greater than 140, or an average diastolic blood pressure equal to or greater than 90, or a patient who reported taking hypertension medication. CDC defined high cholesterol as a total cholesterol level equal to or greater than 240. The Medicare + Choice plans we reviewed vary in their specific strategies Medicare + Choice for delivering preventive services, but several common themes emerge Plans Reviewed from their efforts. First, nearly all identify members’ health risks and inform them or their providers about specific services that might be Assess Health Risks needed. For example, some plans mail questionnaires to members, seeking Using Varying information, such as when certain screening tests were last performed; other plans review claims and prescription data to identify at-risk Approaches members who might need a screening test or other preventive service. Second, all plans have follow-up strategies to help beneficiaries obtain needed preventive services, although their strategies and priorities vary. Third, while limited data provided by some plans suggest promising results, most plans have not evaluated the degree to which their strategies improve health outcomes or affect health care costs for Medicare beneficiaries. Page 11 GAO-03-958 Medicare Preventive Services Plans Use a Combination Although all the Medicare + Choice plans we reviewed use questionnaires of Ways to Identify Health to meet the requirement that they conduct health assessments for newly Risks enrolled Medicare beneficiaries,16 they use a combination of approaches to identify health risks. The particular risks that plans seek to identify vary from plan to plan. Risks include those associated with depression or lack of physical activity; risks from not obtaining recommended immunizations or screenings, such as mammography; and more general risk of short-term hospitalization or illness.17 For example, Group Health Cooperative, Highmark Blue Cross and Blue Shield, and Kaiser Permanente use questionnaire information to calculate a risk score meant to represent each enrollee’s probability of using health services heavily in the future. From its questionnaire, Kaiser Permanente also calculates the probability of 3-year survival for enrollees who have an existing advanced illness, as well as the probability that they will become dependent on others for daily care or need nursing home services during the next year (a condition Kaiser Permanente officials refer to as frailty). Oxford Health Plan, on the other hand, analyzes questionnaire data to assign enrollees a risk classification of high, moderate, or low and assigns patients to health management teams or programs appropriate for each risk level. For existing members, plans use slightly different approaches to identify health risks, including information from claims and pharmacy data, annual risk assessment questionnaires, physician visits, and computer systems (called registries) that indicate when patients require specific preventive services. The specific approaches vary from plan to plan. For instance, Group Health Cooperative officials reported that they review the health risks, such as the immunization status, of their existing members through health maintenance visits, which they encourage Medicare beneficiaries to have every 2 years. During this visit, the provider reviews responses to a completed questionnaire that each patient is asked to bring to the visit and updates computer registry data, compiled from previous risk assessment questionnaires and physician visits. AvMed conducts a health risk 16 Medicare + Choice plans are required to make a “best effort attempt” to assess newly enrolled Medicare beneficiaries. 42 C.F.R. § 422.122(b)(4)(i) (2002). 17 The risk assessment questionnaires for some plans are as brief as a one-page form, while others are as long as eight pages. A number of questions focus on identifying functional status, such as the ability to bathe independently; immunization status; current use of prescription medications; the history of screening tests, such as mammography; past health care use, such as the number of times enrollees saw their primary care physician in the preceding 6 months; behavior risks, such as smoking; and past illnesses or existing health conditions. Page 12 GAO-03-958 Medicare Preventive Services assessment for each of its Medicare members and also uses claims and pharmacy data to identify members with specific diseases, so as to target preventive services. For example, using pharmacy and claims data to identify people with diabetes, AvMed invites these members to a health fair featuring services to prevent further progression of the disease. Paying a single copayment to attend the health fair, members can receive a number of services, such as a blood draw for laboratory work and vision and glaucoma screening. Finally, some plans report that they have increased the use of specific preventive services through their participation in CMS-required national performance improvement projects.18 For example, Highmark reported that in 2002 the plan used medical claims data to identify female Medicare beneficiaries who had not received a mammogram within the past 2 years and notified the beneficiaries and their physicians. As a result, the officials reported that 60 percent of contacted beneficiaries went on to receive mammograms. Plans Use a Variety of After identifying the health risks of Medicare beneficiaries—whether new Follow-up Means to enrollees or existing members—plans we contacted reported that they Reduce Identified Risks also make efforts to follow up on that information by providing feedback to enrollees about risks and referring them to specific, risk-related preventive services. For example, all plans have approaches to prevent disease progression for individuals identified as having chronic health conditions. The plans sometimes differ in their types of follow-up and in their emphasis on different types of preventive services. Some plans we reviewed, for example, stress primary prevention activities, such as exercise programs for all members, to a greater degree than others. To provide feedback, many plans contact members directly through letters or phone calls, encourage contact with primary care physicians, or combine written or oral feedback with follow-up physician examinations (see table 1). 18 CMS generally requires each Medicare + Choice plan to undertake one national quality assessment and performance improvement project per year to measure and improve its own performance in a CMS-defined national focus area. Past national focus areas include improving diabetes care and increasing vaccination rates for influenza and pneumonia. Page 13 GAO-03-958 Medicare Preventive Services Table 1. Feedback Processes Described by Medicare + Choice Plans Health plan Feedback process Group Health Using data available on computer registry, health professionals can Cooperative review specific health risks with members. Health professionals also monitor the computer registry to track services members use. Kaiser For new enrollees, physicians review a summary report and provide Permanente feedback during an initial office visit. In San Diego, existing members who visit health assessment centers receive a letter, based on a completed questionnaire and tests estimating “health age,” that discusses ways of decreasing specific health risks, and they receive a second visit for a complete exam. Oxford Health Various departments receive health risk reports based on risk Plans assessment questionnaires. Reports for high-risk members go to teams of registered nurses, who contact the members and their primary care physicians to coordinate care. Highmark Blue Plan sends results of health risk assessment to physicians to facilitate Cross and Blue discussion with patients. Members with risks related to smoking, heart Shield disease, or osteoporosis receive letters. New members identified as at risk for being frail are referred to case managers, and members identified with chronic disease are referred to a condition management program for targeted interventions. AvMed Health Physicians receive health risk information from risk assessment Plans questionnaires and pharmacy and claims data. Members identified as having specific risks are contacted directly by the plan if health promotion or disease management programs are available for them. Source: Plan officials and plan documents. In addition to educating members about their health risks, some plans also link members to specific preventive services to reduce or mitigate these risks. For example, plans may send targeted health promotion materials; offer 24-hour telephone access to a nurse to discuss health concerns; or offer access to fitness programs, nutrition courses, immunizations, exams, and disease management or care coordination programs. These care coordination programs resolve health care issues through various means, such as in-depth telephone evaluations, communication with primary care physicians, in-home visits, or connections with community resources like Meals on Wheels. To refer Medicare members to preventive services, one plan we contacted emphasized directing them to primary prevention services, such as physical activity programs, while another plan emphasized connecting members to tertiary prevention services, such as disease management programs. For example, identifying physical activity and social isolation as two important predictors of overall health outcomes for seniors, Group Health Cooperative refers Medicare members to physical activity benefits Page 14 GAO-03-958 Medicare Preventive Services and other primary prevention services. In contrast, acknowledging that most individuals age 65 or older have more than one chronic health condition, AvMed focuses more on identifying members with existing conditions and referring them to preventive services that can mitigate the condition. AvMed has created eight disease management programs covering conditions such as congestive heart failure, asthma, and diabetes. The goal is to provide members having these conditions with a series of condition-specific care interventions. For example, interventions for AvMed enrollees in the congestive heart failure program include prescribing specific drugs (such as ACE19 inhibitors, diuretics, and beta- blockers), providing self-directed care plans, and monitoring weight. Some plans described how they track the success of their efforts to provide people with specific preventive care interventions. Highmark, for example, offers financial incentives to physicians who follow specific clinical guidelines for a given condition. The plan also gives physicians quarterly report cards, generated by a computer registry, that indicate whether their patients have received all the care recommended by the management programs in which the patients are enrolled. AvMed, on the other hand, tracks the number of members identified as eligible for specific disease management programs, whether the program was offered to all eligible members, and the number who enrolled. AvMed also reported setting, monitoring, and reporting on performance goals for the percentage of members receiving specific care interventions. For example, for enrollees in the congestive heart failure management program, AvMed tracks the percentage receiving an ACE inhibitor drug. Assessments of Health Few of the health plans we contacted had specifically evaluated whether Outcomes or Cost Savings their approaches to risk identification and reduction lead either to for Medicare Beneficiaries improved health outcomes for Medicare beneficiaries or to cost savings for the plan. From those plans that have such information, the available Are Limited data suggest that offering disease management programs to people who have existing health conditions may hold promise, but most plans lacked evidence from controlled studies of a specific benefit to their Medicare members. AvMed and Oxford are among the plans that have evaluated whether their approach improves health outcomes and saves money. For example, 19 Angiotensin-converting enzyme. Page 15 GAO-03-958 Medicare Preventive Services AvMed plan officials observed that, in all AvMed plans, including its Medicare + Choice plan, AvMed members with existing chronic conditions spent fewer days in the hospital during the same period when more of their members with existing conditions were enrolled in disease management programs. According to AvMed officials, between 2001 and 2002, shorter hospital stays of Medicare congestive heart failure patients led to total savings of $1 million, and shorter hospital stays of asthma patients from all plans (not limited to Medicare beneficiaries) led to savings of $400,000. Similarly, Oxford has estimated savings attributed to various interventions, such as a mean savings of $219 per member per month from Medicare beneficiaries who voluntarily participated in a self- management workshop for diabetes, as compared with a random group of diabetic members who did not attend the workshop. Although these findings show potential to improve health and decrease costs, it is unclear from this information whether the decreased length of hospitalization and cost savings resulted from disease management or from other factors. It is also not clear what the long-term effects may be on Medicare beneficiaries and whether these observations would also apply to beneficiaries in a fee- for-service environment. Some plans are evaluating specific aspects of their approaches as a first step in determining which approaches are effective. For example, Kaiser Permanente officials provided data demonstrating their ability to identify a certain type of health risk among Medicare beneficiaries, but they did not provide data demonstrating that their overall approaches to risk identification or risk reduction resulted in improved health outcomes or cost savings.20 Specifically, they found that three questions on the risk assessment questionnaire, along with the patient’s age, predicted with a high degree of accuracy whether a person would need daily assistance from another person during the following year. Kaiser identified these people as at risk for frailty and through additional study found that, over the next decade, frail people spent more days in nursing homes than individuals who were not frail.21 Kaiser Permanente officials told us that they have not identified interventions that decrease or prevent frailty from 20 Specifically, over the next decade, people designated as “frail” spent 800 percent more days in nursing homes than individuals who were not frail. K.K. Brody, R.E. Johnson, and L.D. Ried, “Evaluation of a Self-Report Screening Instrument to Predict Frailty Outcomes in Aging Populations,” The Gerontologist, 37 (1997): 182–191. 21 K.K. Brody et al., “A Comparison of Two Methods for Identifying Frail Medicare-Aged Persons,” Journal of American Geriatrics Society, 50 (2002): 562–569. Page 16 GAO-03-958 Medicare Preventive Services developing but were instead focusing on identifying interventions to improve outcomes for those people once they were identified as frail.22 In addition to reviewing the efforts of contacted Medicare + Choice plans, we reviewed several studies that evaluated the effectiveness of employer- sponsored approaches to providing preventive services, such as health risk assessment and feedback, to both employees and retirees. Although these studies conclude that employer-sponsored approaches hold promise in terms of increasing preventive services, improving health outcomes, and lowering cost, we found the results limited in how they might be generalized to all Medicare beneficiaries. For example, General Motors evaluated its companywide prevention program, which offered health risk assessments, individualized health profiles, a quarterly newsletter, a self- care book, and a toll-free health information line. The company reported that providing risk assessment and feedback helped participants lower their health risk status and that nearly half of this benefit was realized within the first of 5 years. Although General Motors provides a similar risk appraisal program to retirees, this study did not include them, so the study’s finding cannot be generalized to the Medicare population. Several options have been suggested for improving the provision of New Ways to Improve preventive services within Medicare’s fee-for-service program. They the Provision of include adding a new benefit for a nonillness-related examination, either a one-time “welcome-to-Medicare” examination for new beneficiaries or an Preventive Services examination available to all beneficiaries on a periodic basis. Although within Medicare’s covering a one-time or periodic nonillness examination could be easily administered and could increase the receipt of some preventive services, Fee-for-Service doing so could also increase Medicare costs without necessarily ensuring Program Are that beneficiaries receive the full range of preventive services. CMS has Promising but tested similar options in the past and found that they produced mixed results. It is now examining an alternative that would essentially create a Untested different structure using nonphysician providers to assess health risks and connect individuals with preventive services. The design work will be completed at the end of 2003, and if the decision is made to conduct a demonstration, results would not be available for several years after that. Additional demonstrations also under way—such as one exploring 22 Once frail people are identified, for example, Kaiser encourages medical providers to follow guidelines intended to detect conditions such as depression and to prevent outcomes such as injuries from falls. Page 17 GAO-03-958 Medicare Preventive Services effective smoking cessation approaches and one giving physicians incentives to coordinate and manage the overall health care needs of beneficiaries—may provide additional insights into coordinating and delivering appropriate preventive services within the Medicare fee-for- service program. Two Proposed Options A one-time “welcome-to-Medicare” examination for new beneficiaries has Center on Adding a been proposed as a means to better ensure that health care providers have Preventive Examination to enough time to identify individual Medicare beneficiaries’ health risks and provide preventive services appropriate for their risks.23 Proponents assert the Medicare Fee-for- that a one-time benefit could combine a health evaluation with screenings Service Program and immunizations, along with counseling about health promotion and disease prevention. It could also orient new beneficiaries to Medicare and encourage them to make informed choices about providers and plans. Health risk assessment and behavior counseling could be provided by a range of nonphysician professionals, including nurses, counselors, and dietitians. A similar option would have Medicare cover an annual or periodic preventive visit available to all fee-for-service beneficiaries. In theory, many of the advantages of a one-time preventive visit would also apply to periodic examinations. For instance, dedicated preventive visits might provide greater opportunities for health care providers to assess and address health risks. Some evidence also suggests that a periodic health examination may increase use of preventive cancer screening and counseling services. For example, a National Cancer Institute-supported study surveyed general internists and family physician practices and their patients in 1992 and found that patients who had received a periodic 23 Partnership for Prevention, A Better Medicare for Healthier Seniors: Recommendations to Modernize Medicare’s Prevention Policies (Washington, D.C.: Partnership for Prevention, 2003), and Gilbert S. Omenn, “Historical and Current Policy Issues in Establishing Coverage for Clinical Preventive Services under Medicare,” cited in the Partnership for Prevention’s report. Page 18 GAO-03-958 Medicare Preventive Services health examination within the previous year were substantially more likely to have received appropriate cancer screening and counseling.24 While these options have benefits, they also have potential drawbacks. Adding a benefit for a one-time or periodic examination to the Medicare fee-for-service package could increase the program’s costs without necessarily ensuring that beneficiaries receive the full range of preventive services. The Congressional Budget Office in June 2002 estimated that a one-time physical examination benefit for new enrollees could cost as much as $1.6 billion over the 2003–2012 period.25 According to a Congressional Budget Office official, the agency has not recently estimated the potential costs of a Medicare benefit for examinations provided on a periodic basis. This cost, however, would likely be substantially higher than that of a one-time visit for new beneficiaries. At the same time, establishing such a benefit would not necessarily ensure delivery of the full range of preventive services. In addition, primary care physicians typically cannot provide services such as mammography screenings for breast cancer and colonoscopies for colon cancer, because these services usually require specialists. It also remains uncertain whether covering a one-time or periodic examination would be an effective means of improving beneficiary health outcomes. A previous CMS initiative that included preventive health care visits ended with mixed results. In the late 1980s and early 1990s, the agency conducted a congressionally mandated demonstration to test varied health promotion and disease prevention services, such as free preventive visits, health risk assessment, and behavior counseling, to see if they would increase use of preventive services, improve health outcomes, 24 C.H. Sox et al., “Periodic Health Examinations and the Provision of Cancer Prevention Services,” Archives of Family Medicine, 6 (1997): 223–230. This study reviewed a random selection of community general internists and family physician practices in New Hampshire and Vermont. Care was assessed for those who were patients of the study physicians for at least 1 year, were age 42 or older, had no life-threatening illness, and had recently visited the physician. 25 See Congressional Budget Office cost estimate, H. R. Rep. 107-539, pt. 1, at 238. Beginning in 2004, the bill would have required Medicare to pay for a routine physical examination and associated services when furnished within 6 months of a beneficiary’s enrollment in part B. Beneficiaries already enrolled would not have been eligible for this benefit. H.R. 4954, 107th Cong. (2d Sess. 2002). Page 19 GAO-03-958 Medicare Preventive Services and lower health care expenditures for Medicare beneficiaries.26 The agency’s final report, published in 1998, concluded that the demonstration services were marginally effective in raising the use of some simple disease prevention measures, such as immunizations and cancer screenings, but did not consistently improve beneficiary health outcomes or reduce the use of hospital and skilled nursing services.27 CMS Is Exploring an CMS is exploring one alternative for Medicare preventive care that would Alternative for Assessing provide systematic health risk assessments to fee-for-service beneficiaries Health Risks and through a means other than physician visits. In the late 1990s, the agency commissioned the RAND Corporation to evaluate the potential Delivering Preventive effectiveness of health risk assessment programs. Similar to the Services approaches taken by the Medicare + Choice plans we reviewed, such programs collect information from individuals; identify their risk factors; and refer the individuals to at least one intervention to promote health, sustain function, or prevent disease.28 The study concluded that health risk assessment programs have increased beneficial behavior (particularly exercise) and improved physiological variables (particularly diastolic blood pressure and weight) and general health status. It also concluded that more research would help clarify the programs’ effects on preventive services such as clinical screening.29 In addition, the study stated that to be 26 A 4-year demonstration was mandated in the Consolidated Omnibus Budget Reconciliation Act of 1985, Pub. L. No. 99-272, § 9314, 100 Stat. 82, 194 (1986), and extended for 1 year by the Omnibus Budget Reconciliation Act of 1990, Pub. L. No. 101-508, § 4164, 104 Stat. 1388, 1388-100. At the time, CMS was known as the Health Care Financing Administration. 27 Donna E. Shalala, Medicare Prevention Demonstration: Final Report, RC 87-172 (Washington, D.C.: Department of Health and Human Services, 1998). The report tempered these results by pointing out that the relatively brief period during which the services were provided (roughly 2 years) and the limited number of provider contacts and follow-ups (one to two) may have been inadequate to achieve measurable outcomes. In addition, the grouping of the health risk assessment and preventive services into a preventive package may have obscured the relative effects of individual components of the package. 28 A typical health risk assessment obtains information on demographic characteristics (e.g., sex, age), lifestyle (e.g., smoking, exercise, alcohol consumption, diet), personal health history, and family health history. In some cases, physiological data (e.g., height, weight, blood pressure, cholesterol levels) are also obtained, as well as a patient’s status regarding cancer screens and immunizations. 29 Southern California Evidence-Based Practice Center/RAND, Health Risk Appraisals and Medicare (Baltimore: Centers for Medicare & Medicaid Services, 2001). RAND identified 267 articles, unpublished reports, and conference presentations, of which 27 contained data that project staff deemed necessary to be included as evidence of the effectiveness of health risk assessments. Page 20 GAO-03-958 Medicare Preventive Services effective, risk assessment questionnaires must be coupled with follow-up interventions such as referrals to appropriate services. The study found limited but encouraging evidence on the effectiveness of health risk assessment programs but concluded that the evidence was insufficient to accurately estimate the programs’ cost-effectiveness. The study recommended that CMS conduct a demonstration to test cost- effectiveness and other aspects of the health risk assessment approach for Medicare beneficiaries. Following up on the study’s findings, CMS has begun designing a fee-for- service-focused demonstration project, called the Medicare Senior Risk Reduction Program, to identify health risks and follow up with preventive services provided by means other than physician visits. The program will use a beneficiary-focused health risk assessment questionnaire to assess health risks, such as lifestyle behaviors, and use of clinical preventive and screening services. Because the demonstration is still in its design phase, the particular set of risk factors to be included is not yet final. Risk factors that might be addressed include preventable accidents such as falls, lack of exercise, high blood pressure, obesity, and use of preventive services. The Medicare Senior Risk Reduction Program will test different approaches to administering health risk assessments, creating feedback reports, and providing follow-up services, such as referring beneficiaries to health-promoting community services including physical activity and social support groups. According to project researchers, the program will tailor preventive interventions to individual risks; track patient risks and health over time; and provide beneficiaries with self-management tools and information, health behavior advice, and end-of-life counseling where appropriate. The design phase is scheduled for completion in late 2003, when CMS will decide whether to conduct a full demonstration.30 According to CMS officials, the potential demonstration’s final cost was uncertain at the time our report was completed. CMS is spending approximately $1 million on the developmental work. Unlike some health risk assessment programs, CMS’s program will be limited to questionnaires and follow-up contacts; it will not directly provide clinical screening such as blood pressure or cholesterol measurements. Instead, the program will concentrate on identifying, through information provided by the beneficiary, any modifiable lifestyle 30 According to CMS, the demonstration would also require approval from the Office of Management and Budget. Page 21 GAO-03-958 Medicare Preventive Services and behavioral risk factors and on referring beneficiaries to services for reducing those risks. CMS officials and researchers did indicate, however, that the program’s risk assessment tools will collect information on needed immunizations and cancer screenings and alert beneficiaries and their physicians to any needed services. CMS Is Also Exploring CMS has other initiatives under way that may help improve the delivery of Ways to Improve Care for preventive services within the fee-for-service program. The first is the Those with Identified Medicare Stop Smoking Program, a smoking cessation demonstration project for fee-for-service beneficiaries. Recognizing that smoking is the Health Risks and single most preventable cause of disease and death in the United States, Conditions posing a significant health risk to the aged, CMS launched the demonstration to identify the most effective service to help beneficiaries stop smoking. The demonstration will evaluate the effectiveness of different smoking cessation services. The four services being tested are: (1) reimbursement for provider counseling, (2) reimbursement for provider counseling and for smoking cessation drugs or nicotine replacement therapy, (3) access to a telephone counseling quit-line plus reimbursement for nicotine replacement therapy, and (4) provision of written information on smoking cessation. Seven states are participating in the demonstration: Alabama, Florida, Missouri, Ohio, Oklahoma, Nebraska, and Wyoming. The study will be completed in 2004, with the results published in 2005. CMS has budgeted approximately $14 million for this project. CMS is also developing a physician group-practice demonstration that was required by the Medicare, Medicaid, and SCHIP Benefits Improvement and Protection Act of 2000.31 The aim of this demonstration is to provide incentives for physicians to coordinate and manage the overall health care needs of Medicare fee-for-service beneficiaries, especially those with chronic health conditions. Under the 3-year demonstration, physician groups will be paid on a fee-for-service basis and may, in some circumstances, earn a bonus from savings achieved if the average Medicare expenditure for beneficiaries in their group of patients is below an established target.32 Up to six physician group practices will be selected 31 Pub. L. No. 106-554, app. F, § 412, 114 Stat. 2763, 2763a-509. 32 Annual performance targets will be established for each participating physician group, equal to the average Medicare expenditures of beneficiaries assigned to that group during the base period and adjusted for health status and expenditure growth. Page 22 GAO-03-958 Medicare Preventive Services to participate in the demonstration, which is expected to start during 2003. Under the mandate, the aggregate expenditures for this demonstration must be budget neutral. Any bonus payments made to physician groups must therefore be taken from savings produced by the participating organizations. Finally, a 4-year coordinated-care demonstration is currently under way at 16 sites. Authorized by the Balanced Budget Act of 1997, this demonstration examines private-sector best practices for coordinating the care of patients with complex chronic conditions.33 These conditions include congestive heart failure, other heart and lung diseases, liver diseases, diabetes, psychiatric disorders, Alzheimer’s disease or other dementia, and cancer. CMS is testing whether care coordination programs—such as those that develop a plan of care after a complete assessment of patient needs and offer patient education, health care service arrangements, and coordination with providers—can, without increasing program costs, improve the quality of care and reduce avoidable hospital admissions among Medicare beneficiaries with chronic diseases. The selected sites mix case management and disease management models in their practices;34 operate in urban and rural settings around the country; and include hospitals, retirement communities, and academic medical centers. CMS is required to formally evaluate the projects every 2 years after implementation and report to the Congress on its findings. HHS officially announced the selected sites in January 2001, and as of May 2003, the 16 sites had enrolled approximately 10,000 Medicare beneficiaries in the demonstration. CMS officials stated that the demonstration could eventually enroll more than 36,000 beneficiaries, although half of these will serve as a control group who will not receive coordinated care. CMS officials told us that they expect this demonstration to also be budget neutral. That is, they anticipate that overall costs to Medicare for providing the services will be offset by savings achieved from providing the care coordination services. 33 Pub. L. No. 105-33, § 4016, 111 Stat. 343, 345. 34 Case management services would be provided to help manage general health, and disease management services would be provided to help manage a specific disease. Page 23 GAO-03-958 Medicare Preventive Services Most Medicare beneficiaries receive some preventive services, but many Concluding do not receive services that can help prevent and manage their health risks Observations and conditions early, before significant health problems occur. Services recommended for all people in this age group are not delivered consistently. Perhaps of most concern, nearly one-third of beneficiaries who were screened and identified as having elevated blood pressure or high cholesterol measures in a nationally representative survey had not previously been told by their physicians or other health providers that they had these conditions. Projected nationally, the survey results translate into millions of people who could be unaware that they have a health condition whose treatment could prevent or delay much more significant health concerns. The solutions to ensure that beneficiaries receive needed services are not obvious. The experience of selected Medicare + Choice plans shows that no single approach stands out. All plans we contacted had a means to identify health risks, to provide feedback on risks to patients or their physicians, and to follow up with interventions to reduce those risks. But the follow-up programs, approaches, and priorities differed among the plans we contacted, and few had evaluated their approaches in a manner that would indicate whether these programs could, without significantly increasing costs, improve health outcomes for Medicare beneficiaries. Nevertheless, some current research shows promise for improving the delivery of preventive services—particularly when there are follow-up interventions, such as referrals to appropriate services. We obtained comments on our draft from HHS as well as from the health Agency Comments plans we contacted. HHS generally concurred with our findings and provided examples of CMS’s successes in promoting existing preventive services and in identifying strategies that might be used in future health promotion efforts. HHS also clarified the status of its program evaluating the use of individual health risk assessments, which is in development, and clarified its Medicare Stop Smoking Program, which will assess options for a new benefit for smoking cessation but not necessarily lead to CMS coverage for these benefits. HHS emphasized that only the Congress can decide which preventive services or benefits Medicare covers. HHS also updated its estimate of this program’s budget. We incorporated these clarifications in the draft. HHS also commented that without sufficient evidence, the report links beneficiaries’ lack of knowledge that they may have certain conditions, such as high blood pressure, with evidence that they are not receiving the Page 24 GAO-03-958 Medicare Preventive Services full range of preventive services. We did not intend to link these statements, but we have independent evidence for each of them and have added information to our summary of results to help clarify this evidence. HHS’s comments are reproduced in appendix IV. HHS and the health plans also provided technical comments that we considered and incorporated where appropriate. As arranged with your office, unless you release its contents earlier, we plan no further distribution of this report until 30 days after its issue date. We are sending copies of this report to the Secretary of HHS, the Administrator of CMS, the Director of CDC, and others who are interested. We will make copies available to others on request. In addition, the report will be available at no charge on the GAO Web site at http://www.gao.gov. If you or your staff have any questions, please contact me at (202) 512-7119 or Katherine Iritani, Assistant Director, at (206) 287-4820. Other individuals who made contributions to this report include Matthew Byer, Sophia Ku, and Tina Schwien. Sincerely yours, Janet Heinrich Director, Health Care—Public Health Issues Page 25 GAO-03-958 Medicare Preventive Services Appendix I: Scope and Methodology Because no single source contained all the information we needed to assess the extent to which Medicare beneficiaries receive preventive services through existing physician visits, we used data from four national health surveys: three conducted by the Centers for Disease Control and Prevention (CDC) and one conducted by the Centers for Medicare & Medicaid Services (CMS) (see table 2). For example, CMS’s Medicare Current Beneficiary Survey samples Medicare beneficiaries, asking them for detailed information on their demographic characteristics, insurance coverage, and health status but asking only a few questions about specific preventive services received during physician visits. In contrast, CDC’s National Ambulatory Medical Care Survey samples physicians about office visits, rather than the people who made those visits. The survey contains information about reasons for office visits and about diagnostic and preventive services provided during visits, but it cannot be used to determine the extent to which Medicare beneficiaries received these services.1 Table 2: Four National Health Surveys with Preventive Services Data, 1999–2000 Survey Data year Sample size Description Behavioral Risk Factor 2000 Annual target of 189,450 A state-based random telephone survey of U.S. adults Surveillance System, CDC adults covering a wide range of behaviors affecting health. The largest continuing telephone survey in the United States, it provides national as well as state-specific estimates. National Ambulatory 2000 27,369 office visits, of A national sample survey of visits to office-based physicians in Medical Care Survey, which 7,381 were made the United States. Detailed information about each visit, such CDC by people age 65 and as major reason for the visit and diagnostic and preventive older services ordered or provided, is collected through a patient record form completed by the physicians’ offices. National Health and 1999–2000 9,965 people, of which This survey gathers nationally representative data on the Nutrition Examination 1,392 were age 65 and health and nutrition of the U.S. population through direct Survey, CDC older physical examinations and interviews. Medicare Current 2000 About 16,000 Medicare A continuous survey of a representative national sample of the Beneficiary Survey, CMS beneficiaries Medicare population that collects detailed data on beneficiaries’ insurance coverage, health status and functioning, and health care use and expenditures. Source: CDC and CMS. For our analyses of these surveys, we extracted data for people age 65 and older to represent Medicare beneficiaries, because almost 95 percent of 1 The National Ambulatory Medical Care Survey is conducted by CDC’s National Center for Health Statistics. See the Web site http://www.cdc.gov/nchs/about/major/ahcd/ahcd1.htm for details on the survey design. Page 26 GAO-03-958 Medicare Preventive Services Appendix I: Scope and Methodology the population in this age group was enrolled in Medicare in 2000.2 Also, because the National Ambulatory Medical Care Survey samples office visits to physicians, not the people who made the visits, to estimate the average number of physician visits made by Medicare beneficiaries, we first estimated the number of visits made by patients age 65 and older using this database, and then divided this number by the U.S. Bureau of the Census estimates of the civilian noninstitutionalized population age 65 and older. To determine the major reasons for physician visits and the specific types of preventive services provided to Medicare beneficiaries in the fee-for-service program, we used visit data in this survey for patients age 65 and older who did not belong to a health maintenance organization and whose visits were not paid on a capitated basis.3 Tables 3 to 5 show the estimates and standard errors in data from the National Ambulatory Medical Care Survey 2000 on major reasons for physician visits and on the preventive diet counseling services provided during those visits. We also tested at the 95 percent confidence level the statistical significance of differences we observed between nonillness and other types of visits in the proportion of visits where preventive screening tests (e.g., cholesterol and blood tests) were provided. Table 3: Estimated Proportion of Fee-for-Service Physician Visits Made by People Age 65 and Older, by Major Reason for the Visits, 2000 Estimated Standard Sample number (in Estimated error of Major reason size thousands) percentage percentage Acute problem 1,155 32,843 25.8 1.7 Chronic problem, routine 2,081 53,701 42.2 1.7 Chronic problem, flare-up 532 13,254 10.4 0.8 Pre- or postsurgery, injury follow-up 577 12,533 9.8 1.1 Nonillness care 395 12,479 9.8 1.1 Blank or unknown 84 2495 2.0 0.4 Source: GAO analysis of the National Ambulatory Medical Care Survey, CDC. 2 According to data from CDC’s Behavioral Risk Factor Surveillance System, in 2000, almost 95 percent of adults age 65 and older reported having Medicare coverage. 3 “Capitated” refers to a method of payment for health services in which an individual or institutional provider is paid a fixed amount for each person served, without regard to the actual number or nature of services provided to each person in a set period of time. Page 27 GAO-03-958 Medicare Preventive Services Appendix I: Scope and Methodology Table 4: Estimated Proportion of Fee-for-Service Physician Visits in Which Diet Counseling Services Were Provided or Ordered, by Major Reason for the Visits, 2000 Estimated Standard Sample number (in Estimated error of Major reason size thousands) percentagea percentage Acute problem 1,155 4,138 12.6 3.0 Chronic problem, routine 2,081 11,785 22.0 3.0 Chronic problem, flare-up 532 1,673 12.6 2.5 Nonillness care 395 2,295 18.4 3.6 Source: GAO analysis of the National Ambulatory Medical Care Survey, CDC. a The differences in rates of services provided among the different types of visits were not statistically significant. According to CDC, diet counseling services could be underreported because the survey captured this information only if it was contained in the medical record. If the physician provided counseling but did not write it in the chart, counseling would not have been captured in the survey. Table 5: Estimated Proportion of Fee-for-Service Physician Visits in Which Blood Pressure Measurements Were Provided or Ordered, by Major Reason for the Visits, 2000 Estimated Standard Sample number (in Estimated error of a Major reason size thousands) percentage percentage Acute problem 1,155 18,491 56.3 3.2 Chronic problem, routine 2,081 31,706 59.0 2.9 Chronic problem, flare-up 532 7,870 59.4 4.8 Nonillness care 395 7,762 62.2 4.8 Source: GAO analysis of the National Ambulatory Medical Care Survey, CDC. a The differences in rates of services provided among the different types of visits were not statistically significant. To estimate the proportion of Medicare beneficiaries who had health conditions that they were not previously aware of—specifically, high blood pressure or high cholesterol—we used data from both the interview and the physical examination portions of CDC’s National Health and Nutrition Examination Survey (see app. III for methodology and results from this analysis). To describe the preventive care approaches of Medicare + Choice plans, we consulted with national experts and officials from the American Page 28 GAO-03-958 Medicare Preventive Services Appendix I: Scope and Methodology Association of Health Plans and chose five plans considered to have innovative preventive care programs. Together, these five plans serve more than 1.2 million Medicare beneficiaries in 15 states and the District of Columbia (see table 6). We interviewed officials from each plan and reviewed documents, including plan-provided studies or evaluations of their preventive services programs. We reviewed the scope and methodology of the studies done by some of the plans, but we did not independently verify the accuracy of the data. Table 6: Medicare + Choice Plans Included in GAO’s Study Beneficiaries Medicare + Choice plans Geographic areas served served AvMed Health Plans Florida 24,400 Group Health Cooperative Washington 59,300 Highmark Blue Cross & Pennsylvania Blue Shield 182,000 Kaiser Permanente California, Colorado, District of Columbia, Georgia, Hawaii, Maryland, Ohio, Oregon, Virginia, Washington 880,000 Oxford Health Plans Connecticut, New Jersey, New York 72,000 Source: Plan officials and plan Web sites. To examine the alternatives for identifying and reducing health risks and CMS’s efforts in exploring them, we reviewed available literature, including results of past demonstrations and congressionally mandated studies, and interviewed experts in the field, including those conducting studies and developing position papers for the Partnership for Prevention, a nonprofit organization funded by the Robert Wood Johnson Foundation. We also interviewed Department of Health and Human Services and CMS officials and reviewed documents on planned and present CMS demonstrations related to preventive services. Page 29 GAO-03-958 Medicare Preventive Services Appendix II: Preventive Services Recommended by the U.S. Preventive Services Task Force or Covered by Medicare Year first covered by Task force recommendation Medicare as preventive Medicare cost-sharing a Service for age 65+ service requirements Immunization Pneumococcal Recommends 1981 None Hepatitis B No recommendation 1984 Copayment after deductible Influenza Recommends 1993 None Tetanus-diphtheria (Td) Recommends Not coveredb N/A boosters Varicella Recommends Not coveredb N/A Screening Cervical cancer: pap smear Recommends againstc 1990 Copayment with no deductibled e Breast cancer: mammography Recommends 1991 Copayment with no deductible d Vaginal cancer: pelvic exam Not evaluated 1998 Copayment with no deductible Colorectal cancer: fecal-occult Strongly recommends 1998 No copayment or deductible blood testf Colorectal cancer: flexible Strongly recommends 1998 Copayment after deductibleg f sigmoidoscopy or colonoscopy Osteoporosis: bone mass Recommends (women only) 1998 Copayment after deductible measurement d Prostate cancer: prostate- Insufficient evidence to 2000 Copayment after deductible specific antigen test and/or recommend for or against digital rectal examination Glaucoma Insufficient evidence to 2002 Copayment after deductible recommend for or against Vision impairment Recommends Not covered N/A Hearing impairment Recommends Not covered N/A Height, weight, and blood Recommends Not covered N/A pressure Cholesterol measurement Strongly recommends Not covered N/A Problem drinking Recommends Not covered N/A Depression Recommends Not covered N/A Counseling Smoking cessation, injury Recommends Not covered N/A prevention, dental health Aspirin for primary prevention Strongly recommends Not covered N/A of cardiovascular events Source: U.S. General Accounting Office, Medicare: Use of Preventive Services Is Growing but Varies Widely, GAO-02-777T (Washington, D.C.: April 12, 2002), and U.S. Preventive Services Task Force, Guide to Clinical Preventive Services, 2nd ed. (Washington, D.C.: 1996) and related updates. Page 30 GAO-03-958 Medicare Preventive Services Appendix II: Preventive Services Recommended by the U.S. Preventive Services Task Force or Covered by Medicare a Applicable Medicare cost-sharing requirements generally include a 20 percent copayment after a $100 per year deductible. Specifically, each year, beneficiaries are responsible for 100 percent of the payment amount until those payments equal a specified deductible amount, $100 in 2003. Thereafter, beneficiaries are responsible for a copayment that is usually 20 percent of the Medicare-approved amount. For certain tests, the copayment may be higher. 42 U.S.C. § 1395(a)(1) (2000). b Although the tetanus-diphtheria (Td) and varicella (chickenpox) booster vaccinations are not now covered under Medicare as a “preventive” service, these treatments might be covered under Medicare if necessary to a beneficiary’s medical treatment. Medicare provides coverage for medical treatment and services that are “reasonable and necessary for the diagnosis or treatment of an illness or injury,” provided that the services or products used are “safe and effective” and not merely “experimental.” 42 U.S.C. § 1395(a)(1)(A) (2000). c The task force recommends against routinely screening women older than 65 for cervical cancer if they have had adequate recent screening with normal Pap smears and are not otherwise at high risk for cervical cancer. d The costs of the laboratory test portion of these services are not subject to copayment or deductible. The beneficiary is subject to a deductible, copayment, or both for physician services only. e The task force recommends screening mammography, with or without a clinical breast examination, every 1–2 years for women age 40 and older. f Data are insufficient to determine which strategy is best to balance benefits against potential harms or cost-effectiveness. Barium enemas are covered as an alternative if a physician determines that their screening value is equal to or greater than sigmoidoscopy or colonoscopy. g The copayment has increased from 20 to 25 percent for services rendered in an ambulatory surgical center. Page 31 GAO-03-958 Medicare Preventive Services Appendix III: National Health and Nutrition Examination Survey Methodology and Results Conducted by the Centers for Disease Control and Prevention’s (CDC) Background National Center for Health Statistics, the National Health and Nutrition Examination Survey (NHANES) is a nationwide population-based survey designed to estimate the health and nutrition of the noninstitutionalized U.S. civilian population. Our analysis was based on data gathered during NHANES 1999–2000, which represent the most recent information available. This survey comprises two parts: an in-home interview and a health examination. During the in-home interview, participants are asked about their health status, disease history, and diet; during the health examination, participants receive a number of tests, including blood pressure readings and a blood test to determine total serum cholesterol.1 Details of the survey design, questionnaires, and examination components are available at http://www.cdc.gov/nchs/nhanes.htm. For our analysis, we used the NHANES data described in table 7 to Scope, Methodology, determine if participants age 65 and older2 had high blood pressure or high and Results total serum cholesterol. We used the same criteria for these conditions as CDC and the National Heart Blood and Lung Institute use to estimate the conditions’ prevalence. Table 7: NHANES Data GAO Used to Determine if Participants Had Measures of Specific Health Conditions Health condition NHANES data High blood pressure a Averageb systolic blood pressure ≥ 140 during NHANES exam or Averageb diastolic blood pressure ≥ 90 during NHANES exam or Participant reported during NHANES interview that he or she took hypertension medication High total cholesterola Total cholesterol level ≥ 240 at NHANES examination Source: CDC criteria and GAO methodology. a CDC’s definitions of high blood pressure and high total cholesterol. 1 Which examinations and blood tests a participant had depended on that participant’s age and sex. 2 Of the 9,282 individuals participating in both the NHANES interview and examination components, 1,196 were age 65 and older. Page 32 GAO-03-958 Medicare Preventive Services Appendix III: National Health and Nutrition Examination Survey Methodology and Results b Participants’ blood pressure was measured three or four times during the 1-day physical examination. For our analysis, we determined the average of these blood pressure measurements and applied CDC’s definition of high blood pressure. To determine whether the participants age 65 and older found by examination to have elevated measures of these health conditions were previously unaware of having them, we used patients’ responses from the NHANES interview. During the interview, participants were asked if they had ever been told by a physician or health professional that they had certain conditions, including high blood pressure and high cholesterol. Tables 8 and 9 show the estimates and standard errors from 1999–2000 NHANES data for specific health conditions and level of awareness among participants age 65 and older. Table 8: People Age 65 and Older in the United States Found to Have Measures of Specific Health Conditions, NHANES 1999–2000 Estimated Sample number in the Estimated Standard error Health condition size U.S. population proportion of proportion High blood pressure 835 21,000,000 71.6% 2.07 High total cholesterol 250 7,100,000 25.6% 1.76 Source: GAO analysis of NHANES. Table 9: People Age 65 and Older in the United States Found to Have Measures of Specific Health Conditions and Who Reported They Had Not Previously Been Told They Might Have the Condition, NHANES 1999–2000 Estimated Standard Not previously told of Sample number in the Estimated error of the health condition size U.S. population proportion proportion High blood pressure 254 6,600,000 31.6% 2.02 High total serum cholesterol 87 2,100,000 32.1% 4.65 Source: GAO analysis of NHANES. Estimated numbers, proportions, and standard errors were obtained using SUDAAN, a computer program for analyzing data from complex sample surveys, as suggested in the NHANES Analytic Guidelines. Page 33 GAO-03-958 Medicare Preventive Services Appendix IV: Comments from the Department of Health and Human Services Page 34 GAO-03-958 Medicare Preventive Services Appendix IV: Comments from the Department of Health and Human Services Page 35 GAO-03-958 Medicare Preventive Services Appendix IV: Comments from the Department of Health and Human Services Page 36 GAO-03-958 Medicare Preventive Services Appendix IV: Comments from the Department of Health and Human Services (290204) Page 37 GAO-03-958 Medicare Preventive Services The General Accounting Office, the audit, evaluation and investigative arm of GAO’s Mission Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO’s commitment to good government is reflected in its core values of accountability, integrity, and reliability. The fastest and easiest way to obtain copies of GAO documents at no cost is Obtaining Copies of through the Internet. GAO’s Web site (www.gao.gov) contains abstracts and full- GAO Reports and text files of current reports and testimony and an expanding archive of older products. The Web site features a search engine to help you locate documents Testimony using key words and phrases. You can print these documents in their entirety, including charts and other graphics. Each day, GAO issues a list of newly released reports, testimony, and correspondence. GAO posts this list, known as “Today’s Reports,” on its Web site daily. The list contains links to the full-text document files. To have GAO e-mail this list to you every afternoon, go to www.gao.gov and select “Subscribe to e-mail alerts” under the “Order GAO Products” heading. Order by Mail or Phone The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. General Accounting Office 441 G Street NW, Room LM Washington, D.C. 20548 To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202) 512-6061 Contact: To Report Fraud, Web site: www.gao.gov/fraudnet/fraudnet.htm Waste, and Abuse in E-mail: email@example.com Federal Programs Automated answering system: (800) 424-5454 or (202) 512-7470 Jeff Nelligan, Managing Director, NelliganJ@gao.gov (202) 512-4800 Public Affairs U.S. General Accounting Office, 441 G Street NW, Room 7149 Washington, D.C. 20548
Medicare: Most Beneficiaries Receive Some but Not All Recommended Preventive Services
Published by the Government Accountability Office on 2003-09-08.
Below is a raw (and likely hideous) rendition of the original report. (PDF)