Effective Clinical Practice
Effective Clinical Practice, January/February 1999
Gary Gaumer, Terry Moore, Meryl Friedman, Barbara Bix
For author affiliations, current addresses, and contributions, see end of text.
Context. The Medicare program is encouraging its beneficiaries to enroll in capitated health plans.
Objective. To determine how prepared these plans are to handle chronically ill and frail elderly persons.
Design. Telephone survey of 28 health plans that together serve about one fourth of all enrollees of the Medicare Risk program.
Measures. The degree of readiness (high, intermediate, or low) of health plans in seven domains that experts believe are important to the management of an elderly population.
Results. None of the 28 health plans had high readiness scores for all seven domains. The two domains for which the plans were most prepared were risk assessment and member self-care. The plans were least prepared for the domains of cooperative team care and geriatric consultations.
Conclusions. Many plans do not offer the programs that experts believe are important for Medicare enrollees. They may hesitate to adopt strategies that lack data on effectiveness.
The Medicare program is moving quickly to offer its beneficiaries the option to enroll in capitated managed care plans. Currently, such plans enroll about 5.5 million Medicare beneficiaries. Although some plans have recently left the market, enrollment projections for 2002 have been as high as 10 million benificiaries. The Medicare Plus Choice program, passed by the U.S. Congress as part of the Balanced Budget Act of 1997, aims to expand the Medicare Risk program so that HMOs can allow other types of provider organizations to enroll beneficiaries under capitated arrangements.
Recent research warns that elderly persons who enroll in capitated health plans may have lower functional outcomes than elderly persons who receive care in traditional fee-for-service settings (1). This research supports a widespread--but anecdotal--perception that many plans are not yet ready to manage care for chronically ill and frail elderly persons. To better delineate the specific abilities that plans do and do not have, we surveyed large health plans that currently serve Medicare risk-contract enrollees.
In the summer of 1997, Abt Associates, Inc., conducted a telephone survey of Medicare risk plans to learn more about how these plans are preparing for an influx of elderly members. The survey focused on the presence of programs and care technologies that are thought to be required for the management of a population that is at higher risk for the development of chronic illnesses. In general, the survey respondent was a vice president or director of Medicare, Medicare operations, or utilization management. In some cases, we also interviewed medical directors, marketing managers, and personnel charged with utilization management.
We selected health plans that offer Medicare risk programs in the 20 largest markets in the United States. We sought to identify the largest plans in each market, and we constructed a candidate set of 80 plans (4 per market). We attempted to contact each of these plans, but 4 had failed or had merged with other plans, 10 did not have Medicare risk products approved for sale, 14 could not provide decisions about survey participation or appropriate interviewees in a timely way, and 24 refused to participate in the survey. It is not clear whether these delays and refusals introduced bias, but one might suspect that some of the plans that refused to participate did so because of low levels of readiness.
Twenty-eight health plans in 17 cities responded to our survey. Together, these 28 plans enrolled more than 1.1 million Medicare beneficiaries, or about 25% of all Medicare Risk program enrollees, at the time of the study. More than half of the plans (16 of 28) were independent practice associations. The other 12 were pure group- or staff-model plans or offered members a choice of models. Slightly fewer than half of the plans (12 of 28) were independent; the rest had counterparts in other states. About one third (9 of 28) had entered the Medicare Risk business in the past year.
Measures of Preparedness
To measure the preparedness of health plans to cope with an influx of elderly members, we identified a list of activities and programs that are common features of health delivery systems aimed at frail elderly and chronically ill populations. Although publications pertaining to some interventions and a large body of research on the effects of various home and community-based care programs (2) discuss assessments, case management, adult day care, and interdisciplinary team management of patients (3), this literature does not provide information on the effects of specific interventions, only bundles of interventions. Neither these nor similar studies have studied the effectiveness, or even the efficacy, of these various activities. Nor do we attempt this here.
On the basis of an exploratory survey of 11 plans that we had conducted 18 months earlier, we selected seven domains of clinical programming for further investigation. These domains are similar to those suggested by the HMO Workgroup on Care Management (4). In selecting these domains, we excluded other critical requirements for various reasons. In particular, we chose not to examine access to appropriate networks of hospital, rehabilitative, transitional, and skilled care facilities because Medicare requires such access for certification. Similarly, we did not include changed financial incentives for primary care providers, although we agree that these may be among the most important changes that some plans will need to make to create better incentives for effective care of chronically ill and frail elderly persons.
For each domain that we did select, we asked what each plan currently had in place or was implementing immediately. These domains and our measurement criteria for high, medium, and low readiness are shown in Table 1. Scores for level of readiness (2 = high readiness, 1 = intermediate readiness, and 0 = low readiness) for each of the seven domains were added to create a summary readiness score (range, 0 to 14). We also asked respondents whether we had overlooked any programs that they felt were essential to readiness or whether they wanted to tell us about other special programs that they were implementing for seniors.
Most of the 28 plans that we surveyed have implemented programs in several domains, but only 4 show evidence of broad progress in special programming for chronically ill or elderly persons. No plan scored high on all seven domains (Table 2), and no plan scored low on more than four domains. Most plans have made good progress on some aspects of readiness but no progress on other aspects: Twenty-one plans achieved at least one high and one low score across the seven domains.
Figure 1 details the degree of readiness seen in the seven domains of chronic care programming. The pattern shows that the plans as a group are more prepared for some domains than for others. More plans were strong in domains where investments have been made: risk assessment, member self-care, individualized case management, and the availability of utilization data. Even in these areas of relatively high preparedness, however, no more than about half of the plans had high readiness scores. More plans were weak in making geriatric consultations and cooperative team care available; about half of the plans fell below the intermediate-readiness threshold for these domains.
Most plans screen all Medicare members at enrollment to determine who is at highest risk for chronic illness. Half of the plans have formal follow-up programs; follow-up may include further assessment of care needs or assignment to case management programs. The rest of the plans that screen for high-risk members merely give primary care physicians a list of high-risk patients. Six plans reported not screening at the plan level and leaving the decision about whether to implement screening programs completely at the discretion of individual physicians or groups.
Most plans offer some form of senior-specific self-care to all members. About half of these plans restrict their offerings to newsletters or manuals prepared especially for seniors. The other half provide additional self-care programs, such as community-based seminars (e.g., on foot care, pharmacy management, influenza shots, and arthritis management), vaccination-reminder mailings, and referrals to community resources.
Cooperative Team Care
Half of the plans that we surveyed offer some type of group medical visits or support groups, including the senior-specific Chronic Care Health Clinics (CCHCs), disease management and wellness programs, and community support groups (through referrals). Six plans offer the CCHCs to patients in some locations, and all note some measure of success for this approach in the treatment of chronically ill members. Despite respondents' positive impressions about the CCHCs, none of the plans make these clinics available to all chronically ill members. Members can participate in the CCHCs only if their physician chooses to participate in the program.
Senior-Specific Provider Training
More than half of the plans offer some form of clinical training for physicians about seniors. Ten plans offer continuing medical education credit or other incentives for attendance. Plans report that only a small percentage of providers participates in these training opportunities.
About half of the plans surveyed provide no access to geriatric specialists, and several note shortages of geriatricians and recruiting difficulties. Of the 15 plans that offer some form of access to geriatricians, 4 have formal programs with multidisciplinary teams, 8 have geriatricians on staff (although not in all locations), and 3 provide referrals.
Twenty plans offer case management services, and 15 of the 20 offer senior-specific case management services. In-house, multidisciplinary teams provide most of these services; outsourcing is rare.
More than half of the plans track utilization for all services, but six plans do not have access to utilization data at the plan level.
In addition, 11 plans reported that they use other clinical programs to improve the care of chronically ill or frail elderly persons. Examples of these efforts include:
Finally, we examined the relation between overall plan readiness and plan type. Figure 2 shows the readiness scores (possible range, 0 to 14) for group- or staff-model plans compared with independent practice associations. In both groups, readiness scores are widely and similarly distributed, although one outlier in the independent practice association group had a score of 11.
It is important to emphasize that, in many cases, Medicare risk plans offer benefits or services that members would not routinely receive under traditional fee-for-service Medicare. Specifically, plans are beginning to establish programs to 1) identify chronic illnesses early, 2) intervene to prevent ill persons from becoming disabled, and 3) help disabled persons cope with and prevent further loss of function. Some plans are already well prepared to deal with elderly populations. Surprisingly, little mention of these positive aspects of managed care appears in the popular media.
Our most striking finding was the degree of variability in readiness in the seven domains of clinical programming. This variability is particularly noteworthy given our high refusal rate (24 of 80 plans refused to participate in our survey), which we would expect to bias our results toward high levels of readiness. The number and type of programs implemented varied widely across plans, within market areas, and even from medical group to medical group within plans in particular market areas. There was no obvious relation between readiness and any of the following factors: the plan's success in enrolling Medicare beneficiaries, the size of the plan, the length of time that the plan had been in business, the plan model (e.g., independent practice association or the staff or group model), or whether the plan operated independently or was affiliated with plans in other cities.
Our findings have several important implications. Many Medicare beneficiaries are now enrolling in plans across the United States that do not offer the full complement of programs and services from which they could benefit if they became chronically ill. Among the most common reasons cited for not adopting many of these services is a lack of evidence about the cost-effectiveness of particular interventions or the most effective methods or sequences of implementing programs aimed at chronically ill persons.
It is apparent from our data that in most health plans, the approach to policy with respect to these items is very decentralized. It is also apparent that most plans expect local management to determine policy about clinical programming. Surprisingly, even national plans do not seem to have processes in place to ensure that they replicate "best-in-class" programs--or even processes that clearly reduce costs--across the United States.
From a business perspective, the variation across plans in a single market is especially noteworthy. One would hypothesize that competitive pressures tend to remove differentials in offerings within a service area. Although our survey did not report on marketing activities, the high degree of local variation suggests that marketing promotions do not emphasize the availability of clinical programs for chronically ill and frail elderly persons. The plans' marketing departments may worry that promoting these programs might lead to adverse selection.
Finally, a considerable opportunity may exist for outsourcing care for persons at the highest risk for chronic illness who enroll in health plans through Medicare risk contracts. Plans seem to be moving deliberately to adopt clinical technology and show a preference for making (as opposed to buying) such capabilities. This may signal a "carve-out" business opportunity for organizations more accustomed to dealing with the needs of frail elderly persons. Offered services could address the gaps between desired and actual practice; for example, they could provide multidisciplinary care planning, mobilization of family and community resources, and management of functional outcomes.
|Take Home Points
1. Ware JE Jr, Bayliss MS, Rogers WH, Kosinski M, Tarlov AR. Differences in 4-year health outcomes for elderly and poor, chronically ill patients treated in HMO and fee-for-service systems: results from the Medical Outcomes Study. JAMA. 1996;276:1039-47.
2. Beck A, Scott J, Williams P, et al. A randomized trial of group outpatient visits for chronically ill older HMO members: The Cooperative Health Care Clinic. J Am Geriatr Soc. 1997; 45:543-9.
3. Capitman JA. Community-based long-term care models, target groups, and impacts on service use. Gerontologist. 1986;26:389-97.
4. HMO Workgroup on Care Management. Essential components of geriatric care provided through health maintenance organizations. J Am Geriatr Soc. 1998;46:303-8.
Gary Gaumer, PhD, Abt Associates, Inc., 55 Wheeler Street, Cambridge, MA 02138-1168; e-mail: email@example.com.