Effective Clinical Practice
Effective Clinical Practice, January/February 1999
Dorothy A. Bazos, Elliott S. Fisher
For author affiliations, current addresses, and contributions, see end of text.
Context. The Medicare program has promoted capitation as a way to contain costs. About 15% of Medicare beneficiaries nationwide are currently under capitation, but tremendous regional variation exists.
Practice Pattern Examined. The proportion of Medicare beneficiaries who have enrolled in risk-contract plans in individual states and in the 25 largest metropolitan areas in the United States.
Data Source. Health Care Financing Administration data files.
Results. Medicare beneficiaries are most likely to be under capitation in Arizona (38%) and California (37%). Eight other states have capitation rates greater than 20%: Colorado, Florida, Rhode Island, Oregon, Washington, Pennsylvania, Massachusetts, and Nevada. Thirty states, largely in the Great Plains area and the southern United States, have capitation rates less than 10%. Four major metropolitan areas have market penetration rates greater than 40%: San Bernardino, California; San Diego, California; Phoenix, Arizona; and Miami, Florida. Little penetration exists outside of metropolitan areas.
Conclusion. Capitation in Medicare is a regional and predominantly an urban phenomenon.
The U.S. Congress expects increased enrollment in capitated health plans (risk-contract plans) to slow the rate of growth of Medicare spending. During the first 10 years in which risk-contract plans were available to Medicare beneficiaries, enrollment in these plans increased slowly, reaching only 4.4% by 1992. (1) Since then, enrollment has surged: It reached 15% in 1998. (2) Although recent decisions by some health plans to stop serving Medicare beneficiaries may impede further growth, Medicare capitation continues to be promoted as a strategy for cost containment. Congress has enacted legislation to implement the Medicare Plus Choice program--a plan to broaden the range of organizations allowed to offer capitated plans to Medicare beneficiaries.
In addition to addressing Medicare's interest in cost containment, these capitated health plans offer elderly persons some advantages over traditional fee-for-service plans. Almost all offer expanded benefits, such as free preventive services, low-cost or free prescription drugs, and reduced cost sharing. Most require no additional premiums.
Despite these attractive features, the growth of Medicare risk plans has been uneven. The number of risk plans has steadily increased since 1987, but the areas of greatest growth have been concentrated in a few states and largely in metropolitan areas. (3) In this paper, we use the most recent data on enrollment in Medicare capitated plans to provide an update on this regional variation. We focus our analysis on market penetration in individual states and major metropolitan areas. In addition, we compare variation within states and metropolitan areas with that in nonmetropolitan areas.
We obtained the June 1998 report on Medicare HMO enrollment by plan, county, and state from the Health Care Financing Administration (HCFA) (www.hcfa.gov/medicare/mpscpt1.htm). We then added these data to create summary files for each state, for each Metropolitan Statistical Area (MSA), and for the non-MSA counties for each state. For each geographic area, we determined the proportion of Medicare-eligible beneficiaries enrolled in risk-contract HMOs and the number and proportion of enrollees in each major plan in that area. The characteristics of each plan were ascertained from HCFA reports (Medicare Managed Care Monthly Report [www.hcfa.gov/stats/monthly.htm]).
Figure 1 highlights the variation of market penetration for Medicare risk contracts nationwide. Medicare beneficiaries are most likely to be under capitation in Arizona (38%) and California (37%) and are least likely to be under capitation in West Virginia and North Dakota (both 0.06%). Eight other states have enrollment rates greater than 20%: Colorado, Florida, Rhode Island, Oregon, Washington, Pennsylvania, Massachusetts, and Nevada. On the other hand, 30 states have fewer than 10% of eligible beneficiaries enrolled in risk-contract plans. More than half of the Medicare beneficiaries enrolled in risk-contract plans live in four states: Arizona, California, Florida, and Pennsylvania.
Enrollment data for the 25 largest metropolitan areas in the United States (largest in terms of the number of Medicare beneficiaries) are summarized in Table 1. The percentage of Medicare beneficiaries enrolled in capitated plans ranges from 51.8% in San Bernardino, California, to 5.9% in Detroit, Michigan. Table 1 also shows that enrollment is concentrated in a few plans. The market share for the top three plans ranges from 63% in Cleveland, Ohio, to 100% in Pittsburgh, Pennsylvania; St. Louis, Missouri; and Minneapolis, Minnesota. When all metropolitan areas in which at least 10% of Medicare beneficiaries are enrolled in capitated plans (n = 124) are considered, the 3 top-ranked plans control 60% to 100% of the market. The Appendix Table provides details on the identity of the largest plans in the 25 largest metropolitan areas (Pacificare or Kaiser Foundation in one third of cases).
Metropolitan versus Nonmetropolitan Areas
Although 18% of Medicare beneficiaries living in metropolitan areas are enrolled in capitated plans, fewer than 3% of beneficiaries living outside of these areas are enrolled. Figure 2 summarizes this difference for the states with the highest proportion of enrollees in risk-contract plans overall. Even in the states with the most capitation overall, few beneficiaries in nonmetropolitan counties are in capitated plans: 17.1% in Arizona, 6.3% in California, 5.8% in Oregon, 3.5% in Colorado, and 7.4% in Florida.
Market penetration of risk contracts in nonmetropolitan areas is uneven and seems to be unrelated to that in metropolitan areas. In Arizona and California, where the percentage of risk contracts for metropolitan areas is high (42% and 38%, respectively), approximately a threefold difference in market penetration is evident outside of metropolitan areas (17% in Arizona and 6% in California). In Massachusetts and Connecticut, where metropolitan penetration is lower (22% and 19%, respectively), similar variation in nonmetropolitan areas is found (5% in Massachusetts and 15% in Connecticut).
The proportion of Medicare beneficiaries enrolled in capitated plans varies dramatically across communities in the United States. Enrollment is concentrated in several states and is low in rural areas.
This regional variation might be the result of enrollee preferences or plan availability. Not much is known about the opinions of Medicare enrollees on "managed care" and how these opinions differ across the United States. It is known, however, that capitated plans are simply unavailable to Medicare beneficiaries in many areas of the country. As of 1996, 37% of Medicare beneficiaries lived in areas that had no plans available. The availability of plans varied across states, with half of the beneficiaries in 27 states having no plans available. (3)
Plan availability is particularly influenced by two factors. First, variability in payment rates influences plan availability and the benefit package that plans can offer. Monthly payments to the plans, which have long been based on the average per capita costs of fee-for-service care in the enrollee's county of residence, range from $221 in one Nebraska county to $748 in Miami, Florida. (3, 4) Payment rates are highest in urban areas and lower in rural areas. Plans in areas with higher payments are generally able to offer greater benefits and, consequently, to enroll more beneficiaries. In addition, areas of low population density (i.e., nonmetropolitan areas) have payment rates that are less predictable than those in metropolitan areas. McBride and colleagues (5) report that the local volatility of payments to plans is 4.05% in rural counties that are not adjacent to urban areas and 2.59% in urban counties.
Second, variability in the local infrastructure influences plan availability. The Medicare population is viewed by health plans as the last untapped market for expansion. To keep Medicare risk contracts viable, plans pay particular attention to developing provider networks, improving access to specialty care, and increasing ambulatory case management and outpatient services. A focus on medical utilization management targets cost savings from inappropriate or unnecessary services. (6) The infrastructure, personnel, and network of supports required to provide services managed in this way are often not developed outside of metropolitan areas.
Medicare managed care must be understood in its local context. It is important to recognize that studies reporting national data primarily reflect the practice patterns of a relatively small number of plans in several states. Data reported at the plan or market level would be more informative. Disenrollment rates, for example, are five times higher in Florida than in Minneapolis. (7) Surveys now under way will provide detailed data on all Medicare risk plans.
Geographic variation in Medicare managed care availability suggests that expansion will continue to be constrained by plan availability. Although Medicare Plus Choice has increased the types of organizations that are able to offer Medicare risk plans, many beneficiaries may not have access to them. Many beneficiaries live in states that have less than 10% market penetration, and fewer than 3% of beneficiaries who live in rural areas (one quarter of all Medicare beneficiaries) are enrolled in risk plans. The low population density of rural areas contributes to lack of competition among plans, volatility of payments, and inadequate local infrastructure. These factors are likely to remain barriers to increasing capitation rates. A "one-size-fits-all" federal policy for Medicare beneficiaries may be impractical. New incentives or models for care are needed for nonmetropolitan areas.
|Take Home Points
1. Lamphere JA, Neuman P, Langwell K, Sherman D. The surge in Medicare managed care: an update. Health Aff (Millwood) 1997;16:127-33.
2. Office of Managed Care, Health Care Financing Administration "Medicare Managed Care Market Penetration for All Medical Plan Contractors—Quarterly State/County/Plan Data Files." June 1998. http://www.hcfa.gov/medicine/mpscpt1.htm.
3. Physician Payment Review Commission. Annual Report to Congress. Washington, DC: Physician Payment Review Commission; 1997.
4. Wennberg J, Cooper M. The Dartmouth Atlas of Health Care in the United States. Chicago: American Hospital Publishing; 1996.
5. McBride TD, Penrod J, Mueller K. Volatility in Medicare AAPCC Rates: 1990-1997. Health Aff (Millwood) 1997; 16:172-80.
6. Hurley RE. Expanding Medicare managed care. J Healthc Resour Manag. 1996;14:17-23.
7. Riley GF, Ingber MJ, Tudor CG. Disenrollment of Medicare beneficiaries from HMOs. Health Aff (Millwood). 1997;16: 117-24.
Dorothy A. Bazos, RN, MS, Center for the Evaluative Clinical Sciences, Dartmouth Medical School, Hanover, NH 03755; e-mail: firstname.lastname@example.org.