Health Insurance and Medicine in the 1980s

Health insurers have been challenged repeatedly in recent decades to respond to new medical treatments, technologies, and innovations. Slowly and haltingly they have developed institutions and strategies that ask for evidence of medical effectiveness as an input to coverage decisions. These developments have often brought them into conflict with medical innovators, as they did in the HDC/ABMT case.

Health insurance developed in the pre- and post-World War II period well before medical research began generating a continuing stream of new medical interventions. In recent years, insurers' response to new treatments has become a continuing challenge. "Medical necessity" provisions not only have anchored that response but also have revealed major problems with that reliance. Bergthold (1995) described medical necessity as "rarely defined, largely unexamined, generally misunderstood, and idiosyncratically applied in medical and insurance practice" (p. 181).

Historically, the term medical necessity has been a placeholder for health insurers that has served two functions over time. In the main, it has defined the medical services that are covered in insurance contracts under various benefit categories (i.e., hospital services, physician services, drugs, and durable medical equipment). In this respect, the term has essentially set the limits or boundaries of what is included or covered. In addition, typical medical necessity provisions, as defined in the evidence of coverage documents of insurers, have excluded coverage of experimental or investigational treatments a priori. Other provisions in a medical necessity definition that limit its scope under covered benefit categories include medical appropriateness for diagnosis, direct care, or treatment; standards of good medical practice; the most appropriate level of services (e.g., physician services, hospital services, drugs, durable medical equipment) that could safely be provided; and services not primarily for convenience (Bergthold 1995).

Elsewhere in health plan evidence of coverage documents, there are often additional exclusionary provisions that not only provide a general exclusion for experimental or investigational services (as specifically defined) but also exclude specific medical interventions or services, such as cosmetic surgery, dental implants, or in some cases during the 1990s, HDC/ABMT for breast cancer. A key point is that when coverage is excluded by benefit design (coverage categories) or by specific line item exclusions, medical necessity is irrelevant as coverage will always be excluded. Denials of specific services on the basis of the experimental or investigational exclusion or based on medical necessity, however, could be challenged on the basis of the process and the rationale for the decision. We discuss this in detail regarding HDC/ ABMT litigation in chapters 3 and 4.

Notwithstanding medical necessity clauses and coverage exclusions of experimental or investigational treatments, these provisions were seldom invoked rigorously under fee-for-service medicine. Coverage of and payment for new procedures was often almost automatic among many insurers. Continuing double-digit growth in health expenditures in the 1980s forced change and stimulated a number of efforts to rein in health care costs. In 1983, Congress enacted a system for prospective payment for hospital services by Medicare and, by the decade's end, extended that to outpatient services. Private insurers began to follow, and there occurred a general movement away from fee-for-service medicine.

New medical technologies constituted a major factor in health care cost increases and became one driver of cost containment. A number of expensive new technologies had diffused rapidly before evaluation, including dialysis and kidney transplantation in the 1960s, computed tomography in the 1970s, and magnetic resonance imaging in the 1980s (Institute of Medicine 1985; Rettig 1991). This phenomenon drew the attention of health economists, who began to analyze the elements of increasing costs of health care. They concluded that roughly half of the annual increase in costs of health care could be attributed to the effects of new medical technologies (Newhouse 1992; Weisbrod 1991). Insurers were quite aware of these effects and the pressure they exerted on insurance premiums. Major corporations, as purchasers of health care, also became sensitive to new medical technologies as a source of increasing costs for employer-financed health insurance.

The awareness that many new medical procedures and technologies were not always evaluated for either their medical effectiveness or cost implications provided a major stimulus to technology assessment in medicine. An important Institute of Medicine (IOM) report published in 1985 enumerated the methods of technology assessment: randomized clinical trials, the series of consecutive cases, case studies, registries, sample surveys, epidemiologic studies, surveillance, meta-analysis, group judgment, cost-effectiveness and cost-benefit analysis, and mathematical modeling.1 The appendix of that report and a later directory catalogued in great detail a large number of organizations engaged in technology assessment (IOM 1988).

One technology assessment approach emphasized group judgment and typically focused on the National Institutes of Health's (NIH's) consensus development conferences, which had begun in the mid-1970s. Consensus conferences raised the methodological questions of how physician experts make decisions about effective medical interventions when data are scarce or absent and about the validity of such decisions. A different approach developed in the mid-1980s, the systematic examination of the evidence of effectiveness as found in the medical literature. This movement toward evidence-based medicine would in time displace concern for the weaker methods of consensus development.

Technology assessment based on a systematic review of the evidence took hold among a number of insurers in the late 1980s and early 1990s as a way to make more informed coverage decisions (see chapter 7). The Blue Cross Blue Shield Association (BCBSA), representing 70-some independent Blue Cross plans, was a leader in this effort. In the mid-1980s, it established an internal analytical capability, the Technology Evaluation and Coverage (TEC) program, to evaluate systematically the effectiveness of medical interventions and advise the independent plans on coverage decisions. Assessments were available only to BCBSA member plans. A medical advisory committee, consisting entirely of Blue Cross medical directors initially, aided the program. (Craig Henderson was appointed as the first external member of this committee; other outside experts were added later.) Central to this effort were explicit and rigorous evaluation criteria (first established in 1985) for judging whether evidence of clinical effectiveness existed for a given intervention (see chapter 7, table 7.1). These developments would push the BCBSA into evaluating investigational procedures, which were becoming important for their coverage, reimbursement, and cost implications.

The TEC program was expanded in 1993 and renamed the Technology Evaluation Center. Kaiser Permanente joined as a sponsor of the program. The medical advisory committee was changed to include a majority of voting members with no Blue Cross affiliation, and Dr. Wade Aubry was named chairman. TEC ceased making coverage recommendations, restricting itself to determinations that an intervention did or did not meet the evaluation criteria. Assessments were made available to non-Blue Cross organizations by subscription. It has since generated a growing library of assessments of diagnostic and therapeutic interventions to assist member plans.

But, the BCBSA was not alone. Many other insurers and health plans developed technology assessment capabilities to aid in making coverage decisions. Similar efforts were initiated at Kaiser Permanente in southern and northern California that laid the foundation for joining the BCBSA effort. Group Health Cooperative of Puget Sound developed its own capability. In Minnesota, the Institute for Clinical Systems Integration, a collective effort of managed care organizations in that state, established both clinical guidelines and technology assessment efforts. Aetna created a technology assessment program, recruiting William McGivney from the American Medical Association (AMA). In addition, several analytical organizations, such as ECRI, began providing assessment services to smaller insurers (Rettig 1997).

For most of the health insurance industry, however, the inertia of medical necessity dominated coverage decision making well into the 1990s. The language of coverage was binary and blunt—either a procedure was covered or it was not. If experimental or investigational, then it was not covered. Not all insurers were committed to evidence-based coverage decision making. Policies, practices, and decisions varied greatly across insurers. In many cases, companies relied simply on the judgment of individual medical directors. As a result, the coverage decisions of individual health plans often appeared quite arbitrary at both the policy level and for individual patients (Newcomer 1990; Peters and Rogers 1994).

The evolution of technology assessment during this period allowed a number of major insurers and health plans to become much more sophisticated than they had been in making coverage decisions. The resulting increased scrutiny of new treatments often surprised both clinical researchers and community physicians, given the pattern that had existed under fee-for-service medicine. Some were aware of the development of technology assessment among health insurers. Many became aware of this development when a specific coverage request for a particular patient was denied. Many more were surprised at insurers asking for evidence of medical effectiveness.

0 0

Post a comment