Investigation of inpatient utilization reveals a steady decline in utilization in the three psychotherapy groups from the "year before" to the "second year after," with the three remaining "years after" maintaining the level of utilization attained in the "second year after." In contrast, the control sample demonstrated a constant level in number of hospital days throughout the six years studied. These results are shown in Table 7, which indicates that the approximately 60 per cent decline in number of days of hospitalization between the "year before" and the "second year after" for the first two psychotherapy groups is maintained to the "fifth year after"; this decline is significant at the .01 level. The inpatient utilization for the "long-term therapy" group in the "year before" was over twice that of the nonpsychiatric sample, and about three times that of the first two psychotherapy groups. The significant (.01 level) decline of 88 per cent from the "year before" to the "second year after" is maintained through the "fifth year after," rendering the inpatient utilization of the third psychotherapy group comparable to that of the first two psychotherapy groups.

TABLE 7.-NUMBER OF DAYS OF HOSPITALIZATION AND AVERAGES BY PSYCHOTHERAPY GROUP FOR THE YEAR BEFORE AND THE 5 YR AFTER PSYCHOTHERAPY, AND THE CORRESPONDING PERIOD FOR THE NONPSYCHOTHERAPY GROUP

In terms of decline in use of inpatient services (days of hospitalization), however, the long-term psychotherapy group and the control group are different, in that the former patients significantly reduce their inpatient utilization from the "year before" to the "fifth year after." However, the small size of the samples limits the conclusions that can be drawn.

DISCUSSION

The original pilot study of which this project is an outgrowth was proposed by the senior author as an aid in planning for psychiatric care as part of comprehensive prepaid health-plan coverage. It had long been observed that some of this psychiatric clinic's patients, as well as many patients in the hospital for whom a psychiatric consultation was requested, had very thick medical charts. It was also repeatedly noted that when these patients were treated from a psychiatric point of reference, i.e., as a person who might have primarily emotional distress which was expressed in physical symptoms, they often abandoned their physical complaints. It seemed reasonable to expect that for many of these people, psychiatrically-oriented help was a more specific and relevant kind of treatment than the usual medical treatments.

This would be especially true if the effects of psychiatric help were relatively long-lasting, or if a change in the patient affected others in his immediate environment. In the long run, the interruption of the transmission of sick ways of living to succeeding generations would be the most fundamental and efficient kind of preventive medicine. It therefore seemed imperative to test the intuitive impressions that this kind of patient could be treated more effectively by an unstructured psychiatric interview technique than by the more traditional medical routine with its directed history.

The Balints [2, 3] have published many valuable case reports which describe the change in quantity and quality in patients' appeals to the general practioner after the latter learns to listen and understand his patients as people in distress because of current and past life experiences. It would be difficult, however, to design a statistical study of those patients and of a matched control group treated for similar complaints in a more conventional manner.

Psychiatry has been in an ambivalent position in relation to the rest of medicine welcomed by some, resented by others, often, however, with considerable politeness which serves to cover up deep-seated fears of and prejudices against "something different." In a medical group associated with a prepaid health plan, conditions are favorable for integrating psychiatry into the medical fraternity as a welcomed and amiliar (thereore unthreatening) member specialty. The inherent ease of referral and communication within such a setting would be much further enhanced by the factor of prepayment, which eliminates the financial barrier for all those who can afford health insurance. For many reasons, then, this setting provides both the impetus and the opportunity to attempt an integration of psychiatry into general medical practice and to observe the outcome. In the past two decades, medicine has been changing in many significant ways, among which are prepaid health insurance, group practice, increasing specialization, automation, and a focus on the "whole person" rather than on the "pathology."

Forsham [7] and others have suggested that at some not-too-distant date the patient will go through a highly automated process of history, laboratory procedures and physical tests, with the doctor at the end of the line doing a physical examination but occupying mainly the position of a medical psychologist. He will have all the results of the previously completed examinations which he will interpret for the patient, and he will have time for listening to the patient, if he wishes to do so. The "Multiphasic Health Check," [4] which has been used for many years in the Northern California Region in the Kaiser Foundation Medical Clinics and which is constantly being expanded, is just such an automated health survey, and Medical Group doctors are in the process of becoming continually better psychologists. Eventually many more of the patients who are now seen in the psychiatric clinic will be expertly treated in the general medical clinics by more "compleat physicians."

A study such as this raises more questions than it provides answers. One question alluded to above is whether, with an ongoing training program such as Balint has conducted for general practitioners at Tavistock Clinic, internists might not be just as effective as psychiatric personnel in helping a greater percentage of their patients. A training seminar such as this has been conducted by Dr. Edna Fitch in the department of Pediatrics of Permanente Medical Group in San Francisco for many years and has been effective in helping pediatricians to treat, with more insight and comfort, emotional problems of children and their families and physical disorders which are an expression of emotional distress.

Using a broader perspective than the focus on the clinical pathology, one can wonder what social, economic or cultural factors are related to choice of symptoms, attitudes toward being "sick" (mentally or physically), attitudes toward and expectations of the doctor, traditions of family illness superstitions relating to bodily damage, child raising practices, etc. How often is the understanding of such factors of crucial importance for effective and efficient treatment for the patient? Of special interest in general medical practice and overlooked almost routinely by physicians (and by many in the psychological field) are the "anniversary reactions" in which symptoms appear at an age at which a relative had similar symptoms and/or died.

Health Plan statistics indicate an increase in medical utilization with increasing age in adults. This is consistent with the relatively flat curve seen in the "medical utilization" of the control sample over the six year period and is in marked contrast to that of the experimental sample. There is the implication in this that some of the increasing symptoms and disability of advancing years are psychogenic and that psychotherapeutic intervention may in some cases function as preventive medical care for the problems associated with aging as well as preventive medicine in children.

A certain percentage of the long-term psychotherapy group seems to continue without diminution of number of visits to the psychiatric clinic; these patients appear from the data to be interminable or life-long psychiatric utilizers just as they had been consistently high utilizers of non-psychiatric medical care before. They seem merely to substitute psychiatric visits for some of their medical clinic visits. A further breakdown of the long-term group into three parts, e.g., less

than 50, 50 to 150, and more than 150 visits, would probably help to sort this population's utilization into several patterns. More precise data on these groups would suggest modifications in classifications and methods of therapy or might suggest alternatives to either traditional medical or traditional psychiatric treatment in favor of some attempt to promote beneficial social changes in the environments of these chronically disturbed people.

Sources of criticism

(1) one problem in providing a control group comparable to an experimental group in this kind of study is that, although undoubtedly having emotional distress, and in a similar "quantity" according to our yardstick, the control group did not get to the psychiatric clinic by either self- or physician referral. The fact that the control patients had not sought psychiatric help may reflect a more profound difference between this group and the experimental group than is superficially apparent. One cannot assume that the medical utilization of this control group would change if they were seen in the Psychiatric Clinic. (This objection will be minimized in the "prospective" part of this study, which will be reported in another paper.) Although the average inpatient utilization for the three combined psychotherapy groups is the same as that of the control group in the year before (1959), the inpatient utilization of the long-term psychotherapy group is two and a half times that of the control group. If the study were extended to several years before, rather than just one year, it would become evident whether this was just a year of crisis for the long-term group or whether this had been a longer pattern of high inpatient utilization.

(2) Patients who visit the psychiatric clinic may, for one reason or another, seek medical help from a physician not associated with the Medical Group so that his medical utilization is not recorded in the clinic record, the source of information about utilization. In the long-term-therapy group the therapist is usually aware if his patient is visiting an outside physician, and although it is an almost negligible factor in that group, there can be no information in this regard for the one-session-only and brief-therapy groups without follow-up investigation.

(3) There is no justification in assuming that decreased utilization means better medical care, necessarily. Criteria of improvement would have to be developed and applied to a significantly large sample to try to answer this important question.

(4) Patients may substitute for physical or emotional symptoms behavioral disturbances which do not bring them to a doctor but may be just as distressing to them or to other people.

(5) The "unit" of utilization cannot be used as a guide in estimating costs, standing as it does for such diverse items. In itself the units are not an exact indicator of severity of illness nor of costs. A person with a minor problem may visit the clinic many times, while a much more severely ill person may visit the clinic infrequently. Even more striking is the variation in the cost of a unit, varying from about a dollar for certain laboratory procedures to well over a hundred dollars for certain hospital days (with admissions procedures, laboratory tests, x-rays, consultations, etc.) each worth one "unit." To arrive at an approximation of costs, the units have to be retabulated in cost-weighted form. Suggested further studies

(1) The question of treatment of patients by non-medical professional clinicians has been argued for more than a half century. It is generally recognized that there are not enough psychiatrists now and that there will not be enough in the foreseeable future to treat all those persons who have disabling emotional disorders. In the late President Kennedy's program for Mental Health this lack was recognized; the recommendation for professional staff for community Mental Health Centers included clinical psychologists, psychiatric social workers and other trained personnel. Having little distinction in our psychiatric clinic between the various disciplines as far as their functions are concerned, it would be feasible and interesting to compare therapeutic results of the disciplines as well as individuals with various types of patients and various types of psychotherapy.

(2) Is length of treatment correlated with diagnostic category, original prognosis by therapist, socio-economic level of patient, discipline and orientation of therapist, or "severity of pathology"?

(3) What happens to the spouse, parents, and children of the patients who are seen in psychiatry?

(4) Are there distinguishing patterns of complaints in the three psychotherapy groups?

(5) How do blue-collar patients differ from white-collar or professional patients in number of interviews, diagnostic label, use of medication, recommendation of hospitalization, and type of complaints?

(6) What is the nature of the illness that resulted in hospitalization before the patient came to psychiatry-and after? How often was this a diagnostic work-up because the internist could not find "anything wrong" in the clinic?

SUMMARY

The outpatient and inpatient medical utilization for the year prior to the initial interview in the Department of Psychiatry as well as for the five years following were studied for three groups of psychotherapy patients (one interview only, brief therapy with a mean of 6.2 interviews, and long-term therapy with a mean of 33.9 interviews) and a control group of matched patients demonstrating similar criteria of distress but not, in the six years under study, seen in psychotherapy. The three psychotherapy groups as well as the control (non-psychotherapy) group were high utilizers of medical facilities, with an average utilization significantly higher than that of the Health Plan average. Results of the study indicated significant declines in medical utilization in the psychotherapy groups when compared to the control groups whose inpatient and outpatient utilization remained relatively constant. The most significant ones occurred in the second year group of the initial interview, and the one-interview-only and brief-therapy groups did not require additional psychotherapy to maintain the lower utilization level for five years. On the other hand, after two years the long-term psychotherapy group attained a level of psychiatric utilization which remained constant through the remaining three years of study.

The combined psychiatric and medical utilization of the long-term therapy group indicated that for this small group there was no over-all decline in outpatient utilization inasmuch as psychotherapy visits seemed to supplant medical visits. On the other hand, there was a significant decline in inpatient utilization, especially in the long-term therapy group from an initial utilization of several times that of the Health Plan average, to a level comparable to that of the general adult Health Plan population. This decline in hospitalization rate tended to occur within the first year after the initial interview and remained generally comparable to the Health Plan average for the five years.

REFERENCES

1. Avnet, Helen H.: Psychiatric Insurance: Financing Short Term Ambulatory Treatment. New York, Group Health Insurance, Inc., 1962.

2. Balint, Michael: The Doctor, His Patient and the Illness. New York, International Universities Press, 1957.

3. Balint, Michael, and Balint, Enid: Psychotherapeutic Techniques in Medicine. London, Tavistock Publications Limited, 1961.

4. Collen, M. F., Rubin, L., Neyman, J., Dantzig, G. B., Baer, R. M., and Siegelaub, A. B.: Automated multiphasic screening and diagnosis. Am. J. Pub. Health 54, 1964.

5. Cummings, N. A., Kahn, B. I., and Sparkman, B.: Psychotherapy and Medical Utilization. As cited in Greenfield, Margaret: Providing for Mental Illness. Berkeley, Calif., Berkeley Institute of Governmental Studies, University of California, 1964.

6. Follette, W. T., and Cummings, N. A.: Psychiatry and Medical Utilization. An unpublished pilot project, 1962.

7. Forsham, Peter H.: Lecture before the Permanente Medical Group, San Francisco, 1959.

[From Medical Care, September-October 1970, Vol. VIII, No. 5]

EFFECT OF A SHORT-TERM OUTPATIENT PSYCHIATRIC THERAPY BENEFIT OF THE UTILIZATION OF MEDICAL SERVICES IN A PREPAID GROUP PRACTICE MEDICAL PROGRAM

0

(By Irving D. Goldberg, M.P.H., Goldie Krantz, M.A.,†

and Ben Z. Locke, M.S.††)

A pilot study was conducted to measure the effect of a short-term outpatient psychiatric therapy benefit on the utilization of general medical services at Group Health Association of Washington, D.C. (GHA), a prepaid group practice medical program. The study group consisted of 256 patients who were referred for such outpatient therapy and who were GHA members for a full 12-month period both before and after the psychiatric referral. Study patients experienced a marked reduction during the year after referral as compared with the prior year in the utilization of GHA nonpsychiatric physician services and laboratory or X-ray procedures. The reduction in number of patients seen was 13.6 percent for nonpsychiatric physician services, and 15.7 percent for laboratory or X-ray procedures. In terms of visits made, reduction was approximately 30 percent for each of these services. Basic findings of reduced utilization was still obtained when factors of age, race, sex, psychiatric diagnosis, and number of therapy sessions attended under benefit were taken into account. Results support findings of reduced utilization in other studies and suggest more efficient utilization of appropriate medical services as a result of short-term outpatient mental health benefit in prepaid health plan settings.

Only in the past decade have significant increases in mental health benefits been included in the rapid growth in health insurance protection through private voluntary insuring organizations. Since 1963, the National Institute of Mental Health (NIMH) has actively stimulated this development of encouraging the expansion of private voluntary health insurance coverage for mental health. [5] In a collaborative effort with the NIMH, the United States Civil Service Commission, which administers the Federal Employees Health Benefits program, requested insurance carriers and health plans participating in that program to incorporate new or improved mental health benefits, particularly coverage for outpatient services, into their existing benefit structures.

A total of some four million people are enrolled in community prepaid group practice health plans which are essentially comprehensive in their health coverage. [6] Prior to 1960, when the federal employees program went into effect, these plans in the main were without prepaid mental health benefits. However, all federal employees enrolled in these plans now have some mental health coverage, including outpatient benefits; and similar coverage is also available to other members and contractor groups in these plans.

With the adoption of mental health benefits in prepaid group practice plans, it has become possible to evaluate to some extent the effects that these benefits might have on patient utilization of nonpsychiatric medical services covered by the plans. [1, 2] Group Health Association of Washington, D.C. (GHA) cooperated with the Biometry Branch of the NIMH in conducting a small pilot study directed towards this question. This paper reports on the results of that study which is based on the first year's experience with a mental health benefit at GHA before benefits were expanded and before the total population of GHA was included.

Chief, Evaluation Studies Section, Biometry Branch, National Institute of Mental Health, Chevy Chase, Md.

Program Analyst, Group Health Association, Inc., Washington, D.C.

tt Assistant Chief, Center for Epidemiologic Studies, National Institute of Mental Health, Chevy Chase, Md.

The Federal Employees Health Benefits program, which became effective in 1960 under an Act of Congress, is the largest employer-sponsored contributing health insurance program in the world covering more than seven million persons, including employees, annuitants, and dependents.