Medical Necessity in Private Health Plans

Medical Necessity and the Published Literature

Table 1 presents definitions of medical necessity drawn from a search of peer-reviewed journals, trade journals, and industry and organization publications. A full list of these sources appears in Appendix A. Over the past decade, authors have paid considerable attention to the question of medical necessity as prospective utilization review has come to dominate health insurance.

While variation exists in the opinions expressed, the articles summarized in Table 1 display a significant level of consensus on three basic issues. The first is that merely because a recommended treatment falls within the zone of professionally accepted medical practice does not mean it must be covered. Only one source (the National Health Law Program) confines the evidence to the opinion of the treating physician. The second is that a recommended definition of medical necessity should be multidimensional and should consider factors such as cost, convenience, and relative effectiveness compared to other treatments based on various forms of evidence. Third, the authors uniformly recommend broadening the scope of when an intervention can be considered necessary (i.e., not merely to diagnose and treat an illness but also to improve functioning, avert deterioration, and maintain functioning).

Several authors address the issue of the quality, reliability, and relevance of the evidence considered when making a medical necessity determination; in addition, one article examines the question of who bears the burden of proof in a medical necessity determination, an issue that has not been directly addressed in State or Federal law.

From the health services research community, probably the most seminal work is by Singer, Bergthold, Vorhaus, and Enthoven (1999). The definition of medical necessity they crafted was the result of a consensus process among project participants (Singer, Bergthold, Vorhaus, & Enthoven, 1999):

For contractual purposes, an intervention will be covered if it is an otherwise covered category of service, not specifically excluded, and medically necessary. An intervention is medically necessary if, as recommended by the treating physician and determined by the health plan's medical director or physician designee, it is (all of the following): A health intervention for the purpose of treating a medical condition; the most appropriate supply or level of service, considering potential benefits and harms to the patient; known to be effective in improving health outcomes. For new interventions, effectiveness is determined by scientific evidence. For existing interventions, effectiveness is determined first by scientific evidence, then by professional standards, then by expert opinion; and cost-effective for this condition compared to alternative interventions, including no intervention. "Cost-effective" does not necessarily mean lowest price. An intervention may be medically indicated yet not be a covered benefit or meet this contractual definition of medical necessity. A health plan may choose to cover interventions that do not meet this contractual definition of medical necessity.

This definition requires a review of the treating clinician's recommendation to ensure that it is "for the purpose of treating a condition" and "the most appropriate" intervention in light of the patient's particular condition, benefits, and risks. The definition also assumes plan review of the provider's treatment recommendations. The authors also contemplate that cost-effectiveness will be a basic element of the decision, but clarify that the question of cost-effectiveness is not one of price alone. In addition, the authors create a hierarchy of evidence, with "scientific" evidence classified as the best evidence. No distinction is made by type of condition.

Of particular significance in the Singer/Bergthold analysis is its emphasis on the primacy of coverage limitations, a major concern of insurers. The authors recognize that once a particular type of treatment is excluded for a specific condition as a contractual matter,³ no general finding of medical necessity can override the exclusion. This emphasis on the primacy of the contract in controlling the range of treatments and procedures that will be considered at all in a medical necessity determination is reinforced by the Health Insurance Association of America (Schiffbauer, 1999), which has stated:

When the provider, rather than the health plan or insurer, interprets the scope of coverage under the contract, health plan fiduciaries cannot guarantee to the insured that health care dollars are being spent fairly and equitably on medical treatments that are safe, proven, and effective.

The American Medical Association (AMA), representing physicians (including psychiatrists), has created a prototype medical necessity definition as part of its Model Managed Care Contract project:

Section 1.9 defines medically necessary/medical necessity as health care services or products that a prudent physician would provide to a patient for the purpose of preventing, diagnosing, or treating an illness, injury, disease, or its symptoms in a manner that is a) in accordance with generally accepted standards of medical practice; b) clinically appropriate in terms of type, frequency, extent, site, and duration; and c) not primarily for the convenience of the patient, physician, or other health care provider.⁴

Like the others this definition is multidimensional but it focuses the utilization review on what a prudent physician would conclude based on the evidence rather than what the insurer would determine. While the definition is crafted in such a way as to transfer more medical decisionmaking power back to the provider, the practical impact of this distinction is difficult to assess, since the decision remains reviewable and the review is multidimensional. However, cost considerations as an explicit measure are removed. By using the "prudent physician" rather than the insurer as the standard of measurement where judgment is concerned, the definition seeks to focus the determination on "generally accepted" medical opinion (and thus the phenomenon of multiple schools of thought) rather than the opinion of utilization review professionals who may or may not be physicians and who view their task as selecting the single best form of treatment. Thus, in an appeal made under the AMA definition, a claimant would be able to introduce a wide range of schools-of-thought evidence from "prudent physicians" to show the variation in treatments that prudent physicians might recognize.

Several authors focus on definitions of medical necessity in the behavioral health arena, although their proposed definitions appear to differ more in terminology than in substance. Paul Chodoff (1998) and William Ford (1998, 2000) have called for replacing the term with "health necessity," "treatment necessity," or "clinical necessity." In Chodoff's view, health necessity criteria would be founded on a biopsychosocial rather than on a medical model. The former model requires a view of health as encompassing quality-of-life factors and not just the absence of disease. The terms "biopsychosocial" and "psychosocial" arose from the need to differentiate between mental and physical health.⁵

The practical effects of this distinction would be on the "scope" element of the definition, that is, the range of possible conditions for which treatment, if necessary, would be approved. Interventions would not be solely for the diagnosis or treatment of an illness, but also for the achievement of broader health goals. Furthermore, Chodoff proposes consideration of services for individuals whose diagnoses may not easily fit into categories defined by the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), a reference often cited as a clinical standard in medical necessity definitions for behavioral health (APA, 1994).

Ford's (1998, 2000) behavioral health care definition urges a movement away from covering only acute care to covering longer-term care designed to manage and prevent deterioration of chronic conditions and onset of acute conditions. This definition would include access to psychiatric rehabilitation services when needed for the treatment of chronic mental conditions. (This definition of course would require a dramatic expansion of the terms of coverage under conventional insurance, which, unlike Medicaid, tends to be confined to relatively short-term therapies to help an individual significantly improve or recover in a relatively short period of time.) (Rosenbaum, forthcoming; Rosenbaum & Rousseau, 2001) Like Chodoff, Ford stresses the importance of both the quality of day-to-day functioning as a goal of treatment, and the need to cover treatment designed for alleviation of symptoms in addition to "cure."

Ireys, Wehr, and Cooke (1999) propose a specific definition of medical necessity for persons with developmental disabilities, mental retardation, and other special health care needs. Their article represents a detailed and specific attempt to articulate individualized decisionmaking criteria that can "assist the individual to achieve or maintain sufficient functional capacity to perform age-appropriate or developmentally appropriate daily activities."(p. 19) The authors call for an expanded view of the information sources an insurer should consider beyond "medical evidence" (i.e., information from the patient, the family, collateral providers, and support institutions). They also emphasize the effect of treatment on day-to-day functioning and require that final determinations be made by a physician employed by the insurer (rather than a claims reviewer with lesser qualifications).

Two articles (Appendix A) deal specifically with evidentiary matters and the use of evidence in decisionmaking. David Eddy (1994) posits that when determining the appropriate use of an intervention, analysis of its potential value should shift from qualitative to quantitative, with use of randomized, controlled clinical trials as a definitive evidence base. Furthermore, in his view, a shift from individual-based decisionmaking to population-based decisionmaking is needed, based largely on the utility of controlled clinical trials that demonstrate treatment efficacy across large numbers of people. He advocates for the development of explicit criteria to sort out high-value practices from those of little or no value and believes that the term "medical necessity" is too vague and open to too much variability in interpretation. By contrast, Rosenbaum, Frankford, Moore, and Borzi (1999) recommend an emphasis on individualized decisions rather than across-the-board conclusions based on the application of generalized guidelines and research results to specific cases. They call for strict scrutiny of the reliability and relevance of scientific evidence, as well as for greater emphasis on the facts of an individual case and expert judgment. They also recommend shifting the burden of proof to the health plan in any review of its decision on medical necessity, arguing that the plan has best access to the evidence, and that fairness in allocating the burden of proof would place the burden on the party with the best access to evidence.

Sabin and Daniels (1994) address the question of the utility of medical necessity definitions for mental health services from the perspective of severity of diagnosis. While no question exists that severe mental illness such as schizophrenia, clinical depression, and bipolar depression are covered by traditional medical necessity definitions, Sabin and Daniels investigate the extent to which such definitions also should cover conditions such as shyness, unhappiness, and lack of personal fulfillment. Using six illustrative case studies, such as "The Shy Bipolar," "The Unhappy Husband," "The Cranky Victim," (pp. 5-7) and others, Sabin and Daniels illustrate the differences of opinion between "hard-line" and "expansive" clinicians (p. 5) in deciding whether psychiatric services are needed. Following an analysis of three models of medical necessity, the authors conclude that the most rational model is one that treats a medically defined diagnosis, such as one delineated in the DSM-IV, to decrease the impact of disease or disability. A typical mental health medical necessity definition would be "those mental health services which are essential for the treatment of a Member's mental health disorder as defined by the DSM-IV in accordance with generally accepted mental health practice"(p. 12). Sabin and Daniels note that diagnostic categories continue to change but that society "needs a publicly acceptable and administerable system for defining the boundaries of health insurance coverage." To that end, the DSM-IV (and subsequent editions) provides a workable definition of those boundaries, to the extent that it is "the result of a highly public process open to scientific scrutiny, field testing, and repetitive criticism over time."⁶

Table of Contents | Previous | Next