Over 32 million elderly Americans are enrolled in the Medicare health insurance program. Since 1982, the Health Care Financing Administration has offered a managed care option, with capitated payment, to Medicare enrollees. Currently nearly 15% of Medicare enrollees are members of managed care plans nationwide. 1

The economic incentives of capitated payment have raised concerns that managed care organizations that use capitation might limit provision of effective services as a way of trimming costs. 2345 Public concerns about restriction of service in managed care organizations have led to proposed legislation for a "patient's bill of rights."6 once again being considered in the US Senate. 7 It is becoming clear, however, that "managed care" is not a single entity, and that a wide range of economic incentives and disincentives may be created in various forms of managed care, 8 each likely to have different kinds of influence on clinician behavior. 91011 The proximity of risk to the clinician or clinician group has an especially strong effect on behavior. 12 Systems maintained by managed care organizations to assist clinicians in providing quality care also differ widely, and may strongly influence performance.

Although reduced access to care is a theoretical risk of capitated payment, relatively few studies have addressed whether this concern is borne out in practice. Perhaps because of the variation in managed care arrangements, the results of studies have been mixed. 13 If managed care presents barriers to access to effective care, the elderly would be especially vulnerable because they have a higher burden of illness than the rest of the population. We therefore created indicators in several domains for which it was known that care improved outcomes. We then used these indicators to compare health care received by elderly Medicare beneficiaries in three kinds of payment arrangements: a capitated, staff-model health maintenance organization (HMO) and a partially capitated group-model HMO, both parts of the same managed care organization, and in fee-for-service (FFS) care in the same geographic area.

Demographic characteristics of the study subjects are shown in Table 2. The FFS population in the region under study was older than the nationwide Medicare population, and a slightly larger proportion was female (62% versus 60%, p < 0.001). Those enrolled in the HMO were younger than both the geographically matched, fee-for-service sample and Medicare enrollees in the US as a whole (p < 0.001), and they were less likely to be female (p 0.001).

Age-adjusted rates of breast cancer screening with mammography were substantially higher in the two HMO settings than under fee-for-service insurance. (Table 3). Mammography screening rates in the Staff Model were nearly double those in the FFS sample. The proportion of Medicare enrollees who saw a clinician within a one-year period (Annual visit rate, Table 3) was greater than 85% in all three groups; however, this rate was significantly higher in the Staff Model HMO (94%) than the fee-for-service sample (88%)(p < 0.001).

Proportion with follow up by either a primary care physician or cardiologist within 60 days of hospital discharge after myocardial infarction (Table 4) was similar in the two managed care divisions and, in both cases, higher than under fee-for-service insurance (91% and 93% in the Staff and Group Model respectively, compared to 73% in FFS; p < 0.001). Follow up within 14 days after hospitalization for depression was highest in the Group Model HMO, although small sample size limited our ability to find statistically significant differences (p > 0.3).

Care of chronic diseases is summarized in Table 5. Screening rates for diabetic retinopathy were higher in both divisions of the managed care organization than in the fee-for-service sample. The proportion of diabetics with 2 or more visits was 50% higher in managed care than in fee-for-service (95% and 91% in the Staff and Group Model, respectively, versus 61% in FFS, p < 0.001).

We developed a set of indicators of access to effective care for elderly people insured by Medicare designed to fit varying clinical settings, insurance structures, and data systems. In general, performance in both the Staff and Group Model divisions of the managed care plan studied was equivalent to or better than that in the fee-for-service sector. Other investigators have devised administrative claims-based indicators, 2122 and chart review-based indicators for hospitals, 23 and ambulatory care sites, 24 in efforts to compare hospital performance or health plan performance. In contrast, we explored the feasibility of using data elements available in an automated outpatient medical record, which are increasingly prevalent, in addition to those available in claims databases. The indicators we selected overlap with those chosen to be included in the Health Plan Employer Data and Information Set (HEDIS) for the Medicare population, released in 1997, which also evaluates breast cancer screening and retinal screening for diabetics. 25

Previous studies comparing performance indicators across insurance types have found inconsistent results. HMO performance on preventive services is consistently higher than in FFS settings. 262728 On the other hand, some studies' findings have raised a concern that vulnerable populations such as the chronically ill 29 and the elderly 30 fare less well in HMOs than in FFS care. In the Medicare population, data to support concern about worse outcomes of care in HMOs has come from a national telephone survey about persistence of symptoms. 29

In numerous other investigations, however, care for the Medicare population provided by HMOs has been shown to be comparable to that in FFS. No differences between HMO and FFS samples were found in studies of care for colon cancer, 3132 breast cancer,3334 prostate cancer,35 cardiovascular disease,36 rheumatoid arthritis,37 congestive heart failure, 38 hypertension, 39 acute myocardial infarction,40 acute appendicitis, 41 and stroke. 42 One study found that functional status did not decline faster among Medicare recipients enrolled in HMOs versus those in FFS. 43 Our findings are consonant with these previous studies: for example, a higher rate of mammography screening as seen in our managed care population could explain the trend to earlier stage breast cancers seen in California among HMO enrollees compared to FFS.31, 32 While most of previous work focussed on specific disease entities, we studied the general Medicare population and measured indicators applicable to all enrollees, as well as indicators focussed on specific disease-populations.

There are few studies of the effect of managed care structure, (e.g. Group vs. Staff vs. IPA models) on indicators of quality care. 4445 These models differ both in the extent to which financial risk is attached to individual providers, and also the degree to which providers are effectively formed into an organized system of care. For example some of the differences seen in our study between the Group and Staff Model were probably due to organizational processes and systems in place in the Staff Model, such as physician reminder systems for preventive services leading to higher rates of breast cancer screening.

This study has several strengths. We included a large number of Medicare enrollees covered by the health plan studied, and the universe of fee-for-service Medicare enrollees in the same geographic region. Matching by region reduces the variation that may have affected some previous comparisons of FFS and HMO populations, in which national samples were used. 46 This method also automatically accounts for any "spillover" effect of managed care practices onto fee-for service practice, wherein cost-saving techniques are applied widely by providers, even to those with fee-for-service insurance, in a market with high penetration of managed care insurance. 47 In contrast to studies that focus on inpatient data (in part because of their ready availability,)48 our focus on outpatient care is highly relevant to comparisons of managed care with fee-for-service care49 in terms of ambulatory practices that may prevent hospitalizations. Our work on the time period 1994–95 follows the transformations in managed care in 1991–1993, and thus can been seen as an update to previous studies of the topic.13

Our study also has limitations. People enrolling in managed care organizations may be systematically different from those choosing fee-for-service arrangements. 50 For example, patients who are sicker or more disabled may not enroll,28 or may differentially choose to leave managed care. 51 Those who do enroll could be more aggressive in seeking care. We know that a smaller proportion of HMO enrollees were over the age of 85 years, compared to the matched fee-for-service sample. Differential enrollment could influence our results if patients with higher degrees of co-morbid conditions preferentially enrolled in the fee-for service system and did not participate in cancer screening or secondary diabetic prevention because of their other illnesses. 52 In addition, the annual visit rate could potentially be overestimated in the HMO population if an influx of new patients and a "new enrollee" effect were to cause more patients to come in for visits than otherwise would. Another potentially important limitation is that different data systems created for different purposes, such as claims and clinical databases, may not provide comparable results, even for strictly defined indicators.5354

The use of performance indicators to compare care provided in different settings usually requires expensive data gathering. Indicators based on automated data systems offer partial relief to this problem. Our findings related to staff model and group model HMO types should be replicated in other settings, and extended to other types of managed care. While the staff model HMO share of the market is shrinking, group and network model health maintenance organizations are experiencing increased enrollment. Our finding of concordance on many indicators between the Group and Staff Model divisions suggests that a less tightly organized provider system may perform as well as a staff model HMO. Neither of the HMO divisions studied resembles an independent practice association (IPA) model, so our findings may not apply to that form of provider organization. In addition we studied only one managed care organization, one that is well established, and exists in a region of the country with high level access to specialty services. The generalizability of our findings to other managed care populations is unknown. Furthermore, the single HMO studied is a not-for profit organization, and our findings might not be applicable to for-profit entities. 55

In conclusion, we believe that the indicators we have tested are feasible to implement in other managed care systems, and could be translated into other electronic medical record systems. Indicators such as these might be useful for consumers in comparing health plans to one another, such as those participating in Medicare+Choice. We found no evidence of any decrement in access to health care for the elderly Medicare population enrolled in one managed care plan in this geographic region. The low- or no cost "supplemental" insurance provided by many HMOs may in fact lower financial barriers to care for low-income seniors. Our study describes only two models of managed care, and this is a crude classification by current standards. Yet to be studied are the effects of financial incentives on increasingly smaller units of leverage, e.g. small IPA model groups, and the effectiveness of managed care plans that use sophisticated information systems to track and improve care, as compared to those that use principally financial incentives. 5657 The latter is crucial to understanding how best to create a partnership between providers and health plans that facilitates the provision of excellent care.

None declared