April 11, 2007
Debra A. Draper
Statement of Debra Draper, Ph.D., Associate Director
Center for Studying Health System Change
Before the U.S. Senate
Committee on Finance
Hearing on the Medicare Advantage Program
Chairman Baucus, Senator Grassley and members of the Committee, thank you for the invitation to testify about the Medicare Advantage Program. My name is Debra A. Draper, and I am a health services researcher and Associate Director of the Center for Studying Health System Change (HSC). HSC is an independent, nonpartisan health policy research organization funded principally by the Robert Wood Johnson Foundation and affiliated with Mathematica Policy Research.
HSCs main research tool is the Community Tracking Study, which consists of national surveys of households and physicians in 60 nationally representative communities across the country and intensive site visits to 12 of these communities. We also monitor secondary data and general health system trends. Our goal is to inform policy makers with objective and timely research on developments in health care markets and their impacts on people. We do not make policy recommendations. Our various research and communication activities may be found on our Web site at www.hschange.org.
My testimony today is intended to inform the Committee about what we and other researchers know about commercial health plans care management activities and the impact of those activities on costs, quality and outcomes.1 My comments draw largely from our most recent site visit work in the 12 Community Tracking Study sitesBoston, Cleveland, Greenville, S.C., Indianapolis, Lansing, Little Rock, Miami, northern New Jersey (Newark), Orange County, Calif., Phoenix, Seattle, and Syracuse, N.Y. 2 These sites were chosen randomly to be representative of metropolitan areas in the United States, and we have conducted site visits to each bi-annually since 1996.
Our sixth round of site visits is currently underway, and they include interviews
with key stakeholders of the local health care systems, including employers,
health plans, policy makers, providers and others. We have found our triangulation
(asking different stakeholders about their interactions with other stakeholders)
is critical for getting not only a more complete perspective on a local site,
but such an approach also provides us a mechanism to validate and ensure the
reliability of the data that we collect. As in other recent site visit rounds,
we expect to conduct approximately 1,000 interviews during this round. For our
health plan interviews, we target three leading plans in each community and
within each plan, speak with executives responsible for marketing, medical management
and network management. The quotes found in this document are all from our 2007
My testimony today will focus on three key points:
Commercial Health Plans Care Management Activities
As the cost of health care continues to increase, there has been a resurgence of interest in finding effective ways to curb the escalating trend. Much of the recent impetus has come from employers concerned about the growing financial burden of sponsoring employee health benefits. But it has also been fueled in part by the "consumerism" movement in health care, which among other things, shifts greater responsibility to the consumer for not only their health care costs, but also to more actively engage in managing their health and health care decisions. Many employers are looking to their health plans to offer effective strategies that respond to these issues. Health plans have often responded with a variety of activities around care management.
While many of the care management activities that health plans currently offer are not new, there is growing pressure from not only employers, but from the health plans themselves, to make them more effective. For example, a health plan executive in Seattle recently told us, "When disease management programs were more popular in the marketplace 10-15 years ago, these programs were popular for a period of time, but entities couldnt show quantitatively the impact of the programs so they faded from popularity. People are taking a swipe at it now and advertising better bells and whistles to show it makes a difference. They are promising that it works but whether they are effectively delivering on that promise, Im not sure I know the answer to that. I also think that in the population of people and in the media at large, theres a huge demand for it. So as recognition becomes widespread about the high cost of health care and lack of good health, people are questioning what entities are doing about that. When they read about the obesity epidemic, diabetes, heart disease, cancer and certain way of life issues that are contributing to those issues, it brings back the focus to these programs."
Funding for health plans care management activities comes from several sources. For self-insured employers, they are usually provided at an additional cost, while for fully insured employers, the cost is typically incorporated into the premium.
There is considerable variation among health plans as to what specifically
they offer in the way of care management activities. Much of this has to do
with how they package, brand and market these activities, which is often a way
in which they differentiate themselves in the market. As one Cleveland health
plan executive noted, "All plans offer these things. But it comes down
to how you package these tools so that it looks like what youre doing really
works, try to differentiate yourselves. No health plan has the gimmick that
no one else has. We all try to do it." Health plans provide their care
management activities through internal capacity, through external vendors, or
some combination of the two. Some of the larger national health plans have subsidiary
companies that specialize in these activities.
There are many health plan activities that can be broadly construed as care management. For purposes of my testimony today, however, I am focusing only on selected activities that include both enrollee-focused activities, such as case management and coordination, disease management, health promotion and wellness, and nurse advice lines, as well as process-focused activities, such as utilization management. Conceptually, these are all activities that aim to reduce costs and improve health outcomes by:
Case management and coordination activities target enrollees with health conditions that put them at risk for incurring large medical expenditures. These activities are individually customized to the needs of the enrollee and may include care planning, coordination of follow-up care, and telephone-based support and assistance.
A health plan in Cleveland is doing what a plan executive describes as "situation management." After an enrollee is discharged from the hospital, the health plan assigns a care counselor to ensure that the enrollee understands discharge instructions, to determine if the enrollee has any unmet needs, and to make sure that all medications are being filled. The health plan instituted its "situational management" activities after identifying quality of care issues when enrollees were moving from one care setting to another, such as from hospital to home or hospital to nursing home.
Disease management activities target enrollees with certain diseases or health conditions, commonly diabetes, asthma and other prevalent conditions. The objective of these activities is to encourage enrollees adherence to standardized treatment guidelines and self-care and are generally facilitated by mail and telephone contact.
Across the communities that we study, nearly all health plans offer some type of disease management activities, although the number of these and the diseases or conditions they target vary by plan. Among the more common disease management activities that health plans offer are programs that focus on asthma, chronic obstructive pulmonary disease, congestive heart failure, coronary artery disease, diabetes, high- risk pregnancies and hypertension. Additionally, health plans sometimes report offering programs for other conditions such as pain management and depression. An important challenge that health plans face around disease management initiatives is how to engage the enrollees physician since these activities primarily interface directly with the enrollee. A plan in Cleveland, for example, is currently developing a data integration initiative in collaboration with one of its key provider organizations to better ensure consistency of efforts between the plans disease management efforts and what takes place in physicians offices.
Health Promotion and Wellness
Health promotion and wellness activities target enrollees irrespective of disease status or health utilization. These activities encourage enrollees to pursue healthy behaviors, including healthy eating and exercise. They also provide support to enrollees interested in changing unhealthy behaviors such as smoking. Across the communities we study, we see health plans offering a range of activities, including health risk assessments, gym memberships, weight management support and smoking cessation programs. In Seattle, a health plan executive said, "What we found are that expectations of employer groups are changing. About 55 percent of 1.2 million members spend less than $500 a year. So if health care premiums are $3,600 a year, people are asking whats in it for me? So were being more aggressive on the front end to keep people from deteriorating health. Were giving equal focus to wellness as to disease management. Weve made changes by adding programs with a wellness focus-health coaches, for example. So if you do a health risk assessment and say youd like to lose 10 pounds, the health coaches will work with you to achieve that goal." A health plan in Cleveland has wellness teams, which include nurses that go worksites to do health fairs. In northern New Jersey, a health plan there has a healthy eating program, where enrollees along with their primary care physicians decide on a goal weight, which includes periodic incentives in the form of a gym membership or cash for meeting targeted goals. The growing sophistication of information technology has helped facilitate plans activities around health promotion and wellness. Enrollees are often able to access a variety of health-related information online.
Nurse Advice Lines
Nurse advice lines provide enrollees with telephone access to health plan clinical staff. The lines are usually staffed by registered nurses and operate 24 hours a day, 7 days a week. Nurses provide enrollees who call with a range of services, including education and advice on health conditions and self-care. They also provide triage services working with the enrollee to assess the acuity of symptoms to determine an appropriate course of treatment, and where alternative choices may be available, the most cost efficient. Health plans actively encourage enrollees use of these services. A health plan in Cleveland, for example, includes a nurse line card and magnet in its packets to enrollees.
Utilization management encompasses a range of activities that health plans use to manage the use of health care services. Among other things, these activities are intended to prevent medically unnecessary services. In many of the communities we study, over-utilization of high-end imaging services, such as magnetic resonance imaging (MRI), computerized axial tomography (CAT) scans, and positron emission tomography (PET) scans, has been problematic in recent years. In Cleveland, for example, health plans there reported seeing 20-40 percent growth trends in the utilization of these services, which prompted some to institute pre-certification of these procedures, which requires a medical necessity review, rather than a pre-notification, which typically only requires a courtesy notification to the plan. As one Cleveland health plan executive noted, "The interesting thing is that the denial rate is running only 1.4-1.5 percent. Were not saying no that much, but because people have to justify what difference the test is going to make, they arent requesting it as much."
Identifying and Engaging Enrollees
Health plans typically target enrollees that account for the largest share
of costs for more intensive care management activities - often those with chronic
such as asthma and diabetes, or rare health conditions. There appears to be
a renewed interest on also focusing on enrollees with specific unhealthy conditions
or behaviors, such as obesity, smoking and sedentary lifestyles. As one New
Jersey benefits consultant describes it, "Now there is much more emphasis
on appropriate discharge planning, catching folks before they need hospital
care, figuring out who is at risk for incurring charges and getting those people
into disease management programs or getting them a health coach."
The most common way in which health plans identify potential enrollees for more intensive care management is through claims data, both medical and pharmacy. With these data, for example, health plans identify potentially high-risk enrollees based on claims for hospitalizations, emergency department use, or other high cost services or procedures. Health plans often apply modeling software technologies to the claims data to predict the future health care expenditures of enrollees. Based on the predictive modeling results, enrollees are often stratified based on their projected risk levels. This allows health plans to better tailor the types and intensity of care management activities to the individual enrollee. For example, a designated low-risk enrollee may receive educational mailings, while an enrollee determined to be at higher-risk may receive more intensive services, including telephone contact.
Claims data are also used to identify gaps in care that may indicate that the quality of care that an enrollee is receiving is poor, which if not addressed, might eventually lead to further deterioration in health and higher costs. For example, a review of claims data may reveal that a diabetic enrollee has failed to receive certain services, such as a periodic blood test (hemoglobin A1c) to monitor how well they are controlling their blood glucose levels.
Another tool health plans are increasingly emphasizing to identify potentially at-risk enrollees is a health risk assessment. With the resurgence of interest in health promotion and wellness activities, we have seen interest in and use of health risk assessments grow across the communities that we study. A health risk assessment is a questionnaire, often available online, that collects information provided by the enrollee on items such as personal and family medical history, current diagnoses and symptoms, use of preventive and screening services, and health behaviors such as diet, physical activity, and tobacco and alcohol use. This enrollee-supplied information is then used to predict health risk, which may then flag the enrollee for more intensive care management. As one Seattle health plan executive describes, "Weve introduced a health risk assessment tool for our entire membership and all of this self-reported data is put into electronic records. This tool is just beginning and we have only a small percentage of members using it. We are trying to promote its use more and more so that patients will use it on an annual basis to bring us up to date on health behaviors and if they fit into any chronic disease category."
Once an enrollee has been identified as a potential recipient of more intensive care management, the next step is getting and keeping them engaged. Participation in most health plans care management activities is voluntary on the part of the enrollee. Many health plans report that engaging enrollees in care management activities is challenging. Most believe that some type of incentive is necessary to get and keep enrollees engaged, but few currently exist. The existing incentives vary widely and include, for example, free supplies (insulin and blood glucose monitoring test strips) for diabetics, cash incentives for the completion of a health risk assessment, and discounts off of gym memberships.
Evidence of Impact on Costs, Quality and Outcomes
While it is relatively easy to measure the degree to which data analysis identifies at-risk enrollees with the potential for more intensive intervention, it is much harder to measure how successful these activities are in changing costs, quality and outcomes. It becomes even harder to measure long-term impacts.
Given too that many of the care management activities that health plans currently offer are new or not being extensively applied, the evidence suggests that most health plans do not know how effective they are, let alone researchers.3 Commenting on a wellness program offered by one health plan, a New Jersey broker said, "The wellness program will cost $50,000 a year, but they guarantee the return on that investment will be $50,000 or more. We asked, how do you measure that? Thats where it got very cloudy." The broker went on to add, "They say, we see medical indications that your obesity will result in two heart attacks, we think we can prevent those heart attacks, then well tell you that you saved the price of those two heart attacks."
Assessing effectiveness is complicated further, because health plans often introduce these types of activities on a trial or pilot basis, and depending on their assessment of that limited effort, decide whether to discontinue it or expand. A health plan in Little Rock, for example, recently piloted a health coaching program in its health maintenance organization (HMO) product. It worked with an external vendor to identify high-risk patients with diabetes and heart disease to target for more intensive care management. Based on an assessment of subsequent claims, the decision was made to extend the effort to the rest of its business.
There is, however, growing pressure on health plans from employers to demonstrate the effectiveness of care management activities. There is evidence that health plans are trying to respond, but assessing the effectiveness of many of their care management activities is still evolving. For example, a health plan in Cleveland that provides smoking-cessation support, began tracking the data less than a year ago. Preliminary evidence, however, showed a quit rate of 40-45 percent among a group of enrollees participating in the program that used both nicotine replacement therapy and smoking cessation counseling.
Despite the lack of evidence, some health plans and employers appear willing to invest in these activities-at least for now. Employers that we have interviewed often say that they believe activities like health promotion and wellness are the right thing to do and are particularly important as employees continue to assume a larger share of health care costs. In each of our communities, we hear from health plans, employers and others about the need to better engage consumers to be more aware of what health care costs and to be more proactive in their knowledge of health care and the use of services.
Implications for Medicare Advantage
There is considerable potential for health plans to apply many of their care management activities to Medicare, particularly if they find them to be effective. These types of activities may be even more beneficial for a Medicare population where chronic illness and other high cost health conditions are more prevalent. But the current evidence on their impact on costs, quality, and outcomes is sparse for a number of reasons, including the newness of many of these efforts, as well as the complexity of quantifying and measuring their effectiveness.
Despite the lack of evidence, we have seen some self-insured employers willing to pay for these activities, at least for now. So, the question for Medicare becomes to what extent is it willing to support experimentation and does this justify any of the extra payment to Medicare Advantage plans when plans would reap any savings? Since heath plans are pursuing these activities in their commercial products, Medicares role in supporting experimentation is less clear. But if Medicare does want to pay for experimentation, should it pay directly for selected activities rather than paying higher overall rates? Further, how long should Medicare pay for experimentation and what outcomes over what timeframes should it expect?
Intuitively, health plan activities aimed at improving the quality and efficiency of care for their enrollees are a good thing. But without credible evidence on what impact they have on costs, quality and outcomes, it is difficult to justify financial support unless those providing the funding expect that the impact of health plans care management activities will eventually yield results that justify the investment. To the extent too that care management activities do save money, they are self-financing and may not require extra support.
|1.||See for example, G.P. Mays, M. Au, G. Claxton, "Convergence and Dissonance: Evolution in Private Sector Approaches to Disease Management and Care Coordination," Health Affairs, forthcoming 2007. A. Short, G.P. Mays, J. Mittler, "Disease Management: A Leap of Faith to Lower Cost, Higher Quality Care," HSC Issue Brief, No. 69, (Washington, DC: Center for Studying Health System Change, 2003). S. Felt-Lisk and G.P. Mays. "Back to the Drawing Board: New Directions in Health Plans Care Management Strategies." Health Affairs, vol. 21, no. 5 (2002): 210-217.|
|2.||Additional information on our site visit methodology and findings may be found on our Web site at www.hschange.org.|
|3.||Research evidence has grown in recent years but continues to be quite limited. For example, a review of selected peer-reviewed literature found estimates of return-on-investment from selected care management interventions ranging from zero impact to 640 percent. The literature mainly focused on programs for asthma, diabetes and heart disease that were implemented in controlled clinical setting-not a typical health plan setting. See, H.G. Dove and I. Duncan, Estimating Savings, Utilization Rate Changes, and Return on Investment from Care Management Interventions: Selective Literature Review of Care Management Interventions, Schaumburg, IL: Society of Actuaries, 2005.|