The concept of medical necessity first crossed the medical horizon in the 1940s when it became necessary to describe the services for which doctors and hospitals would receive insurance payments. No specific definition was offered or seemed needed at the time, but the situation changed in the 1960s when disputes arose between providers and insurance companies.
An example of a definitional attempt was that of Medicaid (1): "accepted medical practice or community standards of care; not for the convenience of the patient or provider; not experimental or investigational; and appropriate and effective." This loose definition does not afford adequate guidance for the gatekeeper function of medical necessity. Attempts to clarify and operationalize the concept—notably those by President Clinton's task force on health care reform (1) in 1996—that have focused on either the medical part (2) or the necessity part (1) of the definition have all been unsatisfactory. What is meant by medical necessity remains ambiguous. That point was confirmed by a recent study from the General Accounting Office, which found a substantial variation in rates of denial of claims for lack of medical necessity among different insurance plans (3).
Thus, although current practice decisions about medical necessity are still nominally under the control of physicians, they are actually strongly influenced by the deliberations of courts and by consumers and insurers. As has been noted, although "medical necessity" sounds scientific and objective, that is not, in fact, the case (4). How did this state of affairs come about?
In these comments I describe the limitations of the current concept of medical necessity, especially for psychotherapy, and I propose substituting the concept of "health necessity," which would enable practitioners of psychotherapy to directly claim the necessity of their services.
I submit that any attempt to define "medical" in terms of a physician's specific activities rather than to accept as "medical" whatever a physician does in good faith is bound to flounder in a definitionally impassable swamp. I believe that much of this confusion is derived from the inherent lack of preciseness in the concept of disease and illness (5).
In 1951 Parsons (6) introduced the terms "medical model" and "sick role," which still form a bedrock for the medical domain, although no consensus exists about the definition or limits of these terms. Things would be simpler if we accepted the narrow view of Thomas Szasz (7) that "medical" has meaning only in the context of an organic lesion. However, society has accepted a considerably broader view that now underlies a lack of clarity about medical necessity and enables various parties to introduce parameters and limits on its use in ways that serve their interests.
But whatever the true or apparent intent, medical necessity must be dealt with by practitioners whose patients or clients rely on insurance reimbursement and who are under the scrutiny of managed care organizations. All physicians, whatever their specialty, must deal with medical necessity. Thus, when in general medical practice the components of a benefit package need to be specified, certain questions must be answered, such as "When is mammography medically necessary?" or "When are bone marrow transplants for leukemia medically necessary?" or, looking down the road, "What are the possible benefits of cosmetic gene therapy?" Are prescriptions of medications to help a healthy male grow hair on his head or to improve his erectile potency medically necessary? Under what conditions should health maintenance organizations approve home nursing care?
Psychiatrists have more trouble with medical necessity than their colleagues in other specialties. On the grounds of cost, insurance companies justify the special restrictive treatment imposed on psychiatrists. I suggest that lurking behind this doubtful rationalization is the old conviction that psychiatrists aren't real doctors, that they don't treat really sick people, and that their treatments therefore cannot conform to medical necessity. This conviction seems to be implicitly accepted by the National Alliance for the Mentally Ill when, in their lobbying for parity, they ask for equal coverage only for severe mental disorders, which they label brain diseases.
The difficulties medical necessity presents for psychiatrists in general are compounded for psychiatrists whose main treatment modality is psychotherapy and who are subjected to stringent insurance restrictions. One of the factors responsible for this attitude—that psychotherapy is practiced widely by nonphysicians—apparently rattles the bones of the medical model still imbedded in the present-day concept of medical necessity.
However, the chief problem is the type of patient treated with psychotherapy, especially when treatment is unaccompanied by prescriptions for medications. Questions can be raised, for instance, in the case of patients with some personality disorders. For example, at what point and by what criteria does a scrupulous, orderly, worrisome individual become a person with a compulsive personality disorder and thus merit insurance coverage for psychotherapy?
The psychotherapy patients who pose the greatest challenge for medical necessity are those treated for the rather vague category of problems known as "problems of living." Let me summarize a brief example reported previously by Sabin and Daniels (8). A single woman in her mid-thirties with a successful career and a good social life entered therapy complaining of feeling empty and lost. These feelings were related to her inability, despite numerous efforts, to become involved in a long-term, committed relationship with a man. This woman had a real problem, and there was a good chance she could benefit significantly from psychotherapy. But was she "sick," and how could she be diagnosed?
Cases like the one above pose the problem of the distinction between illness and the degree of psychopathology sufficient to limit the ability of some people to lead satisfactory lives. But psychopathology, in the sense of a deviation from some hypothetical model of perfect functioning, is universal and cannot always be equated with illness. Acknowledging this universality is certainly not to deny that beyond a certain point of intensity and distress, psychopathology can be included under broader versions of the medical model that do not mandate the presence of an organic disorder and thus can satisfy medical necessity. However, if we accept present conceptualizations of medical necessity as binding, we may be going too far if we include every interpersonal or relational problem that comes to psychotherapeutic attention.
I conclude that medical necessity, in its present incarnation, fails to function as a rational gatekeeper for appropriate health services. Lacking a satisfactory definition, it has become cumbersome and unwieldy and in some ways almost meaningless. Operationally, it seems to consist of an inherited medical model skeleton with emphasis on organicity, which is useful in some areas of medicine but which is imbedded in a thick carapace of issues with only a tangential relationship to medicine.
Decisions based on medical necessity seem to be derived from broad considerations of policies that can be criticized as being instituted more to reduce costs or ensure profits than to benefit patients. Medical necessity has placed psychotherapists in a position of defensive mistrust with respect to the insurance industry, forcing them to endure irksome restrictions on therapeutic decision making and even to fudge diagnoses in their patients' interests.
In two articles in the September 1996 issue of this journal, Bennett (9) and Borenstein (10) each thoughtfully analyzed the place of psychotherapy in an era of managed care and defended its compatibility with medical necessity criteria. However, rather than pursue attempts to fit medical, psychiatric, and psychotherapeutic procedures to an unyielding procrustean bed, I suggest that the time has come to rid ourselves of medical necessity and its baggage and to undertake the difficult task of establishing new definitions and new criteria for health coverage.
A possible beginning would be to substitute the concept of "health necessity" for medical necessity. This broad concept would rely on medical criteria when they are relevant but would also acknowledge that the health of the citizenry can be perceived in broader terms. A theoretical foundation for this concept may be found in the biopsychosocial model.
Health necessity would be based on three broad fundaments:
• Uniform qualifications for practitioners, acceptable professional identities, and competence
• Criteria for the kinds of services that would be provided and covered
• A fair mechanism for resolution of disputes about questions of service coverage.
The criteria for services to be covered would include biotechnical medical criteria when appropriate, as would be the case in most ordinary medical practice, but they would be acknowledged to be only a subset of the health necessity criteria. For mental health needs, a broad range of services could be considered (2), including appropriate psychotherapy for individuals who may not fit comfortably within DSM-IV diagnostic categories but who suffer a significant degree of distress and interpersonal impairment.
This proposal may seem utopian, and it would be naive not to acknowledge the cost factor as a significant deterrent to such a major revision of health care services. However, it can be argued that the kinds of services included would be more effectively chosen by rational considerations of their merits and disadvantages based on the demand for them, their effectiveness, and their cost, rather than by a need to make obeisance to the fossil that medical necessity has become. Such a cleansing of the slate would enable psychotherapists to directly make the claim that their services should be covered because of the help they offer to many troubled people in various kinds of distress. They would no longer have to disguise such a claim in an inappropriate "medical" camouflage.
I am aware of the magnitude of the task I am putting forward, but it needs to be balanced against the magnitude of the inconsistencies and confusion that currently result from the effort to stretch the concept of medical necessity beyond the vanishing point of meaning.
Dr. Chodoff is clinical professor of psychiatry at George Washington University in Washington, D.C. Send correspondence to him at 1904 R Street, N.W., Washington, D.C. 20009.