10.

Health maintenance organizations (HMOs) have been described by their advocates as a "cost-effective" way to provide health care (489). It appears that HMOs do provide care to their enrollees at varied but substantial cost savings: Empirical evidence shows that HMO enrollees pay in the range of 10 to 40 percent less in total costs (premium plus out-of-pocket costs) than conventionally insured comparison groups (373).1

Explanations of HMO performance, whether measured by costs, access, quality, or physician or enrollee satisfaction, however, are a hybrid of rhetoric, theory, and evidence. There is very little information available on the details of how HMOs actually function, especially on the analytical tools they use to make internal decisions on how to allocate their resources (295). This chapter presents some preliminary evidence regarding the current and potential use and usefulness of cost-effective analysis/cost-benefit analysis (CEA/CBA) in resource allocation decisionmaking in HMOs. It does not, however, examine directly the question of the cost effectiveness of HMOs themselves.

HMOs, representing the prepaid segment of the health care market, are both insurors and providers of health care (300). Their revenue comes from capitation payments, and they are responsible for delivering care to an enrolled population (371). Theoretically, the HMO has a direct economic incentive to provide "costeffective" care. Because of this, some people have assumed that the HMO might be more receptive than conventional health care delivery settings to the use of CEA/CBA (615).

It is important to note, however, that HMOs exist in a predominantly fee-for-service environ

'This range reflects the differences observed between types of HMOs. Prepaid group practices consistently show the lowest total (premium plus out-of-pocket costs), though there is much variation among them; individual practice associations show among the highest costs (372).

ment. Since potential enrollees have a choice of providers, and physicians a choice of practice settings, the HMO must compete for both. Because its financial viability depends on its ability to attract and retain enrollees (as well as physicians), the HMO will be induced to offer benefits and services comparable to those offered by its fee-for-service competitors.

While the distinctive characteristics of the HMO guarantee control of total costs (by virtue of capitation payments) and promote efficiency (by virtue of the HMO's responsibility to deliver appropriate services to an enrolled population within a constrained budget), there is nothing to require the HMO to be any more or less concerned with the effectiveness and benefits of a service than are providers in other health care delivery settings: Rather, the benefits side of the CEA/CBA problem is largely accepted as being defined by the norms of "good medical practice" that prevail in the marketplace. The expectations of HMO enrollees and physicians tend to diminish the importance of resource allocation questions pertaining to whether or not a benefit or service should be offered.

In general, the HMO plan bears primary responsibility for the financial viability of the HMO (in a management or administrative sense). The plan faces explicit resource allocation decisions upon which economic incentives and financial constraints directly bear. Whether or not physicians are sensitive or responsive to these incentives and constraints in making their implicit resource allocation decisions, however, is the subject of considerable debate (152,371, 589,590). Some claim that physicians are unaware of incentives and constraints, and even point to complete coverage as giving them a free hand (372). Yet others assert that physicians in the HMO setting are sensitive to costs (at least on a subliminal level) and that even a decision, for example, to treat bronchitis, or to perform a coronary bypass, specifically involves the aspect of costs (619).

In fact, the available evidence shows that almost all of the observed cost savings in HMOs are attributable to lower rates (by 30 percent) of hospitalization for HMO enrollees (373). Decisions concerning the need for hospitalization of HMO enrollees are primarily made at the discretion of attending physicians. It is true that the resource allocation decisions made at the administrative level will define the parameters of decisions faced by HMO physicians. Apart from direct and indirect controls imposed by the plan of the HMO, however, there is little consensus on any one explanation of why HMO physicians should exhibit this apparent "cost consciousness." HMO physicians, trained no differently than physicians practicing in the feefor-service environment, bring with them habits and values acquired in medical school or in some other prior practice setting (300). Theoretically, there is no reason to expect that HMO physicians would be any more likely than their fee-for-service counterparts to explicitly consider the aspect of cost, or beyond that, to consider cost concomitantly with the effectiveness of a service.

If anything, the HMO physician can be expected to allocate the available resources to a given medical problem foremostly on the basis of effectiveness and/or efficacy, differing in this respect from the plan, which will be motivated to allocate resources in significant part on the basis of costs, i.e., efficiency. No further attempt will be made in this chapter to investigate

the use and/or usefulness of CEA/CBA in the decisionmaking of HMO physicians.

The preliminary evidence and conclusions that are presented in the discussion that follows pertain largely to the resource allocation decisions made at the organizational, i.e., administrative, level of the HMO plan. The evidence discussed in this chapter should be regarded as tentative because of the great diversity in types of HMOs and the small number of HMOs from which the evidence was gleaned. The influences assumed to determine the resource allocation questions relevant to the HMO theoretically derive from "generic" characteristics of HMOs that distinguish them from other health care. providers. It is important to note, however, that there is probably no such thing as a "typical" HMO. There is a tendency (in the rhetoric and limited literature available on HMOs) to juxtapose HMOs against "other providers," but this has masked important differences among HMOs (295).

The findings presented here do seem to substantiate the preliminary conclusion that formal CEA/CBA, except in its "net cost analysis" forms, is not used to any significant degree in decisionmaking in HMOs. In large part, the evidence is based on: 1) actual resource allocation decisions recently made in a handful of HMOs around the country, 2) the analytic techniques those HMOs used, and 3) the role that analysis played in those particular decisions.

HMOs have occupied a prominent position in Federal health policies during the last decade. They have been promoted as one strategy for controlling health care costs and encouraging a more rational allocation of resources to health care needs (295). Much of HMOs' original and

In 1979, the Office of Health Maintenance Organizations (OHMO) contracted for a CBA to examine the economic costs and benefits of continued Federal assistance to new HMO development. That study determined the estimated rate of return (in costsavings to the community) derived from the Federal HMO development investments. It found that Federal assistance costs are re

sustaining appeal derives from the fact that these organizations have been viewed as a more desirable alternative than Federal regulation for achieving those objectives (95).

The term "health maintenance organization" was the brainchild of the Federal HMO initia

covered in the form of community health care cost savings after 8 years of HMO operation, and projected even more substantial future savings (302). Based on these findings, OHMO has devised a 10-year strategy for focusing Federal funds and support of HMO development in those areas that offer the greatest potential return (489).