Martin T. Orne, M.D., Ph.D.
Despite the dramatic progress in psychopharmacology and the growing use of behavioral techniques, not only in psychiatry but also in general medicine, psychotherapy remains the most widely used approach in the treatment of a broad range of psychological and medical problems. Considerable controversy persists, however, about the effectiveness of psychotherapy, about the kind of training that should be required for those who practice psychotherapy, about how psychotherapy can be most appropriately applied in clinical practice, and about the cost of psychotherapy relative to other interventions. These concerns are particularly germane to the question of when psychotherapy should be reimbursed and by whom.
The Problem
Considering such questions demands addressing the issues that arise when one seeks to determine whether psychotherapy is effective. This includes the present state of knowledge concerning psychotherapy, determining what aspects of psychotherapeutic effectiveness can be meaningfully assessed, considering the changes that third-party reimbursal will bring about in the psychotherapeutic interaction, and focusing on the kind of research approach that in the long run is likely to advance understanding of the psychotherapeutic process.
Beginning with an illustration may be helpful in considering the current state-of-the-art of psychotherapy. Residents in psychiatry were asked to assume they needed their gallbladder removed. They then were to choose from among the great surgeons of the past; for instance, Hippocrates, the great Greek physician; Maimonides, the great physician of the 12th century; Billroth, the great German surgeon of the 19th century; and a mediocre surgical resident from an undistinguished medical school in a second-rate hospital today. Residents immediately pick the mediocre resident in the second-rate hospital today, since obviously his or her results would be vastly better than those of any of the giants of the past.
Then these same residents were asked to assume that they were seeking psychotherapy and were given a similar choice of psychotherapists from among Hippocrates or Maimonides (both of whom practiced psychotherapy), Freud, Jung, or Adler versus a mediocre psychiatric resident from an average medical school today, the answer would be quite different. Few would pick the mediocre resident. Indeed, everyone would want one of the greats. This striking difference tells us something about the nature of psychotherapy and the relationship of art to science in the practice of medicine.
The difference between choice of a surgeon and choice of a psychotherapist reflects the fact that there have been far greater scientific advances in the field of surgery than in the field of psychotherapy. The mediocre surgeon, thanks to advancements in technique, antibiotics, anesthesia, diagnostic procedures, and so on, can accomplish his or her task more effectively than the best surgeons of the past. However, pyschotherapy does not have the kind of incremental knowledge that can readily be taught and thus allow the mediocre psy-
55
56
chiatric resident to stand on the shoulders of the giants of the past. (This is in contrast to the pharmacological treatment of depression, where the mediocre psychiatric resident does better than his outstanding predecessors, due to advances in the science of psychopharmacology.)
This is the dilemma and also the challenge of psychotherapy. It does not mean that psychotherapy is ineffective. When compared with no treatment, it clearly has salutory effects. It does mean, however, that understanding of the psychotherapeutic process has not reached a level at which it can be separated into its essential elements and effectively be taught.
Thus, research findings in a number of studies show that a particular type of psychotherapy is considerably less important than such intangibles as the characteristics of the therapist and aspects of the treatment context. Although psychotherapy is effective overall, differences in effectiveness between two or more kinds of psychotherapy have yet to be solidly documented, especially across a wide range of disorders. Indeed, the relative contribution of specific treatment effects versus nonspecific or placebo factors to the effectiveness of psychotherapy remains unclear. This is a troubling state of affairs, particularly since consumers of psychotherapy are generally satisfied, and an increasing segment of the public demands the benefits of such a treatment.
Unfortunately, the cost of all medical treatment has been escalating at a rapid rate. Although the largest increases are associated with hospitalization of any kind, there is a fear, supported by limited data, that increasing third-party reimbursal for psychotherapy will be a bottomless pit. This fear is often translated into concerns about the effectiveness of psychotherapy. By requiring proof of efficacy for reimbursement, one thereby seeks to limit the application of psychotherapeutic techniques.
It is reasonable to ask, however, whether linking third-party reimbursal to efficacy considerations is either the most important or even the primary issue that requires attention. There are other questions concerning reimbursal for psychotherapy that, if resolved, could contribute greatly to a rational, scientifically based, cost-effective approach to reimbursal questions. These issues center around the theoretical basis for a psychotherapeutic intervention, the type of disorder most appropriate for treatment, and the training and competence of the therapist.
Rational Basis for Psychotherapy as a Treatment
Much has already been written and said concerning the expanding data base on the effectiveness of psychotherapy. Dr. Parloff (chapter 9) summarizes some of the research findings indicating that overall psychotherapy is indeed an effective treatment. There is an historical context within which to look at these findings, a context that reveals other equally important issues at the core of the reimbursal question.
It should be remembered that the issue of effectiveness has never been the sole basis for justifying accepted medical practice or, for that matter, third-party reimbursa -- current trends in technology assessment notwithstanding. For example, although it is true that modern medicine is incredibly effective in treating the bulk of infectious diseases and has made great progress in surgery, biochemistry, and other interventions, no one would seriously claim that medicine can treat all or even the vast majority of mankind's ills. There are a good many medical treatments that have not been shown to be particularly effective and at best result in palliation, but whose costs are reimbursed without question.
Traditionally, the patient sought medical advice and expected treatment that, if not a cure, at least helped ameliorate the effects of the illness. Although patients and physicians alike were generally persuaded that a combination of sedatives and physics, tonics, laxatives, good nursing care, hygiene, diet, bed rest, sleep and the like constituted effective treatment, proofs of specific effects were rarely available and rarely required. With the advent of specific treatments over the past 100 years, very powerful interventions became possible for many conditions. Despite the gains made in the 20th century, however, no one would seriously argue that, because of the absence of a specific remedy, physicians should stop trying to treat certain conditions.
This is a key point; reimbursement has been and should continue to be based upon the use of generally accepted treatments as much as it is upon the proven effectiveness of a treatment. The general acceptance or, as we shall refer to it here, the legitimacy of treatment, has always been a major part of medicine -- a troubled patient whom the physician treats in the best way possible. This does not apply to psychiatry alone, but to all of medicine, which people tend to forget in the heat of debate over the effectiveness of psychotherapy. Psy-
57
chotherapy should be judged by the same standard for reimbursal that is typically applied to traditional medical treatments for such maladies as back pain, acne, the common cold, many allergic conditions, headaches, and most viral infections.
Clearly, effectiveness is neither the only nor the essential prerequisite for reimbursement of many treatments within current medical practice. Simple success, while of obvious interest to both physician and patient, has never been the sole criterion for treatment choice. Other factors enter into treatment considerations. For example, regardless of effectiveness, the faith healer was never accepted as an appropriate provider of medical treatment. Similarly, the chiropractor failed to persuade physicians, despite many testimonials through patients. This was neither blindness nor the parochialism of the establishment. Rather, because it recognized the validity of making a medical diagnosis in the first place, the limitations of medical knowledge, and the importance of nonspecific factors in recovery from illness, the medical community demanded that the claims for specific medical treatment be based on rational principles that make sense in terms of scientific understanding of normal- and pathophysiology.
There was a time as recently as the early part of the 19th century when such schools of medicine as naturopathy and homeopathy effectively competed with orthodox medicine. Homeopathy prescribed infinitesimal amounts of medicine that could not have a real therapeutic effect. But because they were combined with careful attention to hygiene, good nursing care and a strong placebo effect, and because many of the treatments of orthodox medicine of the time were in fact causing more harm than good, the end result was that the homeopaths had better outcomes than orthodox medicine. This did not mean, however, the homeopathic theory was correct. In the long term, the search for specific treatments in medicine, not homeopathy, led to the success of modern medicine. Reviewing some of the discussions of the time shows that the physicians were indeed, for the most part, right about what they said. What history seems to be telling us is that it is more important to have a commitment to the search for a solution than to be satisfied with assessing an individual's success at a given moment in time. Individual outcome is interesting, and it is relevant to future treatment approaches, but it is by no means the only (and at times not even appropriate) criterion for acceptance.
It has always been crucial for the acceptance of a therapeutic procedure that it make sense scientifically. This demand for a rational basis for therapy has stood medicine in good stead, ultimately leading to the development of highly effective specific treatments. Though psychiatry and psychology may lag behind some areas of medicine in the number of specific and highly effective treatments available, as disciplines they are no less committed to a scientific understanding of human behavior and illness.
As with 19th century medicine, the general acceptance of psychotherapy as a scientific treatment rather than its effectiveness per se should be the criterion for legitimacy. One crucial factor is to distinguish between the mainstream, which maintains a commitment to the development of rational communicable knowledge, and the point of view that makes claims of effectiveness based on irrational or untestable conceptual models.
The Proliferation of Providers of Psychotherapy
Reimbursement for psychotherapy might be a relatively simple matter if physicians were the only providers of psychotherapeutic services. The reason for this is that as long as there is no excess of physicians and as long as nonpsychiatric medical practitioners are paid considerably better than psychotherapists, third-party payers would not have to worry about a bottomless pit of the increasing cost of psychotherapy. There would be a clear maximum, as charges are on the basis of time, and, even given a 45-minute hour, a dozen patients a day stretches the capabilities of most physicians. Therefore, a finite top would be placed on the potential costs of psychotherapeutic treatment simply by counting the number of medical people willing to devote their time to such a poorly paid enterprise.
The situation, of course, has been radically changed from this single-provider model over the past 30 years as other professions have laid claim to practicing psychotherapy independently. It has been argued that medical training is not essential to the practice of psychotherapy and that many nonmedical therapists appear to be as effective as those with medical training.
This situation creates some of the difficulties that exist today. Again, the parallel to 19th century medicine seems quite striking. There are different groups of individuals, with very different backgrounds and different lev-
58
els of training, competing for patients. Although rigorous credential requirements or licensing would be tempting as ways to solve the dilemma, such solutions are very unlikely in a pluralistic society. This is particularly so since no objective basis exists for establishing what constitutes the necessary training in order to be an effective therapist. In fact, the movement has been to decreasing qualifications, rather than increasing them! Unfortunately, the less well trained individual may not be a particularly good therapist, but is apt to learn to charge as though he or she were. The difference in costs between variously qualified providers of services is minimized by openmarket competition.
Medical Versus Nonmedical Uses of Psychotherapy
The proliferation of therapists, without hard evidence that one is better than another, presents the third-party payer with an ever-increasing number of would-be providers, without any defensible basis for deciding between their claims. What can third-party payers do under these circumstances? The first major distinction that needs to be drawn is between psychotherapy as a form of medical treatment (regardless of whether the service provider is a physician, a psychologist, or a social worker) and psychotherapy as a nonmedical treatment.
This distinction is by no means new. Cosmetic surgery may be a useful model. After all, cosmetic surgery is a medical procedure always carried out by surgeons. For their own sake, they usually are well trained and carry high malpractice insurance. However, the distinction between cosmetic surgery, at the convenience of the patient, and reconstructive plastic surgery after an accident is very clear. Third-party payers do not pay for a "nose job." They do not pay for a facelift. They do pay for plastic surgery for medical purposes, in burn cases, etc. A clear precedent exists for separating medical treatment with a given set of procedures from nonmedical treatment. Unfortunately, in psychotherapy this distinction has not been made, but it is crucial if the reimbursal issue is to be solved.
Despite the occasional case that requires adjudication in plastic surgery, the distinction is fairly clear cut and does not seem to present a serious problem for the health care delivery system. Within the psychotherapeutic tradition, however, the distinction between medical need and patient's desire to resolve inner conflicts -- in order to maintain increased equanimity, become more creative, increase the ability to work, play, and care for others -- has been obscured rather than emphasized. The psychotherapist sought as his or her task to accomplish as many of these goals as possible with the patient, rather than to limit the intervention to the minimum that must be accomplished in order to achieve the patient's stated goals and needs.
Psychoanalytic therapy is the particular form of psychotherapy that provided the model for the bulk of current treatments. Its goal was to help each patient achieve the highest level of functioning at work, play, and love that he or she could possibly obtain. Although the patient may well have sought therapy originally because of stress, or an incapacitating phobia, or hysterical paralysis -- certainly medical problems -- the scope of the treatment was defined to include an indepth exploration that has appropriately been likened to higher education of the emotions, a growth experience, an aid to full development.
All of these imply goals that greatly transcend medical treatment. They are no more candidates for reimbursement than cosmetic surgery or running exercise, both of which make one a "better" person but neither of which is essential to function normally. Although it has been argued that it was not possible to resolve medical difficulties such as incapacitating phobia, functional pain, or hysterical paralysis without working through the major psychodynamic difficulties of the individual, recent work with short-term psychodynamic therapy, behavior therapy, and cognitive approaches has shown that it is possible to treat many specific difficulties effectively without becoming involved in a massive long-term effort.
Although many psychiatrists would argue that optimal treatment demands the larger effort, third-party payers would seem justified in insisting that only treatments of conditions that either cause acute pain or dramatically interfere with functions or involve lifethreatening behaviors, and only those aspects of psychotherapeutic treatment specifically directed at the relief of such conditions, qualify for reimbursal.
The differentiation here is between treatment for an illness, a severe problem, and treatment as a form of growth or a kind of emotional or behavioral education. The latter should be paid for by the individual, not out of the public purse or out of the third-party payer's pocket. Clearly, some conditions, such as serious character disorders, the schizophrenias -- especially in younger people and adoles-
59
cents -- may require lengthy periods of treatment. Sometimes the severity of the problem and the potential for curing a patient with a long life expectancy would justify reimbursal of treatment over an extended period of time.
Adjusting reimbursement to treatment of specific incapacitating difficulties is important, but one should not expect from the outset to expect to set fixed limits on a treatment procedure, e.g., 10 sessions for one phobia, 15 sessions for two. There is no simple sessions-to-disorder ratio analogous to the dose-to-weight ratio for some drug treatments. A consensus should be developed with providers of services about what can legitimately be reimbursed and what aspects of psychotherapy, as desirable as they may be, do not constitute medical treatment for incapacitating disorders.
Such an effort will require good faith on everyone's part. It may be naive to assume that it will ever be possible to operationalize the difference between medical and nonmedical aspects of psychotherapy to everyone's satisfaction. On the other hand, it is essential for all providers of psychotherapeutic services to understand that if no viable peer review process can be developed, third-party payers are likely to progressively decrease the amount of psychotherapeutic treatment that is covered. Since psychotherapeutic treatment is relatively inexpensive, compared with hospitalization, and often prevents it, and since psychotherapeutic treatment typically decreases the amount of medical services that would otherwise have to be covered, third-party payers also have much to gain by this dialog.
The importance of joint approaches with sensible peer review procedures is underlined by a large group of borderline patients who can be maintained within the community, usually working at some marginal job, by being seen at intervals that may range from biweekly to bimonthly and often aided, of course, by medication. Some psychotherapeutic treatment is analogous to following a diabetic or an epileptic; it must be continued for the life of the patient and is obviously a medical necessity. The task of the therapist here is to see the patient as infrequently as the illness will tolerate and yet maintain the person in a functioning way with a modicum of comfort. This is an extremely cost-effective form of psychotherapy that can readily be practiced and really pays off for the third-party payer.
Sorting out "educational" therapy as a separate entity will allow third-party payers to comfortably identify patients who may need psychotherapy for a lifetime, albeit infrequently. The cost is not very great and will be far less than the alternative of what is bound to happen if the person in need does not have psychotherapy and ends up as a public responsibility.
Increasing the Efficacy of the Psychotherapeutic Process
Perhaps the most important issue for third-party payers to address is that evaluating the more than 150 brands of psychotherapy would be prohibitive in cost, duplicative, and ultimately fruitless as a test for reimbursement suitability. Typically, treatments are more alike than different, and it is by no means clear when their effects result from specific rather than nonspecific processes. The findings, consequently, would at best show small and usually insignificant differences. Few professionals really care whether one treatment is 60 percent effective, another 63 percent effective, and a third 69 percent effective, and there is no guarantee that reimbursement based on such minor differences would result in cost containment. Rather, practitioners and researchers seek to understand through ongoing clinical and basic scientific research where the 60-percent effectiveness comes from. What process accounts for that rate, and how can the procedure be made 90 percent or 95 percent effective? Such discovery would require an entirely different kind of research, directed at the process of what is going on in psychotherapy, rather than at comparing all possible outcomes from different psychotherapeutic treatment packages.
Perhaps the most sensible resolution, in the meanwhile, would be to apply the same criteria that are traditionally used in medicine -- that the treatment makes sense conceptually and is testable. Is there a serious attempt by practitioners to learn about what they are doing, to increase knowledge so they can teach it? Is there a commitment to testing their procedures, while at the same time recognizing the limitations of the technique they have proposed? Such a commitment would provide an ongoing adjustment mechanism for the rational application of the effectiveness-reimbursal equation.
Conversely, treatment providers who refuse to specify what they are doing or why they are doing it, and who do not examine the process by which their treatment works, are in the long run not likely to be productive. The most extreme example, faith healing, again may
60
well be effective, but it should not be reimbursed because, by its very nature, it defies scientific inquiry. In contrast, hypnotherapy, which may involve similar processes, seeks to bring these under close scientific scrutiny so they can be understood, improved upon, and ultimately more effective.
The issue then centers both on the nature of the process and on what is being done to understand the process. This is not a simple matter, nor should there be any illusions that the suggested approaches will be easily put into operation. Perhaps raising these issues will make possible a dialog, however, at least between providers of service and third-party payers who can recognize their ultimate community of interest.
The very inefficiency of psychotherapeutic technique marks this field as one where the greatest changes can and must be accomplished. Although it is easy to argue that the final answers must come from research, the funding for such research will not come from the pharmaceutical industry. Most private sector industry invests well over 10 percent in research and development. In the public sector, biomedical research is well below one-half of 1 percent of the cost of providing medical services. If the service needs of this Nation are ever to be met efficiently and economically, the best investment is likely to be a moderate increase in the percentage devoted to research in the process of treatment.
Summary
The highly touted effectiveness-reimbursal equation being applied to psychotherapy ignores other issues that, if resolved, could provide a framework for a rational cost containment of psychotherapy within the third-party reimbursal system. As with many other medical treatments, reimbursal should be based upon the legitimacy of psychotherapeutic treatment where "legitimacy" refers to generally accepted principles of treatment applied by properly trained and competent professionals to appropriate and debilitating disorders. The rational basis for the treatment should be amenable to scientific scrutiny; and research should focus on discovering those specific and nonspecific aspects of the various psychotherapeutic approaches that provide their effectiveness, with the ultimate goal being to greatly increase the efficacy and cost-effectiveness of applying specific psychotherapeutic interventions.