Mental Health Services in General Health Care: A Conference Report, Volume I (1979)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

145 EXECUTIVE SUMMARY: COORDINATED MENTAL HEALTH CARE IN NEIGHBORHOOD HEALTH CENTERS* Jonathan F. Borus, M.D., Barbara J. Burns, Ph.D., Alan M. Jacobson, M.D., Lee B. Macht, M.D., Richard G. Morrill, M.D., and Elaine M. Wilson, B.A. This report is drawn from the authors' cumulative clinical, adminis- trative, and research experience in both Community Mental Health Centers (CMHCs) and Neighborhood Health Centers (NHCs) over the last decade as well as a review of the relevant literature. From our experience we feel that the neighborhood-based conjoint health-mental health setting of the NHC is an excellent context for the provision of primary mental health services, i.e., problem and diagnostic evaluation, crisis inter- vention, individual, group and family psychotherapies, aftercare services (including psychoactive medication) for the chronically ill in the com- munity, and prevention/educational outreach programs about mental health and mental illness. We have been impressed by the opportunities that such a setting provides for mental health professionals to collaborate with primary physicians in the latter's roles as case-finders and treaters of patients with defined mental disorder, patients with com- bined psychiatric and medical problems, and patients reacting to either external or illness-related stresses. The report's first section traces the history of public mental health and health services in the U.S., discusses levels of care, and presents an overview of the current Suctioning of NHC mental health programs. The second section outlines five hypothesized advantages of providing primary mental health services as part of primary health care in the NHC's organized neighborhood setting and the third section reviews theoretical, clinical, and research data relevant to these hypotheses. A summary of these latter two sections, in which the hypothesized advantages and related findings are reviewed, follows. 1. Advantage: The provision of mental health services within a neighborhood-based, primary health setting can improve their accessibility and - acceptability to and utilization by neighbor- hood citizens. . *An earlier version of this paper was developed at the invitation of the Institute of Medicine and was submitted to the President Is Commission on Mental Health, 1977. It was used as a background paper for this conference. The full text comprises Volume II of the conference proceedings.

146 This hypothesis is corroborated by quantitative data from a six month study at Boston's Bunker Hill Health Center in which mental health utilization rates were five times higher than NIMH data on average national utilization. This higher utilization occurred across sexes, and for all age groups and marital statuses. The services disproportionately reached usually underserved populations including patients from the lower socio-economic classes, children and married adults. It is felt that this increased utilization reflects the easy access to these centers, whose neighborhood loca- taion decreases geographic and travel barriers to care, as well as an increased acceptability of the NHC as a mental health service setting to citizens. Acceptability is encouraged by the frequent use of indigenous care-givers who understand local culture and values and the fact that the mental health services are provided within a trusted health institution with which many people are familiar and comfortable from prior use of its health, dental or heaLth-related social services. Mental health services offered in a multi-service setting are also less stigmatized as the patient does not have to automatically label himself as mentally disordered simply by walking through the door. As below, patients frequently present with somatic complaints to the primary physician and then "slide over" into specialist mental health care. 2. Advantage: The conjoint health-mental health delivery system can improve case-finding, successful referral, coordination of care, long-term follow-up, and preventive/educational efforts to meet general and specific population needs. Primary physicians are major case-finders and treaters of patients with mental disorder. It is estimated that 60% of all patients with mental disorders seen in primary care is not referred to mental health specialists. Working side by side in the NHC, the primary physician can form a collaborative relationship with the psychiatrist and build up the necessary trust to seek consultation about patients with mental disorder whom the primary physician decides to treat alone. In the conjoint setting, the primary physician can also easily refer patients for mental health specialist care whom he cannot best handle. The latter is supported by a study of 19 Boston NHCs in which almost half of all referrals for mental health care came from the medical staff of the health center. The physical proximity and the ability to form professional relationships between primary physician and mental health specialist also facilitates collaboration, mutual training, and communication about treatment planning to foster coordinated rather than fragmented care for patients with combined or multiple health and mental health problems. The setting is also an excellent one for long-term follow- up care. Patients with chronic or episodic mental disorders often resist continuing in mental health care during a period of remission of symptoms but will continue their contacts with the NHC for their

147 costs, including the total costs for the multiple outpatient mental health services needed by chronically ill patients living in the community. 4. Advantage: The provision of mental health services within the NHC can increase the priority of and concern for mental health problems among com- munity citizens. - Health has a much higher priority and is a less stigmatized issue for citizens groups to come together around than mental health. The 10,000-50,000 person NHC service area is smaller than the CMHC's 75,000-200,000 catchment and the smaller area's citizens are more likely to have some sense of cohesion and shared destiny which allows them to relate to the NHC as "theirs." It has been our experience that mental health professionals can work closely with citizens groups in NHCs to plan and promote mental health programs, including aftercare programs for the chronically ill, which fit well and are accepted by the neighborhood. 5. Advantage: The conjoint health-mental health setting of the NHC can offer unique opportunities for necessary training in primary health and mental health care. The NHC offers primary practice and mental health trainees opport- unities for both front-line, acute ambulatory care and long-term care of chronic illness in the community Training in their patients' neighborhood can help both sets of providers consider the impact on clinical care of socio-environmental, ethnic and cultural, and public health aspects of care often ignored in institution-bound training. Importantly, the proximity to other trainees and availability of collabor- ating role models can help both primary practice and mental health pro- fessional trainees overcome prior negative inter-professional stereotypes to learn to use each otherts expertise to their patients' benefit. The fourth section of the report develops ten issues relevant to the NHC delivery system and suggests alternatives to current national mental health policy concerning these issues, as briefly outlined below. 1. NHCs are proposed as a~preferred setting for the delivery of primary mental health care in the public sphere. The conjoint health-mental health setting and location in the community allow it to address itself to the medical' psychological, socio-environ- mental, and public health aspects of patient care. 2. Closer linkages are proposed between NHCs and CMHCs. Although the NHC setting is excellent for providing primary mental health services, CMHCs are vital sources of secondary and tertiary level care for many intensive and expensive services such as in- patient care, rehabilitation, etc. Both CMHC and Bureau of Com- munity Health Services legislation could provide incentives and requirements for closer linkages between NHCs and CMHCs.

148 3. New sources of support are proposed to meet critical needs for consultation, collaboration, inservice training, and other indirect services which foster coordinated rather than fragmented care. Other indirect services, such as education, prevention, and screening out- reach programs to increase access to care, training of health and mental health professionals, and evaluation research efforts to under- stand these delivery systems also need to be supported. Although essential to a coordinated care system, few of these critical indirect services are paid for by current reimbursement schemes. 4. Methods are proposed to interlink NHC and CMHC citizen groups and to provide citizen and professional input into Health Services Agencies concerning mental health needs and issues. 5. Additional resources are proposed to stimulate the develop- ment of facilities in NHCs to provide long-tenm aftercare and day treatment for the chronically ill in the community. 6. Linkages are proposed between NHCs and Health Maintenance Organizations, multi-specialty private group practices, and solo private physicians to collaborate in providing coordinated care. 7. Support is proposed for greater in-service and professional training of both primary physicians and mental health professionals in NHCs and other coordinated primary health and mental health care settings. Part of residency training for both primary physicians and psychiatrists could occur in the NHC setting as well as subs specialty training for selected professionals who wish to develop the clinician-executive skills necessary to effectively lead such conjoint health delivery systems. 8. Evaluation research outcome studies are proposed of the NHC system of care to complement existent preliminary studies of service provision and utilization. 9.-10. Alternative funding mechanisms to pay for the direct and indirect services provided in NdCs prior to (9) and as part of (10) National Health Insurance plans are proposed. The report concludes with three recommendations. These are that national policy, as embodied in federal health services program legislation, government and private health financing programs, and National Health Insurance plans, should encourage and fiscally support: 1. further development of Neighborhood Health Centers and other relevant population-based settings which provide mental health services as a coordinated part of primary health care. those indirect services which facilitate coordination rather than fragmentation of primary health and mental health care and outreach into the community to increase citizen accessibility to and acceptance of necessary mental health services. 3. needed professional training in and critical evaluation research about Neighborhood Health Centers and other con- joint health-mental health primary care settings. 2.

149 TRAINING FAMILY PHYSICIANS IN MENTAL HEALTH SKILLS IMPLICATIONS OF RECENT RESEARCH FINDINGS David Goldberg, M.D. University of Manchester England Many of the well-known surveys of mental disorders among patients attending primary care physicians have shown very wide variations between individual practitioners in their estimates of the frequency of such conditions. Shepherd 1/, for example, showed a nine-fold variation between family physicians in London, and surveys in the United States have shown even wider variations. The results of five recent surveys are summarized in Table 1, which shows the percentage of consecutive attenders thought "psychiatric" by the primary care The mean is the figure which is most usually quoted when and it is often conveniently . ~ . . ~ physician. reference Is mace co these surveys, forgotten that this mean is arrived at by averaging the pronouncements of very heterogeneous observers. Table 1. INTER-PRACTICE VARIATION IN THE DETECTION OF PSYCHIATRIC MORBIDITY: THE FIGURES REFER TO PERCENTAGES OF CONSECUTIVE ATTENDERS Number of Physicians Mean Range Marks, Goldberg & Hillier 2/ Manchester, U.K. 91 14.2% 15-64% Locke et al 3/ Prince George's County, Md. 79 9.0% 0~44% Locke & Gardner 4/ Monroe County, N.Y. 58 16.9% 0-37% Leopold, Goldberg & Schein 5/ W. Philadelphia, Pa. 32 16.3% 0-92% Goldberg & Steele 6/ Charleston, S.C. 45 39.0% 0-85%

150 When psychiatric screening tests are used simultaneously during such surveys, they typically show much less variation between the various practice populations than that suggested by the doctors' own assessments. Figure l shows 14 practices in Michael Shepherd's well- known study , arranged in rank order, so that the prac Lice with the highest rate of ascertainment of psychiatric illness is on the extreme left while that with the lowest is on the extreme right. Figure l cot I 60 ~ 50 - - Z 40 `, 30 At us 20 a u, He 1 0 u' on—-—o SCORE OF 10 OR MORE ON THE M- R SECTION ~ : GENERAL PRACTITIONERS' CLI N I CAL ASSESSM ENT \ iR~ \ on Q \ a\ I o ( ! I I I I I ~ Ox I o,.__O PRACTICES ORDERED BY THEIR PATIENT CONSULTING RATES FOR PSYCHIA. RIC ILLNESS

151 The continuous line shows these assessments -- varying from 63% on the left to 18% on the right. The dotted line shows the probable cases predicted by the Cornell Medical Inventory. It can be seen that so far from there being any association between the two variables the best regression line between the points an the dotted line has a positive slope. Nor can it be concluded that the doctor where the two lines cross is a very clever fellow, and that his colleagues perched at either end of the line are various degrees of fool. The doctor where the lines cross could be identifying completely different individuals from-those picked out by the questionnaires: it may just happen that they pick out the same proportion of the population. What we can say is that the doctor on the extreme left has some sort of bias towards perceiving patients as psychologically sick, and that the reverse may be true of the doctor on the right. Figure 2 has a familiar look about it. But we are now using the General Health Questionnaire instead of the Cornell Medical Inventory; we are studying 29 second and third year family practice residents rather than established physicians; the time is 1979, and the place is Charleston, South Carolina. But nothing else has changed. Figure 2 . . . .. Percent , 80— 70- 60- 50- 40 - 30- 20- 10- Conspicuous Morbidity '7 ~ `q ,~ Probable Prevalence 1 111 8, h~ too '--- r I ~ I - I I Al 1 st 5th 1 Oth 1 5th 20th 25th 30th Ronk Order (CO'lspiClJOUS Morbidity)

152 Table 2 shows that there is in fact no correlation between the level of disorder reported by the doctors and the level of disorder in their population predicted by a screening questionnaire. The results now include a large survey that we carried out in Manchester with similar results. Table 2. RELATIONSHIP BETWEEN THE LEVEL OF THE DISORDER REPORTED BY THE DOCTOR AND THE LEVEL ASSESSED BY THE PSYCHIATRIC SCREENING TEST Shepherd et al. 1/ 14 General Practitioners, London, England -0.31 (NS) Marks, Goldberg & Hillier 2/ 22 General Practitioners, Manchester, England Goldberg & Steele 6/ 29 Family Practice Residents, Charleston, S.C. -0.17 (NS) +0.08 (NS) Let me be very clear about the conclusion to be drawn from these data. One cannot conclude that there is no association between an individual physician's assessments and the symptom levels of his patients. The reverse is in fact true, and we shall be returning to that. But it is reasonable to assume that a doctor who tells you that 90% of his patients are mentally sick is no more likely to have a greater number of sick patients attending his office than a doctor who tells you that only 10% are sick. The difference between them resides not in their patients, but in their concepts of psychiatric disorders and the threshold that they adopt for case identification. Figure 3 shows the ratings made by an imaginary physician "A" for 60 patients, each of whom has completed a screening questionnaire. The scores on the screening questionnaire are shown on the vertical axis, and the doctor's rating of the degree of psychiatric illness is shown on the horizontal axis. It can be seen that the correlation between the two measures is quite good, at +0.6. This doctor identi- fies 62% of his patients as sick. His colleague, physician "B", makes exactly the same ratings and so has the same correlation coefficient; but he has a more restrictive view of illness. We can see in figure 4 that his threshold has been moved to the right, so that only 17% are cases. Physician "B" can be said to have a negative bias towards identifying psychiatric illness; he will report a low rate and will tend to fail to identify cases among his patients with high scores (i.e. he will have a low "identification index": see reference 2~. He will have the compensatory advantage that he will rarely accuse asymptomatic patients of being psychologi- cally sick (i.e. he will have a high "stability index": reference 6~.

153 Figure 3 SCORE ON PSYCHIATRIC SCREENING QUESTIONNAIRE High Low . · · ~ ·- -. 1~. ~ ·~. ~ · I .- ·~. ·- · · 1.~. .. · .. I .~e · l · -. Be- ·e ·~. ~. · ~- _ O 0 1 j 2 . Normal Subclinical 1. Mild _~ 1 "Non-Cases" 3 4 Moderate Severe "Cases" CLINICAL ASSESSMENT BY DOCTOR (% Cases= Conspicuous Psychiatric Morbidity, CPM). Physician "A" Correlation Between Doctor and Screening Test +.60

154 Figure 4 SCORE ON PSYCHIATRIC SCREENING QUEST ONNAIRE High Low 1 ·. ·- ~ -. ·~. 1 ~ o o · - ~ ·. ·. 1 ·~. 1 .. - 2 3 1 4 LOW BIAS -These doctors miss many probable cases. (Identification Index- Low Stability Index - High) j CLINICAL ASSESSMENT BY DOCTOR

155 Inevitably, physician "C" has a high bias towards diagnosing illness, with a conspicuous morbidity of 83%. Such doctors rarely fail to identify symptomatic patients as " sick"; yet this advantage will be purchased at the price of frequently identifying asymptomatic patients as "sick" probably because as raters, they tend to guess " s ick" when in doubt . Figure 5 SCORE ON PSYCHIATRIC SCREENING QUESTIONNAIRE , ! High Low · ~ 1 .~. ·~. O O 1 .. r · ·~. · ~ .~. HIGH BIAS -These doctors falsely identify asymptomatic patients as "sick" (Identification Index - High Stability Index - Low) . CLINICAL ASSESSMENT BY DOCTOR

156 These mythical physicians have been shown to emphasize the point that a doctor's assessments of the level of mental illness tells us more about him than about his patients: what it tells us about him has been studied by Michael Shepherd and ourselves. Shepherd and his group 1/ were able to account for 51Z of the non-random variance of the doctor's assessments in terms of two variables: the "mobility index" was an ecological variable related to the turnover of patients in each practice, and the "psychosomatic score" was an attitude scale which measures the extent to which the doctor thinks psychogenic factors are important in the etiology of various illnesses. The more important he thinks they are, the higher will be his estimate of psychiatric morbidity. Our recent study in Manchester 2/ was able to include rich observational data made by a psychiatrist observing over two thousand interviews by 56 family doctors, and succeeded in accounting for 53% of the variance of the doctors' assessments. A dimension we have called "interest and concern" accounts for a quarter of the variance, and is characterized by the doctor being rated as an empathic interviewer by the observing psychiatrist; having a tendency to ask patients about their home and their family; and being interested in psychiatry. The next dimension, "psychiatric focus," replicates Shepherd's earlier finding. These doctors score highly on Shepherd's scale concerned with psychogenic factors in illnesses, are once more rated as "interested in psychiatry," and have a tendency to ask questions of a psychiatric nature during their diagnostic interviews. Finally, older doctors who have spent longer times in the practice being observed are likely to have rather higher rates than younger, newly arrived doctors. The data presented so far support the view that the level of disorder reported by individual doctors tells us very little about the psychological abnormalities among their patients. This however is not to say that their collective assessments are of no value. Shepherd showed a correlation of +.19 between CMI score and the doctors rating of "case" versus "non-case." More recent work, using the GHQ as the screen, and allowing the doctors a six point scale of degree of psychological disturbances rather than the two- way classification used by Shepherd, shows figures for Spearman's rank order correlation coefficient `>f between .3 and .4. 2/,6/ These correlation coefficients measure the degree of association between the patients' symptom levels on the one hand, and the physicians' ratings of the degree of disturbance on the other. They take no account of the threshold that an individual doctor uses for case identification, or indeed of the threshold score used for the screening test. We are measuring the ability of a doctor to make ratings that are congruent with his patients' symptom levels, and the striking thing about such a measure is how variable doctors are among themselves in this important respect.

157 Figure 6 shows that although the mean coefficient for 9l British GP's was +.36, some have coefficients as high as those obtained by trained psychiatrists using lengthy research interviews despite the hurried nature of their own interviews, while others have coefficients that indicate their total inability to make accurate ratings. It can be seen that the situation in the United States is similar to that in England. What determines whether a doctor will have a high coefficient or a low one? Figure 6 Number of Doctors 25- 20- 15 10 5~ O- ~ an.. . . ~ -.1 0 .1 ............... ........... 5 ::.:.: :a, . ~_ ~ . .2 .-.- :-:.:-:.:. :-:.:.:.: ·:.:-:.:-: :-:.:-:-:-. :-:-:-:-:- ·.:.:-:.:.:- ·:-:-:---:- .:-:-:.-.:. :-.-.:.:.:-. ...... . , ·O .4 :,:. .:.:. :-:-:-:-- ::: .5 .6 Correlation Coefficient Between Doctor and GHQ. ~ 91- Family Doctors, Manchester, U.K. 45 Family Practice Residents; Chor/eston, U.S.A.

158 It proved possible to account for almost two-thirds of the variance of correlation coefficient. 2/ Once more, the dimension "interest and concern" turned out to be important, since it accounted for 40% of the variance. This time, however, the dimension of "conservatism" was accounting for a further 19% of the variance. Conservatism is con- ceived as a broad personality dimension reflecting inflexibility, resistance to change, and authoritarianism rather than a political dimension. It tells you very little, for example, about voting habits. Less conservative doctors are not only better able to rate their patients' disorders accurately, but are also more likely to ask "probe" questions with a psychiatric content. It seems likely, therefore, that ability to make accurate psycho- logical assessments is partly related to attitudinal and personality factors, and partly related to the possession of certain skills as an interviewer. It is the latter that offer more promise to the planner of training programs for family practice residents. The random rating behaviors of certain family doctors on each side of the Atlantic necessarily indicates that much psychiatric morbidity is missed. The amount that is missed of course depends upon the doctor; an average Manchester GP detects rather more than half of the cases predicted by the screening questionnaire. 2/,8/ Once more, the range is very wide. Some detect all of them; others almost none. Does this matter? Preliminary evidence suggests that it does. In one practice in Yorkshire 9/, we randomly assigned patients with high scores on the screening questionnaire who had been rated by the doctor as psycho- logically normal to an index condition in which the doctor was in- vited to discuss the responses on the GHQ with the patient, and a control condition in which the information was withheld from him. Three months later, 90% of the index group were well, but only 40% of the control group. Patients in the control group gradually improved with time, but had to endure an average of 5.3 illness- months following initial consultation, in contrast to the 2.8 illness-months for the patients whose disorder had been discussed with their doctor. Before we consider the implications that such findings have for the future training of family physicians, let us turn briefly to con- sider the treatment that emotionally disturbed patients actually re- ceive from family physicians. In their research among depressed women found during a community survey in the London borough of Camberwell, Brown and Ginsberg report that the majority of such patients attend their doctors hoping that they will have an opportunity to discuss their problems with him. What they typically get is a short interview and a prescription for a psychotropic drug. In Ginsberg's words:

159 "Women go to the doctor with the hope that they might gain help but usually with low expectations that the kind of help they will receive -- that is, a prescription for a psychotropic drug -- will be effective." 10/ The evidence that we have certainly suggests that although lip- service is paid to psychotherapeutic help, pharmacology is what the patients actually get. Shepherd 1/ asked family doctors to respond to a set of clinical vignettes, each of which was followed by a range of possible actions. One of these was always a drug therapy, one was always referral to a specialist service, one was an appro- priate form of psychotherapy, and so on. The psychotherapeutic choice was usually the one chosen by his doctors, yet he observed that in practice his survey showed that psychotherapy was not often used. In Shepherd's words: "It is evident how much our practitioners relied on drugs even though they were clearly aware of the role of psychological and social factors in the presenting conditions. While a substantial number of patients presented somatic symptoms, for which a somatic form of treatment might have been regarded as understandable, this would not account for more than a small part of the whole picture." Our survey in Manchester confirmed Shepherd's findings. The research psychiatrist frequently saw family doctors giving psychotherapeutic help in the form of allowing patients to ventilate problems, empathic listening and giving them reassurance during the course of their clinical interviews. Yet of the 2,()92 interviews observed, not a single one was an interview in which the patient had expressly come for psychotherapeutic help and the session was formally devoted to that kind of help. In fairness, I should perhaps say that in the Family Practice Residency Training Program at Charleston, I have quite often seen such sessions, and that I freely acknowledge that there are many General Practitioners in England who do give such help. All that I am saying is that such help is relatively infrequent in the actual working conditions observed by us in a large sample of general practices in the north of England. Since the topic of this paper is training, it is also worth recalling that the older genera- tion of the family doctors studied will have had an undergraduate training in psychiatry which was wholly inadequate by present day standards, and that very few of them will have had any training at all aimed at helping them to acquire psychotherapeutic skills. Hawton and Blackstock _ / have recently interviewed family doctors who had been looking after 122 patients who had deliberately poisoned themselves with psychotropic drugs. 63% had seen their doctors in the month before the attempt, and indeed 45% had seen him six or more times in the previous year. The family doctors

160 had identified many interpersonal and social problems among these patients, but had responded to them with drugs rather than with counseling. At least 80% of the patients used drugs of the class prescribed for their overdose. The authors are not opposed to the use of psychotropic drugs by family physicians, but they observe that they should be prescribed only for clear indications: "Antidepressants in therapeutic doses where there is clear evidence of underlying depressive illness; tranquillizers for short periods to help patients through crises when levels of anxiety are such that their coping ability is grossly impaired; and non- barbiturate hypnotics for short periods where insomnia, secondary to stress, is undermining an individual's resources, such as during a severe grief reaction... Automatic represcribing, found to be so common for psychotropic drugs by Freed, should be avoided if at all possible." 11/ To what extent do primary care physicians prescribe psychotropic drugs which are appropriate to the symptoms experienced by their patients? Hard facts are thin on the ground. One problem is that the patients cannot be neatly divided into those with anxiety states and those with depression: the majority of patients seen in primary care settings have combinations of depression and anxiety, with or without somatic symptoms. Downing and Rickels 12/ succeeded in showing-that those prescribed anxiolytics were somewhat more anxious than those who were depressed, and that the reverse was true for those prescribed antidepressants. However, all the physicians included in their study were members of a Psychopharmacology Research Group, and indeed many were psychiatrists. It is not clear whether physicians without a special interest in the action of psychotropics would be so discriminating; available evidence is not encouraging. In his extensive study of the prescribing of psychotropic drugs in general practice, Parish 13/ showed that many of the physicians studied used psychotropics in an idiosyncratic and undiscriminating way. Thus one relied almost solely on phenobarbitone, another on butobarbitone, and a third prescribed either "Librium" or "Mandrax." Furthermore, it seemed likely that use of a favorite drug was linked with attachment of a favorite diagnostic label. "One practitioner's frequent diagnosis was 'nervousness' or 'debility' for which he constantly prescribed Beplete... in another instance 'nervous dyspepsia' was frequently diagnosed and treated with Melleril, and in a third practice tension headache was the most commonly diag- nosed disorder and (was usually treated with) Librium and Valium."

161 Should it be thought that British physicians are particularly remiss in their prescribing of psychotropics, Hyams, et al 14/ studied 227 primary care physicians in Oregon and New York: although drugs and physical examinations were preferred to insight-oriented psychotherapy as treatment methods, 32% of the physicians failed to name even one correct drug for the management of anxiety or depression. I wish to draw four harsh conclusions from this body of research before surprising you all and ending my remarks in a constructive and optimistic fashion. 1. Estimates of the level of psychiatric disorder in consulting populations made by primary care physicians are almost com- pletely valueless if one's aim is to arrive at valid estimates of prevalence. 2. Although many primary care physicians make accurate assess- ments about emotional disorder, many others are very poor at making such assessments. Thus in some practices emotional disorder commonly goes undetected, while in others asympto- matic patients are commonly labelled as sick. The ability to make accurate assessments is related to personality variables and interview style. 3. Failure to detect emotional disorder can have undesirable - consequences for the patients. Although minor disorders will tend to remit spontaneously with time, more severe disorders are favorably affected if the doctor detects them. - Treatment of detected disorders is often haphazard at present. Inappropriate prescription of psychotropics appears to be widespread and inadequate training in counselling skills is the rule rather than the exception. Implications for Training Family practice residents need to acquire skills in three related areas, all of them concerned with the way in which they are taught to interview their patients. 15/ The first group of skills is concerned with the detection of psychiatric disorder. The way the interviewer starts is often of crucial importance: things often go wrong from the beginning. It is essential for the doctor to ask open questions at the outset, and to spend some time clarifying the patient's complaint. The doctor needs to be sensitive to verbal and non-verbal cues relating to psychological distress, and to possess skills to help shy patients to talk and to direct and focus communications from over-talkative, circumstantial patients. The most effective methods for helping trainees to acquire these skills are probably programmed learning

162 videotapes followed by feedback from behavioral science teachers of their own interviews with patients during the first year of training. The second group of skills is related to assessment and diag- nosis. Trainees need not only skills, but knowledge of the more important syndromes seen in a primary care setting, in order to be effective in "making their diagnoses simultaneously on physical, psychological and social levels." 16/ Teaching in this area should be combination of seminars and lectures for the required factual knowledge, and feedback from teachers of actual recorded interviews as in the first area. The last area concerns management skills. Doctors who are not confident of their ability to manage psychological disorders will not wish, or dare, to become proficient at eliciting such disorders. The teaching required here is important but time consuming for the teachers: supervision by experienced teachers of the trainees t treatment of individual patients throughout the training years. Now who is to carry out such teaching? Dr. Hiram Curry, Chairman of the Department of Family Practice at Charleston, writes: "We can anticipate...that psychiatrists will criticize as behavioral science becomes a more important part of the curriculum at the expense of traditional psychiatric training. Psychologists will vie for a larger piece of the action, and in my judgment they should get it. There has been heavy emphasis in the past on understand- ing marked abnormal behavior...(but) we now realize that we should make a heavy investment in understanding normal as well as unusual behavior, both in our patients and in ourselves." 17/ There is much to be said for this point of view; but there have been undesirable consequences. The behavioral and social scientists imported to replace psychiatrists as teachers about psychological disorders often attack their own caricature of the "medical model" with relish, but may neglect to ensure that the trainee becomes proficient in the assessment of the more important syndromes of psychological disorders. The baby is thus thrown out with the psychiatric bathwater. Dr. Curry continues: "Some traditionalists imply, by their actions, that only doctors have information which is truly valuable to aspiring physicians. This is far from true. In the future, psychologists, social workers, epidemio- logists, nutritionists and others will have more important roles as teachers."

163 Dr. Curry reaches this conclusion -- with which I largely agree -- with the following argument: "It is not a safe assumption that the study of psycho- logy in college permits us (as doctors) to understand the average person who is not well...l7/ But this knife cuts both ways. How will this motley army of social scientists learn to "understand the average person who is not well" themselves? Certainly not by studying psychology at college. All too often, it seems to me, psychiatrists and social scientists alike assume that what the family physicians need are the skills they themselves possess -- perhaps suitably scaled down. But both psychiatrists and social scientists are often unfamiliar with the typical patient seen in a primary care setting. Where the social scientist is concerned, a college sophomore with anxiety about public speaking, or an articulate middle class client with a minor sexual problem, are both rather poor role models for the kinds of problem with which his trainee will daily have to deal. In contrast to the psychiatrist, the primary care physician is likely to see transient affective disorders rather than long-standing personality disorders, and to have patients who present with a baffling melange of somatic and affective symptoms. By virtue of his medical qualifications, the psychiatrist has a unique contribution to make to the training of the family physician. He will, of course, assist other social scientists in helping the trainee to acquire appropriate counseling skills; but he has four special contributions to make. First, his knowledge of physical diagnosis will assist him in helping trainees to appreciate the interactions between psychologi- cal and physical factors in the causation of pains and other somatic symptoms. It is not easy for a social scientist without medical quali- fications to know how reasonable a trainee is being when he persists with closed questions concerned with physical disease to the relative neglect of psychological functions. Second, his knowledge of the syndromes of psychological disorder should help his trainees acquire the probe questions which are neces- sary for accurate assessment of the patient's current psychological status. The syndromes which the trainee should be able to assess are those in which diagnosis implies special management procedures: for example, those depressive states which are especially responsive to drug therapy; the presentations of substance abuse in a primary care setting; and the assessment of confusional states in the elderly. Third, he should be largely responsible for teaching the proper use of psychotropic drugs, and helping his trainee to understand how

164 any drug therapy is to be incorporated into a general plan of psycho- logical and social help tailored to the needs of the individual patient. Finally, he should be responsible for helping trainees to grasp the indications for referral to a psychiatrist. All too often, the new omnipotent brand-image of the Family Physician makes the trainee feel that he has failed if he asks for specialist help. Nor is it the case that all major syndromes should be referred, and that all minor syndromes should be treated by the family physician. On the one hand, the family physician will often have a role to play in the manage- ment of long-term major disorders, and will therefore need the skills to make appropriate clinical assessments, when the occasion demands, of patients who are receiving depot injections of phenothiazines or who are on long-term lithium therapy. And on the other hand, neurotic and characterological problems often require specialized psychothera- peutic skills for their management which it would be unreasonable to expect every family physician to possess. O —_ _~ - _ . . . . ~ ~ ~ _ ~ ~ ~ I _ I will finish on an hortatory note. Our first task is not to train today's family physician; nor is it to train the physicians of tomorrow. Our first task is to train the teachers. Psychiatrist and social scientist will both be required on the faculty, for each has expertise that the other all too often lacks. But there must be an agreement about the aims of the teaching programs; and, above all, the teaching must be adapted to the particular needs of the family physician. -

165 References Shepherd, M., Cooper, A.B., Brown, A.C. & Kalton, G. (1966) Psychiatric Illness in General Practice, Oxford U. Press. 2. Marks, J., Goldberg, D., Hillier, V.F. (1979), "'Determinants of the Ability of General Practitioners to Detect Psychiatric Illness," Psycho. Med. 9 (May). 3. Locke, B.Z., Finucane, D.L., Hassler, F. (1967), "Emotionally Disturbed Patients Under Care of Non-Psychiatric Physicians," Psychiatric Epidemiology & Mental Health Planning, Monroe, R. et al, (eds) Psychiatric Research Report (A.P.A), pp. 235-248. Locke, B.Z., Gardner, E.A., (1969) "Psychiatric Disorders Among Patients of General Practitioners & Internists." Public Health Reports 84, 167-173. - 5. Leopold, R.D., Goldberg, D., & Schein, L. (1971), "Emotional Disturbances Among Patients of Private Non-Psychiatric Physicians in an Urban Neighborhood." NIMH Contract HSM-42-69-79. U. of Pennsylvania. 6. Goldberg, D.P. & Steele, J. (In preparation). 7. Wilson, G. (1975), "Manual of the W.P.A.I.," NFER, London. 8. Goldberg, D., Kay, C. & Thompson, C. (1976) "Psychiatric Morbidity in General Practice and the Community," Psycho. Med. 6, 565-569. Johnstone, A. and Goldberg, D. (1976) "Psychiatric Screening in General Practice," Lancet, 1, 605-608. 10. Ginsberg, S. & Brown, G., (1979), Personal communication to author. 11. Hawton, R. & Blackstock, E. (1977) "Deliberate Self Poisoning implications for psychotropic drug prescribing in general practice." J. Royal Call. Gen. Pract. 27, 560-563. 12. Downey, R.W. & Rickels, K. (1974) "Mixed Anxiety-Depression," Arch. Gen. Psychiat. 30, 312-317. Parish, P.A. (1971) "The Prescribing of Psychotropic Drugs in General Practice." J. Royal Call. Gen. Pract. Suppl. No. 4, Vol. 21 (No. 92~.

166 14. Hyams, L., Green, M.R., Haar, E., Philpot, J., Meter, K (1971) "Varied Needs of Primary Physicians for Psychiatric Resources," Psychosomatics 12, 36-45. 15. Goldberg, D. (1979) "Detection & Assessment of Emotional Dis- orders in a Primary Care Setting," Int. J. Ment. Health, in press. 16. Browne, K. & Freeling, P. (1966) Doctor-Patient Relationships, Baltimore: Williams & Wilkins e 17. Curry, H.B. (1977) "Strategy for Achieving Academic Objectives of Family Medicine" in Academic Missions of Family Medicine. Bryan, T.E., (ed) DREW Publication (NIB) 77-1062, pp. 93-104.

167- THE PRIMARY CARE PHYSICIAN: MENTAL HEALTH ISSUES Robert S. Lawrence, M.D. Division of Primary Care and Family Medicine Harvard Medical School Boston, Massachusetts Kafka wrote in A Country Doctor that "to write prescriptions - is easy, but to come to an understanding with people is hard." 1/ That truth undoubtedly explains in large part the prominence enjoyed by psychotropic drugs among the 1.4 billion prescriptions written last year for the seventy million Americans who regularly take drugs ordered by physicians. How did this gap emerge between the techni- cal skills and knowledge exemplified by the use (and abuse) of modern pharmaceuticals and the all-too-often deficient mental health skills of the prescribing physician? Rasmussen traces the separation of medicine and psychiatry back four centuries to the decision by the Church to permit the study of the human body by dissection. In exchange for this permission, the medical community accepted a tacit prohibition against corresponding scientific investigation of man's mind and behavior--attributes which were firmly identified with the soul. 2/ Engel observes that with "mind-body dualism firmly established under the imprimatur of the Church, classical (medical) science readily fostered the notion of the body as machine, of disease as the consequence of the breakdown of the machine, and of the doctor's task as repair of the machine." 3/ Into this gap between mind and body steps - or is pushed the primary care physician who is often expected by his patients to be part mechanic and part priest. Functioning as he does in what Friedson terms a client-dependent form of practice, 4/ the primary care physician should, and usually does, feel an obligation to fulfill or at least acknowledge the patient's expectations. Furthermore, one of the distinct attributes of primary care is an understanding of the doctor-patient relation- ship and how this relationship can be used to ameliorate physical illness and mental health problems. Carmichael has argued that family medicine as one form of primary care is more identified with psychiatry as a "behavioral discipline" in contrast to the traditional surgical and medical disciplines. 5/ The functions of the "behavioral disciplines" are mainly helping

168 and caring rather than healing or curing, and the relationships are characterized by mutual participation and not activity-passivity or even guidance-cooperation. The primary care physician trying to pull medicine and psychi- atry together soon learns that each discipline brings with it very different personal rewards. Physicians and medical students value making a diagnosis, providing successful treatment or controlling the patient -- actions which provide prompt or immediate gratifica- tion for the doctor. Less valued are giving help and care and making personal affiliations -- actions which are slower and for which gratification is usually delayed. 6/ The pace of these latter services and rewards is matched by the process of slow assimilation of material needed to learn the psychiatric contributions to medicine - learning which has difficulties involving "not only intellectual understanding but also emotional processes, mobiliza- tion of anxiety, empathy with patients, and the growth and maturity of the student," resident or physician. 7/ Put another way, the primary care physician extends his knowledge of medicine by including new ideas and discoveries -- by reading journal articles or attending conferences. In psychiatry, on the other hand, "knowledge generally increases not through new information but through deeper understand- ing of the known." 8/ Abundant opportunities exist for the primary care physician to acquire this deeper understanding. Lipowski estimated that 50-80% of medical outpatients have psychiatric illness serious enough to warrant special attention. 9/ Goldberg diagnosed a psychiatric illness in 34% of patients with small bowel disease attending a follow-up clinic at St. Thomas' Hospital. 10/ This rate is close to that in patients seeing a general practitioner and higher than rates in the community. 11/ After the Mid-town Manhattan study revealed that 81.5% of the population had emotional symptoms it became clear that we needed to distinguish between mental disorders and the emotional problems of living. Using the nosology of mental disorder in Section V of the I.C.D.A. to classify psychiatric visits, the National Ambulatory Care Survey of 1973-74 provided the following data for 645 million physician visits (about 70% of total ambul`~- tory visits in the United States): "Mental disorder was diagnosed in 78Z of visits to psychiatrists, 4% of visits to general practi- tioners, and 9% of visits to internists. Because of the larger number of visits to non-psychiatrists, 59% of all visits by patients with mental disorders were to these physicians." 12/ How did the physicians respond to these patients? Overall, 4% of visits included psychotherapeutic/therapeutic listening therapy. Psychiatrists provided this therapy in 73% of visits with a mean

169 duration of 45 minutes while non-psychiatrists used psychotherapeutic listening in 2% of visits with a mean duration of 19 minutes. Again because of greater numbers the non-psychiatrists accounted for 46% of all visits which provided psychotherapeutic/therapeutic listening. In patients with a principal diagnosis of mental disorder the non- psychiatrists all used other treatments, such as drug therapy, medical counselling/advice or general history and physical exam, more than psychotherapeutic/therapeutic listening. Inadequate or inappropriate response to patients with mental disorders is by no means limited to the ambulatory setting. Moffic and Paykel detected 43 depressed patients among 150 medical in- patients; but only six had any mention of depression in the chart, four were receiving anti-depressants and two had been referred to psychiatry. 13/ The behavior of non-psychiatrists in both settings may reflect the fact that psychiatry as a body of knowledge is in- trinsically anxiety-provoking. 14/ Lazarson notes that "even more anxiety-provoking than the theoretical body of knowledge is work with the clinical source of all theory -- the patient."' 15/ The affect of patients with mental disorders re-evokes in the physician those "problems and conflicts whose existence and solution took place exclusively in his unconscious, and towards whose revival there are the strongest resistances." 16/ The consequences of this behavior for primary care physicians already burdened with anxiety about their patients' medical needs are that emotional problems often present as an enigma. Some physicians completely fail to recognize mental disorders or are uncomfortable with patients whom they label malingerers, hypochondriacs, crocks or "problem patients." Management of these patients often includes extensive diagnostic or treatment plans with the minority of patients receiving time for ventilation and therapeutic listening through repeated visits. In their study of 88 general practitioners Peterson et al found that most physicians recognized the emotional situation of some patients, some physicians attempted to treat emotional problems without any real basis of knowledge, and only a few physicians were prepared to do something about the problems. 17/ Paradoxically, the characteristics of the therapeutic relation- ship in primary care practice, with more spontaneous intimacy and greater continuity, should enable the physician to treat mental illness more effectively than specialists. 18/ Many primary care physicians do possess good interviewing skills, develop therapeutic relationships with their patients, are empathic, and support patients through normal life crises. But these skills and attitudes which suffice for the emotional problems of living are inadequate to deal with mental disorders such as: (in order of frequency in primary care practice) neurotic disorders, psychophysiologic disorders, behavioral disturbances of childhood, character dis- orders, and more severe depressive and schizophrenic psychotic disorders. 19/

170 No one is more aware of these deficiencies in mental health skills than the primary care physicians themselves. Forty-six percent of family physicians surveyed by Fisher,et al.were dis- satisfied with their own competence to treat mental illness, and only 10% thought their psychiatric training very good in relation to their present practice. 20/ Similar results have been reported for internists. 21/, 22/ When one compares the mental health skills primary care physicians think important with the opinion of psychiatrists and psychologists, a high level of agreement exists. A survey of psychiatrists, psychologists and non-psychia- tric physicians produced the following list of the ten most important topics in psychiatry which the physician should know: Interviewing Suicide evaluation Psychopharmacology Chronically-ill or dying patients Psychophysiologic disorders Psychiatric referral Doctor-Patient relationship Drug and alcohol abuse Differential diagnosis of mental illness Sexual problems. _/ Rural general practitioners ranked all but drug and alcohol abuse and sexual problems in their list of most important topics in psychiatry. 24/ In another study of family practice Werkman,et al. found alcohol abuse and marital problems in the most common psychia- tric disorders. 25/ The fact that they are often the most difficult disorders to treat may account for the rural practitioner's failure to include them among the most important topics -- an example of inner resistance against threatening material in the body of psychiatric knowledge. 26/ In the face of these difficulties what accounts for the current emphasis on teaching psychiatry and behavioral science to primary care residents? Certainly it would be easier to support the resident's enthusiastic and skillful pursuit of organic disease while makir~ other arrangements for dealing with those mental disorders which are less well-defined and frequently viewed as less legitimate aspects of patient care. 27/ But in moving beyond the biomedical model we acknowledge that the primary care physician is ideally placed for the prevention, diagnosis and treatment of mental disorders for the following reasons: "many mentally disturbed patients present with physical complaints; physical and mental illness frequently co- exist; some stigma is still attached to seeing a mental health specialist; the important social and family context of the patient are already known; and the potential for long-term follow-up is present.'' _ / To take advantage of these opportunities the primary care physician needs to acquire specific mental health skills.

171 Several authors have described curricular objectives for teaching these skills, 29/, 30/ and a detailed list of educational goals was submitted to the Bureau of Health Manpower by the American Psychiatric Association in 1977. As part of a curriculum project for the Harvard Care Program, Lipsitt, et. al. described the mental health skills which when mastered by the primary care physician would permit him to function effectively in treating the "whole" patient as well as deal with problems commonly seen in general practice. 31/ These skills are broadly defined in three categories: sensitivity skills, therapeutic skills, and referral skills. First, the sensitivity skills; the overall objective in this category is to educate primary care physicians to be more sensitive to patients and their needs for treatment and understanding. Implicit to the achievement of this objective is the physician's awareness of his own reactions and the effect on him and on the patient of the treatment process or the building of a therapeutic alliance, To have this awareness the physician must be capable of sufficient intro- spection to assess his own attitudes and values. One measure of this sensitivity to patients is the degree to which the physician shares Carmichael's conviction that it is his responsibilty to "establish a setting in which the patient can state his thoughts and feelings with complete freedom." 32/ Sensitivity to patients also requires an understanding of the process of normal growth and development through the life cycle, and of the problems of adaptation and solutions at each stage of the cycle. Understanding of economic, class and environmental influence on illness is also necessary for physician awareness of treatment needs in the context of the patient's own frame of reference. 33/ Knowledge of the influence of these psychosocial factors must then be combined with knowledge of community resources for planning the most appropriate treatment. Only when these skills of sensitivity are developed can the primary care physician make intelligent judgments about the proper mix of therapeutic listening, crisis intervention, or empathic support for a patient with a problem of living. The therapeutic skills are built on the foundation of sensitivity skills. Here the primary care physician learns how to counsel various kinds of patients with problems complicating, causing, or resulting from their medical and/or psychosocial conditions. To provide this counselling the physician needs: to understand the "psychosomatic approach" in its broadest sense; to be experienced in the rudiments of history taking, interviewing skills and minor psycho-therapeutic techniques; and

172 to recognize and deal with the acute (though often low level) anxiety and depression frequently related to illness. The importance of the history-taking and interviewing skills cannot be stated too strongly. Engel states that "the most essen- tial skills of the physician involve the ability to elicit accurately and then analyze correctly the patient's verbal account of this ill- ness experience. The biomedical model ignores both the rigor required to achieve reliability in the interview process and the necessity to analyze the meaning of the patient's report in psychological, social and cultural, as well as in anatomical, physiological, or biochemical terms." 34/ By application of these principles the primary care physician can recognize and deal with emotional and mental problems in patients presenting with essentially non-psychiatric conditions, understand the issues which impact on patient compliance, and support the patient in following the medically indicated regimen. In addition to managing common anxiety and depression these skills enable the primary care physician to deal with problems of living; couples' counselling; and marital, sexual, adolescent and family problems. Therapeutic skills enable the physician to manage problems of grief, terminal illness and adjustment to disability; and provide the base for acquiring knowledge and experience in the disorders of craving (obesity, alcoholism, and drug abuse) and awareness of personality patterns that place the patient at risk of developing habituation to medications prescribed by the physician. The category of referral skills includes learning to recognize serious psychiatric disorders and developing skills of the refer- ral process, utilizing appropriate resources. To make appropriate referrals for specialized mental health services the primary care physician must: develop diagnostic skill for the major psychiatric disorders; be able to elicit data relevant to suicidal or homicidal potential; - know how to arrange for psychiatric hospitalization when needed; be able to provide initial management of psychiatric emergencies until the services of a psychiatrist can be arranged if necessary; know how to work with psychiatrists in providing proper treatment; and

173 referrals are facilitated, and the primary care physician can grad- ually increase the level of responsibility he assumes for managing more seriously ill psychiatric patients or medical patients with significant mental disorders. The disadvantages center around the disruptive influence of severely-disturbed patients on the practice itself and the occasional patient whose mental disorder or the intensity of the transference-counter-transference distracts the primary care physician from appropriate management of medical problems in that patient. At the moment this seems to be a small price to pay for the opportunity to break down distinctions between psycho- logical and physiological events. As Dr. Regier and his colleagues noted, much of the ability of primary care physicians "to deal with more explicitly defined organic illness may well depend on the ease and compe fence wi th which they respond to the pre sentat ion o f mental disorder and broader emotional concerns . " 40 /

174 REFERENCES Kafka, F., A Country Doctor, quoted in Balint, M. et al. Treat- ment or Diagnosis: A Study of Repeat Prescriptions in General Practice, Tavistock Publications, J.B. Lippincott, 1970. 2. Rasmussen, H., Medical Education - Revolution or Reaction, The Pharos 38:53-59, 1975. 3. Engel, G.L., The need for a new medical model: a challenge for Biomedicine. Science 196:129-136, 1977. 4. Friedson, E., Professional Dominance: The Social Structure of Medical Care, New York: Atherton Press, Inc., 1970. 5. Carmichael, L.P., Psychiatry and Family Medicine, the Behavioral Disciplines. The New Physician 19:524-527, 1970. 6. Ford, A.B. et al., The Doctors Perspective, Press of Case Western Reserve University, Cleveland, 1957. Psychiatry and Medical Education, Conference Report of the American Psychiatric Association, Washington, D.C., 1952. 8. Reichard, J.F., Teaching Principles of Medical Psychology to Medical House Officers: Methods and Problems. In Psychiatry and Medical Practice in a General Hospital, ed. Zinberg, N.E. International Universities Press, New York, 1964. 9. Lipowski, Z.J., Review of consultation psychiatry and psycho- somatic medicine. Psychosom Med. 29:201-024, 1967. 10. Goldberg, D., Psychiatric study of patients with diseases of the small intestine. Gut. 11:459-65, 1970. Editorial "Psychiatric Illness Among Medical Patients," The Lancet, 8114, March 3, 1979. 12. Report of the NIMH Workgroup on Mental Health Training of Primary Care Providers, Washington, D.C., January 1977. 13. Moffic, H., Paykel, F.S., Depression in medical in-patients Br. J. Psychiatry 126:346-53, 1975. 14. Training the Psychiatrist to Meet Changing Needs: Report of the 1962 Conference of Graduate Psychiatric Education, American Psychiatric Association, 1963.

175 15. Lazarson, A.M., The learning alliance and its relation to psychiatric teaching. Psychiatry in Medicine, 3:81-91, 1972. 16. Bandler, B. Ego-centered Teaching, Smith College Study, Soc Work 2:125-136, 1969. 17. Peterson, O.L., et al., An analytic study of North Carolina General Practice: 1953-54, J. Med. Ed. 31: iii-iv, 1-165, 1956. 18. Pierloot, R.A., The treatment of psychosomatic disorders by the general Practitioner. Intl J._Psychiatry in Medicine 8:43-51, 1977-78. 19. NIMH Workgroup report. Op. cite 20. Fisher, J.V., Family physicians want more postgraduate psychi- atric training. Patient Care 7:54-57, 1973. 21. Engstrom, W.W.: Are internists functioning as family physicians? Ann. Int. Med. 66:613-616, 1967. 22. Deller, J.J., On the training of the internist for the practice of internal medicine: a point of view. Ala J Med Sci 13:362- 366, 1976. 23. Johnson, W., Snibbe, J., The selection of a psychiatric curriculum for medical students: results of a survey. Am J Psych 132:513- 516, 1975. 24. Callen, K.E., Davis, D., What medical students should know about psychiatry: the results of a survey of rural general practi- tioners. Am J Psych. 135: 2, 243-44, 1978. 25. Werkman, S.L., Mallory, L., Harris, J., The common psychiatric problems in family practice. Psychosomatics 17 :119-122, 1976. 26. Kaufman, M.R., The teaching of psychiatry to the nonpsychiatrist physician. Am J Psych 128:5, 610-16, 1971. 27. Gardner, E., Emotional disorders in medical practice. Ann Ins t Med 73:651-653, 1970. 28. NIMH Workgroup report. Op. cit. 29. Kinzie, J.D., et al., Objectives of psychiatric education in a primary care curricul~m. J Med Ed 52:664-67, 1977. 30. Lazarson, A.M., The psychiatrist as teacher in primary care residency training: the first year. Inter J Psych in Med 7:~2~:165-178, 1976,77.

176 Lipsitt, D.R., Harburger, J.S., Johnson, B.P. "Report of the Behavioral Science Subcommittee" in Training Adult Primary Care Residents in an EMO: An Approach Based Upon Behavioral Objectives by Budd, M.A., Hatem, C.J., and Lawrence, R.S., Cambridge, MA., 1975. 32. Carmichael. L.P. Op. cit. Kleinman, A., Eisenberg, L., Good, B., Culture, Illness and Care. Ann. Intern. Med. 88:251-258, 1978. 34. Engel, G.L. Op. cit. 35. Lipsitt, D.R., et al, Op. cit. 36. Lazarson, A.M., Inter J Psychiatry in Med., Op. cit. 37. Reichard, 38. Balint, M. JF. Op. cit. The Doctor, His Patient 9 and the Illness, International Universities Press, New York, 1964. 39. Schniewind, H.E., A psychiatrist's experience in a primary health care setting, Inter J Psychiatry in Med 7~3~:229-240, 1976-77. 40. NIMH Workgroup report. Op. cit.

177 REIMBURSEMENT BY MEDICARE FOR MENTAL HEALTH SERVICES BY GENERAL PRACTITIONERS - CLINICAL, EPIDEMIOLOGIC AND COST CONTAINMENT IMPLICATIONS OF THE CANADIAN EXPERIENCE Alex Richman, M.D., M.P.H. Murray G. Brown, Ph.D. Dalhousie University Halifax, Nova Scotia This paper describes some Canadian experiences in paying for psychia- tric services by general practitioners under a national health insurance program. The first part of the paper outlines the national health in- surance program; the second part gives national and regional data on utilization; the third part describes some cautions and concerns these data raise for future programs, and the paper concludes with an agenda for future research. CANADIAN NATIONAL HEALTH INSURANCE FOR PSYCHIATRIC SERVICES In 1961 the Canadian Government established a Royal Commission on Health Services to review the state of health services and make recom- mendations for the future. A Royal Commission project on the extent and results of psychiatric services in Canada identified marked limita- tions in insurance reimbursement for psychiatrist's care. 39/ There was very little opportunity for general practitioners to be reimbursed for counselling and psychotherapy. 45/ The 1964 Report of the Royal Commission on Health Services laid the framework for national health insurance. 8/ Recommendation Number 29 stated that: "Henceforth all discrimination and distinction between physical and mental illness in the organization and provision of service, the treatment, and the attitudes on which these discriminations are based, be disallowed for all time as unworthy and unscientific." The authors acknowledge the unpublished tabulations from Health Information Division, Information Systems Directorate, Policy, Planning and Information Branch, Department of National Health and Welfare, Canada; Program Development Division, Nova Scotia Department of Health; Maritime Medical Care Inc., and Carl d'Arcy, Ph.D., University of Saskatchewan; the help of Mrs. B. Brunelle, and the Audio-Visual Division of Dalhousie Faculty of Medicine.

178 The Commissioners specified that the medical services benefit should incorporate insured medical services for "the diagnosis and treatment of all physical and psychiatric conditions including mental retardation" (Recommendation 30a). By 1971 all 10 provinces had introduced comprehensive medicare programs for all age groups. The provincial programs were very similar since, although provision of health care in Canada falls within provincial jurisdiction, the provinces had to meet federal requirements in order to benefit from the federal cost-sharing agreement which began in 1968. These federal requirements included: 7/ (1) comprehensive coverage for all medically required services rendered by a physician or surgeon; (2) universal availability to al' eligible residents on equal terms and conditions, with insurance coverage for at least 95% of the total eligible provincial population; (3) accessibility uninhibited by excessive user charges; (4) portability of benefits among Provinces - a resident of one Province would be able to get equivalent benefits if he moved to another Province; (5) administration on a non-profit basis by a public authority. The provincial medicare programs did not change the delivery of clinical services, which continued to be provided mainly by private practice physicians working on a fee-for-service basis. 24/ The patient has choice of physician; the physician submits a claim for an individual patient. The method by which the provincial costs are financed varies from Province to Province; in Ontario there is an annual premium, Nova Scotia has a sales tax. Currently physician satisfaction with medicare varies, satisfaction being highest in the Maritime Provinces. 33/ UTILIZATION OF PREPAID PSYCHOTHERAPY Table 1 shows the use of prepaid psychotherapy and counsell:ing services in 3 Canadian regions over a 6-year period, 1972-1978. (Consultations and hospital visits are excluded). During 1977-78 the population of 22.6 million used 3.8 million services costing $91.2 million. In relation to population there were 170 services per 1,000 population costing $4,043 per 1,000 population. Over two- fifths (42.75%) of these services were provided by general practi- tioners. Regional utilization ranged from 76 to 194 services per 1,000 (Figure 1~. In relation to total expenditures (excluding pathology, radiology and certain specialized services) psychiatric

1972-73 179 TABLE 1 Psychotherapy Services: Number of Services and Payments, Rates per 1, 000 Insured Persons, Percentage Distribution by Specialty, Nine Provlnces, by Region, 1972-73 to 1977-78 Preliminary Maritimes Central Western 9 Provinces Number of Services (No, ) 58,522 1,792,443 540,933 ~,391,898 Payments ($) 820, 595 30, 830, 389 9, 412, 382 41,13~., 366 Services per 1000 Insured Persons (No. ) 38.22 129.32 93.98 113.11 Payments per 1000 Insured Persons ($) 535,99 2,225.06 1,645.65 1,945.07 Percent done by: General Practice (I) Psychiatry (I) Other (%) 1973-74 Number of Services (Jo. ) Payments ($) Services per 1000 Insured Persons (No. ) Payments per lO00 Insured Persons ($) Percent done by: General Practice Psychiatry Ocher 1974-75 36.16 33.38 59.80 63.54 4.04 3,08 72,054 1,091,206 1,087,816 34,413,849 46, 55 142.18 702.72 2,457.25 (I) (%) (I) 38.68 59.21 2.11 585, 151 10,231,261 q9.97 1,748.04 2,648,411 45, 732, 926 123.72 2,136.4S 35.95 33, 02 35. 38 60.13 64.12 60.99 3.92 2.86 3.64 Number of Services (No, ) 91, 587 2 ~ 306, 866 685, 096 3, C83, 549 Payments ($) 1,702,349 43,732,036 12,376,181 57,810,566 Services per 1000 Insured Persons (No . ) 58. 29 162.48 114. 64 141.80 Patents per 1000 Insured Persons ($) 1,083.61 3,080.15 2,070.98 2,658.57 Percent done by: General Practice (%) 44.69 38.43 34.54 37.75 Psychiatry (I) 51.40 57.89 62.98 58 . 83 Other (%) 3.91 3.68 2.48 3.42 1975-76 . _ Number of Services (No. ) Payment s ($ ) Services per lO00 Insured Persons (to. ) Payments per 1000 Insured Persons ($) Percent done by: General Practice Psychiatry Other 976_77 105,113 2,658,953 2,153,432 51,890,988 65.94 185.11 1,35D,96 3,612.57 (%) (%) (%) 45.13 50.59 4.28 762,824 3,526,890 16,702,889 70,747,309 125.05 159.8q 2,738.18 :3,207.33 39.55 43.76 S6.38 53,35 4,07 2,89 40.63 55.56 3.82 Number of Services (No,) 113,914 2,741,233 851,717 3,7C6,864 Payments ($) 2,525,174 58,493,260 19,603,574 809712,008 Services per 1000 Insured Persons(No, ) 70.58 188.86 137.02 165.89 Payments per 1000 Insured Persons ($) 1,564.54 4,029.85 3,168,21 :3,612.08 Percent done by: General Practice (%) 48.46 39.53 46.90 41.50 Psychiatry (%) 47.24 58.32 49.83 56.03 Other (%) 4.31 2.16 3.27 2.48

180 TABLE 1 CONTINUED Maritimes Central Western - lq77.-78 _ Number of Services Payments 123,431 3,096,355 Services per 1000 Insured Persons(No.) 75.77 Payments per 1(~(30 Insured Persons ($) 1,90(J,77 Percent done by (No,) ($) r 9 Provinces 2,83Q,857 66,384,856 194.24 4,540.68 885,17S 21,756~644 140.08 3,443.05 3,848,463 91,7~37,SS5 170.53 4,042.80 General Practice (%) 49,90 40,00 50.58 42.75 Psychiatry (%) 47.07 59.16 46.40 55.84 Other (%3 3~03 0.85 3.02 1.42 NOTES 1. The figures are based upon fee-for-ser~nce payments made by the medical care insurance plans of nine provinces to physicians residing in the province making the payments. Out-of-province payments are excluded. Also excluded are services not covered by provincial plans, e.g. Workers' Compensation, services rendered by physicians paid on a salary, sessional or other non-fee basis. 2. Psychotherapy includes individual, group and family psychotherapy or psychoanalysis, hypnotherapy, and certain "counselling" services. 3. Figures are adj usted to make them as camp arab le as pass ib le . For example, in provinces where psychotherapy services are paid in terms of 15 or 30 minute units, it is assumed that the average psychotherapy service takes 45 minutes. 4. Data for the two central provinces and for Manitoba are on a date-of- service basis. Those for the other provinces are on a date-of-payment basis . 5. The "Maritimes" consist of Prince Edward Island, Nova Scotia and New Brunswick. Newfoundland is excluded because comparable data for earlier years were not available e "Central" comprises Quebec and Ontario. The three prairie provinces and British Columbia make up the "Wes tern" p rovinces . 6. Grouping and adjustment procedures were revised recently. Checking of the results is still under way, and consequently the figures are treated as "preliminary" until further notice. Health Information Division, Information Systems Directorate, Policy, Planning and Information Branch Department of National Health ~ Welfare, March 19 79 .

181 FIGURE I PAYMENT FOR PSYCHOTilERA PY, CANAD I AN REG I ONS 1 972 - 1 977 4,00O1 ~ 500- c~ ~CENTRA L P ROV I NC ES — NINE PROVINCES WESTERN PROV I NCES ~ARITlME PROVINCES f rem fable 1 250- , . . . . 1972-73 1973-74 1974-75 1975-76 1976-77 1977-78 YEA ~ FIGURE II PROPORT I ON OF PA YMENT S FOR P SYC HOTHERA PY TO GENERAL PRACTITIONERS NINE CANADIAN PROVINCES 10% 20% 30% 40% 1972-731- / \ ~ ' 20% 1974-75 r 10% A 1 2 3 l \ . _ \ 20% 50% 4)~8\ 9' 30% 40% - 50%

182 services amounted to 5.2% of the total costs of insured physician services, with an inter-regional range of 3.0% to 5.~%. In comparison to six years earlier (1972-1973) the absolute number of services in 1977-78 had increased by 61%, the cost by 122%, the services per 1,000 population by 51%. The proportion of medical payments for psychotherapy services had increased from 3.95% in 1972-73 to 5.2% in 1977-78, an increase of 32%. The proportion of psychotherapy services from general practitioners increased from 35.45% to 42~75%. Provincial trends in the proportion of psychotherapy services from general practitioners were not uniform. Figure II shows tile proportion of psychotherapy services by general practitioners in nine Canadian provinces between 1972-73 and 1974-75. In some provinces the proportion of psychotherapy by general practitioners decreased (Provinces 1, 4 and 9), while some provinces (No. 8) showed striking increases. Later this paper will refer to some of the factors related to increased psychotherapy by general practitioners. RECORDING OF PSYCHIATRIC MORBIDITY IN CANADIAN INSURANCE PLANS Other papers in this Conference have shown the wide variations in recording of psychiatric morbidity by general practitioners. There is evidence that the recording of psychiatric morbidity by general practitioners can be enhanced under health insurance. In contrast to the United States, where the majority of private practice office visits for mental disorders are to psychiatrists, 35/ more Canadian patients are given diagnoses of mental disorders by general prac:ti- tioners than by psychiatrists. Table 2 shows the pre- and post--medi- care experience of general practitioners in Quebec. The number of psychiatric visits nearly tripled and variation among physicians in the recording of mental disorder was halved. 31/ The Canadian system meets many of Shepherd's desiderata for a cooperative psychiatric morbidity survey of general practice. 47/ (1) Participation of a large representative group of collaborating doctors. (2) Use of a uniform diagnostic classification. (3) Use of clearly defined standardized indices for measuring morbidity in terms of time periods and in relation to population-at-risk. The Canadian universal prepaid hospital insurance plan does not discriminate against mental disorders. The provincial data bases of individual hospitalizations and medical services enable detailed studies of the use by the general population of specialized psychiatric services, mental health services from private practitioners, general

All Visits and Consultations 183 TABLE 2. PSYCHIATRIC SERVICES BY ACTIVE GENERAL PRACTITIONERS QUEBEC, JANUARY-MARCH, 1971 AND 1975 _ . . 1971 1975 Mean Change 2~003 1~629 -19% Number Coefficient of variation .61 . 50 -18% Mean 9, 735 11, 579 +19% Payments Coefficient of variation .43 .45 + 5% Psychiatric Visits and Consultations Mean .93% 3.24% +248% Visits as Percentage Coefficient of of All Visits and variation 6.86 3.4 -50% Consultations Mean 1.477 4~27% +190% Payments as Coefficient of Percentage of Pay- variation 4.98 2.73 -45% meets for All Visits and Con- sultations From Mathematica Policy Research, 1978 hospital care for mental disorders, and in some provinces with pharma- care the use of prescribed psychopharmaceutical agents. 26/ These data, by and large, have not been extensively used for health services research. The next section describes the uses of these data bases for epidemiologic research in Saskatchewan. 10/-15/ CONSPICUOUS PSYCHIATRIC MORBIDITY - SASKATCHEWAN Saskatchewan led the Canadian provinces in establishing prepaid hospital care in 1947 and prepaid medical care in 1962. Carl d'Arcy of the Psychiatric Research Branch at the University of Saskatchewan recently completed a detailed analysis of conspicuous psychiatric: morbidity recorded for Saskatchewan residents over a 2 year period.

184 He collated the data bases of the specialized psychiatric services, general hospitals and fee-for-service physicians to develop a file of unduplicated persons given a diagnosis of a mental disorder (including ICDA-8 category 793.0) during a two year period in the province of 920,000. In the "pyramid" of Figure III patients are classified in a hierarchical manner -- persons seen in more than one sector during the two year interval are assigned only to the top-most sector and not to lower sectors. The cumulative percentages shown are population-based, not the proportion of patients attending physicians. (Over 20% of residents did not attend a physician during a calendar year.) a) Specialized psychiatric inpatient services - about 2,400 persons were first admissions -- 1/4% of the population. c) b) Specialized psychiatric outpatient services - about 6,000 persons were first admissions. Cumulatively 0.91% of the population were first admissions to the specialized psychia- tric services during the two year period. Specialized psychiatric services - another 10,000 individuals were readmitted during the two year period. Cumulatively, 2% of Saskatchewan residents were seen in the specialized psychiatric services. d) General hospital non-psychiatric wards - another 10,000 individuals were hospitalized in non-psychiatric wards. Up to this point 28,000 individuals, 3.1% of the general population, were recorded from the specialized psychiatric services and general hospital in-patient wards. e) Private practitioners recorded a diagnosis of mental disorder for another 94.6 thousand individuals. Cumulatively, 13.3% (1 in 7.5) of Saskatchewan residents were recorded by medi- cal agencies or physicians as having a mental disorder during the two year period. The age group 20-39 had the highest rate of recorded mental disorder, 25.6% for women and 13.0% for men. Diagnostic Distribution - One Calendar Year (1972) Figure IV shows the diagnostic distribution among various treat- ment sectors during 1972. The height of each bar is proportionate to the number of patients reported in that sector; the width shows the proportion of patients within the specified diagnostic group. Func- tional psychoses made up 30 percent of the psychiatric inpatient service patients, 9 percent of admissions to the mental health clinic, 7 percent of the mental disorders in general hospital wards, and 4 percent of the mental disorders recorded by physicians in private practice. The picture for psychoneuroses and psychosomatic disorders

185 FIGUXE III PSYCH IATR I C . I N - P AT I ENT S E RV I C E 2, 400 (F I RST ADMI SS I ONS) i MENTAL HEALTH C L I N I C 6 , 000 (Fl RST ADMI SS IONS) NOT F I RST A D M I S S I O N S 10 , 000 TO PSYCH IATRIC SERV I CE G E N E R A L HO S P I TAL 10, 100 W A R D S PHYS IC IANS P R ~VATE 94, 600 P R A C T I C E PERSONS WITH DIAGNOSED MENTAL DISORDERS SASKATCHEWAN 1971 - 1972 ( FROM d'ARCY 1976 ) C UMU LATI VE PE RCENTAG .26% . 91~o 1. 99% 3.09% 13. 33% TOTAL = 123, 000 P E R S O N S 13. 33% O F P O P U LA T I O N

186 UP Cut Cat A ~ . ~ _ ~ ~e, ae up o C~ ~ o ,L,,~__i~ ]~ Cat—Z ~ tY o _ — ~ ~ ... Z In y o Z U} o Z tT In ._ A, .. _ ~ ~ ~ o A t_ as, ~ —: lo ~~ o S—° - __ ~ of o ~ ~ ~ lo 11

187 is different. Twenty-five percent of psychiatric in-patients had a diagnoses of neurosis or psychosomatic disorder in comparison to 73 percent of the patients seen by physicians in private practice. Care of Patients with Mental Disorders by General Practitioners Figure V and Table 3 show the distribution of 112,000 patients seen during 1971-72 by private practitioners in Saskatchewan. A1- together general practitioners recorded 99,000 individuals as having a mental disorder. There is a small overlap between psychiatrists and general practitioners -- about 7 percent of the 112,000 patients were seen by both psychiatrists and general practitioners; on the other hand, 77.8 percent (87,000) of the mental disorders recorded by private practitioners were seen only by general practitioners during 1971-72. Figure VI shows the distribution of 540,886 medical services among the 112,000 patients seen by fee-for-service practitioners during the two-year period. The lefthand side of Figure VI shows the distri- bution of 112,000 patients; the righthand side the distribution of 540,886 services for mental disorders; the middle set of figures shows the number of medical services per patient during the two year period. Over three-fourths (77%) of patients had under five services - making up one-fourth (28.1%) of the total medical services for mental disorders during the two-year period. Persons with more than 20 ser- vices were 4 percent of the patients but took up 39 percent of the services. Although there is a mean of 4.8 services per patient the distribution is skewed. Figure VII shows the Lorenz distributions of patients with mental disorders and their associated medical services, for psychiatrists and general practitioners separately. The abscissa shows the cumula- tive percentage of patients, the ordinate the cumulative percentage of services. One-half of the patients seen by general practitioners took up about 10 percent of their mental health services; one-half of the psychiatrists' patients used about 20 percent of the psychiatrists' services. There is inequality in the utilization of mental health services for both general practitioners' patients and psychiatrists' patients; the proportion of services taken up by a minority of men- tally ill persons is actually greater among the general practitioners than among the psychiatrists. PSYCHOTHERAPY BY GENERAL PRACTITIONERS - NOVA SCOTIA Figure VIII shows the rate of prepaid psychiatric services from psychiatrists and general practitioners over an eight-year period, fiscal years 1971 through 1978. In Nova Scotia, psychotherapy is defined as:

188 FIGURE V PERSONS (N=112, 053 ) WITH DIAGNOSED MENTAL DISORDERS BY TYPE OF PHYSICIAN - SASKATCHEWAN 1971-72 (From d'Arcy 1976) - - - 77.8% - - X \ GENERAL \ PRACTITIONER \ 99, 166 = 88. 5 5.7% 3.99b / / \ - / 5.0% 6 . 0 % /12, 785 = 11.4 P SYCH I ATR I ST 13, 707 = 12. 2% TABLE 3. MEDICAL SERVICES FOR MENTAL DISORDERS, SASKATCHEWAN, 1971-1972 Services Total General Hospital Days (Care by psychiatrist or general practitioner) Psychiatrists (13,707 patients) Treatment interview First visit Consultant Other General Practitioner (99,166 patients) Minor assessment Subsequent visit Other service Psychotherapy, counselling Miscellaneous Other Service ~- From d'Arcy, 19 76 Patients Services 2,055 540,886 8,017 7,556 6,335 4,423 8,027 57,856 43,878 24,130 28,004 11,680 ~- 121,301 56,596 6,821 5,854 31,686 103,545 71,415 42,392 72,326 28,050 Per Pat; Ant 4.8 15.1 7.5 1.1 1.3 3.9 1.8 1.6 1.8 2.6 2.5

189 FIGURE VI DISTRIBUTION OF ME D I CA L SE RV I C E S FO R MEN TA L D I S O R DE R S SASKATCHEWAN 1971 - 72 ~ FROM d'ARCY 1976 ) 77% 13.4%: 4.1* U N D E R 5 SERVICES 5 -9 SERV ICES 10-19 20+ 28.1 17.7 5.1 39.1 , , , , , 1 , 1 1 80 60 40 20 0 0 20 40 60 PERCENTAGE OF NUMBER PERCENTAGE OF 112, 055 PAT 1 ENTS SE RV 1 CES 540, 886 SERV I CES FIGURE VI I PERCENTAGE OF SERVICES BY PERCENTAGE OF PATIENTS GENERAL PRACTITIONERS AND PSYCH IATRI STS - 1969 AND 1974 90 - 80 - 70 - 60 - 50 - 40 - 20 - 10 - i,? / ,Y ~,/'1 ./' ,.>' PSYCH I ATR I ST~ ~ ./ 1 974/ 1969 ',,> — —~ GENERAL PRACTITIONERS 1974~ 1969 10 20 30 40 50 60 70 80 90 1 00 CUMULATIVE PERCENTAGE OF PATIENTS / ',, ,'i' from d'Arcy

190 FIGURE VIII PSYCH IATR IC SERVICES 105- B Y P SYC H I ATR I ST S AND GENERAL PRACTITIONERS N O VA SC O T I A, 1971 - 78 1 / u, o 75 ~ 60- an - O 45- o - Hi ~ - 304 r~~ All ~ ~ / Lo An TOTAL / / - __ 15- - v- , _ 1971 1972 1973 1974 1975 GE N E RA L PRACTITION / / / f / / ,, ———~ P S YC H I A TR I S T S Fiscal Years Ending March 31 l 1976 1977 1978

191 I'--. any form of treatment for mental illness, behavioural maladaptions and/or other problems that are assumed to be of an emotional nature in which a physician deliberately established a professional relationship with a patient for the purposes of removing, modifying or retarding existing symptoms, of attenuating or reversing disturbed patterns of behaviour and of promoting positive personality growth and development. Accordingly, a psycho- therapeutic procedure may be charged for if one- half hour, or major part thereof, has been spent in such treatment of the patient." The total number of insured medical services increased 185% from 39 per 1,000 in 1971 to 111 per 1,000 in 1978. In relation to the total expenditures for medical services, the costs for psychotherapy increased from 1 percent to 3.5 percent. In 1974, the services provided by general practitioners increased sharply in association with a change in the fee schedule. In 1974 the fees for counselling and psychotherapy were increased 108 percent for general practitioners and 59 percent for psychiatrists, so that the general practitioner charges for psychotherapy were the same as those of psychiatrists. 25/ (In later years the fee schedules were again changed so that there was a 3;3% differential between psychia- trists and general practitioners.) In 1978, approximately 3.5% of the insured population received medical psychotherapy. General practitioners provided 64 psycho- therapy services per 1,000; a mean of 1.9 services per patient seen. (If one patient were seen by three different practitioners for psycho- therapy, the services would be cumulated for the one patient.) Pri- vate psychiatrists provided 45 services per 1,000; a mean of 5.9 services per patient during the year. In addition to the services from private psychiatrists the Nova Scotia mental health clinics provided another 38 psychiatrists' services per 1,000 population, a mean of 4.2 services per patient during the year. How many general practitioners submit claims for office psycho- therapy? Among the 500 general practitioners who earned more than 25,000 dollars from the insurance program during 1977, 30% did not submit any claims for psychotherapy. One-fourth (26%) of the practi- tioners submitted claims for less than $500 of psychotherapy during the year. The majority of psychotherapy and counselling in general practice was provided by a minority of practitioners. Of the 500 general practitioners, 38 physicians claimed 60 percent of the costs for general practitioner psychotherapy.

192 CAUTIONS AND CONCERNS This section describes some concerns these utilization data raise in three areas -- epidemiology; quality assurance/cost contain- ment; and clinical care. These three areas are interrelated since they all involve the questions: - Who needs treatment? (Person) - What is the level of care needed? 19/ (Place) - When is a lesser level of care appropriate? (Time) Epidemiological Concerns It is apparent from the literature that physicians vary widely in their recognition of mental disorders; that one condition may be given different names; and that different conditions (mental disorder, emotionality, problems of living, distress, etc.) may be called the same thing. Psychiatric epidemiologists need better methods for iden- tifying and categorizing the kinds of mental "disorders" seen by general practitioners. Shepherd's monograph shows the marked differences in diagnostic composition of cases, among physicians grouped by rate of recognition; the differences between physicians were less apparent in the psychoses and more prominent in the psychosomatic and "other" disorders. 47/ What are the relations between symptoms, syndromes, diagnoses? How can we better define the course and outcome of "mental disorders" with similar symptoms seen in different settings? 52/ Although symp- toms may be similar, the course and outcome of alcohol-related problems identified in community surveys differs from that seen in alcohol treat- ment programs. 6/ The course and outcome of physical disorders seen in different settings also varies. Motulsky has described the different outcomes and symptoms among patients with hereditary spherocytosis and mitral-valve prolapse who are seen in medical practice in comparison to those found in community surveys. 34/ We can no longer assume that persons found in population surveys with symptoms of specific mental disorders will have the same course, complications and outcomes as those mental disorders coming for treatment. We also need better methods to define the stage and severity of mental disorders. There has been relatively little progress in defining "stage" because of our limited understanding of the natural history of the various mental disorders. "Severity" is highly sub- jective as yet; 65% of patients seen by private psychiatrists were judged as "serious" in comparison to 19% of patients seen by all specialists. 35/

193 After we have reduced the variability between practitioners (and even epidemiologists) in diagnosis, after we have better infor- mation on course and outcome, after we have better methods for stating complications and severity we can begin to apply Gonnella's staging concept to specific mental disorders, to define different stages on the basis of complications, severity of problems, and potential out- come. _ / Finally, we can begin to establish "Diagnosis Related Groups," to bring together patients with diverse diagnoses who re- quire common treatment services and should have similar outcomes. 49/ Concerns for Quality Assurance and Cost Containment The applications of Quality Assurance and Cost Containment methods to psychiatry do not yet compare with their applications to other medical specialities. 43/ How can we justify the cost and assess the quality of care for the 2-4% of patients who use 30- 40% of mental health services? Resource absorption by a small proportion of cases is the concern of psychiatrists as well as economists, as shown by Guideline Number 1 of the Declaration of Hawaii: "... the psychiatrist shall serve the best interests of the patient and be also concerned for the common good and a just allocation of Health Resources." 53/ To develop effective mechanisms for Quality Assurance and Cost Containment, we should bear in mind the following: 42/ 1) A diagnosis of a mental disorder does not, by itself, justify ambulatory treatment. . A "diagnosis" does not denote a plan of clinical action, nor does it define the person as needing medical care. 26/ By itself, a "diagnosis" is not a disguised prescription for clinical intervention since the usual "diagnosis" does not describe the stage or severity of the disorder, which type of treatment is needed, its duration or frequency, the type of provider or level of care. The need for treatment must be assessed from the nature and severity of the impair- ment, limitation of function and the potential effectiveness of the treatment proposed. 40/ The American Psychological Association 4/ recognizes that a review model which begins with diagnostic categories is disastrous . . . "Instead of starting with diagnosis, one must focus on the objectives of service by authorized providers and the methods used to achieve those objectives in an optimal fashion. Starting from this strategy, empirical norms and other criteria can be developed by peer review operations and from standardized automated, data-processing procedures."

194 2) Mental disorders must be distinguished from psychological and other emotional responses. What sorts of disorders involve "feeling sad for at least two weeks?" Who should provide what sort of care for such conditions for how long? It is as yet difficult to differentiate among: a) everyday personal reactions to everyday problems; temporary and appropriate adaptations to stress; b) problems of "emotionality," personality traits, culturally determined behavior; subjective perceptions of "suffering"; c) isolated, psychological and behavioral symptoms and signs; d) syndromes for which there is good clinical consensus on their nature, severity; and considerable agreement on the type and amount of therapeutic intervention needed, 28/ and e) conditions, difficulties or syndromes not listed above. Clinicians are often unable to discern between life crisis and mental disorder. 23/ Of persons with psychiatric diagnoses seen by medical practitioners the majority are suffering from relatively minor ailments which include "problems of living," as well as "medical problems." 10/ We need reliable methods for differen- tiating these various disorders, situations, behaviors and reactions. 3) The objectives and goals of treatment must be specified. The therapeutic objectives of psychiatric treatment must be specified in advance in order to assess the results of treatment, or the need for modifying the treatment plan. Practitioners vary in their goals for similar patients. 32/ We must be able to identify whether problems for which care is sought by the patient are congruent with those being approached by the practitioners 48/ Continuation of therapy must be based on whether predetermined clinical goals have been attained. We do not yet have reliable methods for assessing progress in attaining treatment goals. 4) The clinical setting must be considered. Which types of ambulatory program can provide the appropriate level of care required by a patient? Within the program, what is the appropriate level of provider for the treatment needed by the patient? Out-patient clinics present a range of services not found in private offices. Solo-practitioners provide a continuity and comprehensiveness of care not found in some clinics.

195 5) Professionally pre-determined screening criteria are essential Screening criteria are not meant to be comprehensive or encyclo- paedic, but are used to focus attention on circumstances for which problems in justifying admission, continued care, and level of care are most likely, and to identify those cases for whom peer review is indicated. These screening criteria should be: 50/ a) understandable to non-practitioners and require no clinical interpretation; b) unambitious and reliable; c) appropriate for the content of the clinical record; and d) measureable, if at all possible. 6) Cost-containment must use mechanisms which complement '~claims ~ . , review." Providers must go beyond the claims review perspective of focussing on contractual definition of "eligibility," "provider," "service," "disorder," etc. We need mechanisms which enable profes- sionals to advise the third party payor whether the care of an individual patient, as well as the statistical pattern of care is appropriate, effective, of the "right" amount and of high quality. A cost-containment system for ambulatory care must allow for: a) conceptual clarification between claims review and other types of review; b) non-practitioner use of predetermined criteria to select cases for peer review; c) peer review focussed on specific variations from pre- determined criteria; d) use of reliable nomenclature for differentiating procedures, treatment goals, and patients' psychiatric problems; e) mechanisms for assessing the level of care needed by the patient in terms of type of provider and type of clinical setting for that provider; and f) experience-based norms of the course of treatment for defining optimum points for focussed review.

196 Clinical Concerns This section describes two concerns for clinical care -- the type and level of care and the "redistributive phenomenon." There are wide variations in types of treatment and types of provider. 46/ Treatment for mental disorders includes pharmacologi- cal or specialized psychological approaches whose nature and effective- ness are well defined and for which the level of requisite training and experience can be specified. 29/ On the other hand, there is suspicion and increasing evidence that personnel varying widely in discipline, training and ideology can perform counselling and psycho- therapy with equivalent results. 21/, 51/ It is necessary to classify treatment procedures not only by their duration and frequency, but the necessary level of training and experience of the provider. 18/, 36/ It is not appropriate to have a few procedure codes and reimbursement rates to cover the wide range of psychotherapies and all providers of psychotherapy. Some mental disorders can be cared for by non-psychiatrists or non- physicians. We need to define appropriate levels of care by dis- cipline, training and experience of provider and to distinguish what procedures should be provided by which types of non-psychiatrists, as well as psychiatrists. 51/ The problem is illustrated in the Request for Research Applica- tions (RFA) issued by the National Cancer Institute on February 15, 1979. Since counselling is considered to be potentially effective for cancer patients, projects are requested to show the relation- ship between specific emotional or psychological problems and speci- fic counselling techniques. (Figure IX). The RFA asked for clear definition of the rationale for matching the problem to the counselling technique and, in addition, enumeration of the special skills, qualifi- cations or experience of the staff required to provide a specific counselling technique for specific emotional psychological problems. A second problem for clinical care is the redistributive phenomenon. In the development of prepaid health insurance pro- grams, economists are concerned with the redistribution of medi- cal resources from higher to lower income populations. 37/ It is possible however, that including psychiatry in health insurance pro- grams may represent a subsidy to the rich from the poor. 2/,9/,27/,30/ There is some pre-medicare evidence that "upper-class" persons were more likely than "lower-class" persons to remain in prepaid psychiatric treatment. One report from British Columbia found that among married women aged 15-44, covered by the same insurance plan, the 33 residents of upper class areas used twice the amount of psychia- trists' services as did the 51 residents of lower social class areas. 41/

197 FIGURE IX S PEC I F I C COUNSELL I NG TECHN IQUE 1 2 3 4 5 6 A SPECIFIC EMOTIONAL OR B PSYCHOLOG I CAL PROBLEMS - 1 1 - RATI ONALE FOR MATCH I NG ENUMERATE S PEC I AL S K I LLS, QU A L I F I C AT I ON S. O R EX P E R I EN C E N.l.H., RFA 15 Feb.1979 Another pre-medicare report 22/ compared the experience of three socio-economic groups who were eligible for similar medical benefits for psychiatric care from general practitioners or psychiatrists during a six month period in Ontario: a) persons on welfare "socially assisted" b) low-income subscribers whose premium was subsidized ("low-income") c) "independent" - or non-subsidized subscribers These three income-groups were continuously enrolled during the The proportion diagnosed as having a mental disorder study period. during the six month period ranged from 4.0% to 5.3% in the various income groups, with socially assisted enrollees having a rate similar to that for non-subsidized ("independent") enrollees. The distribution of the three groups in general practice was similar to their distribution in the plan; however, socially assisted patients had fewer services than "independent" patients. In psychia- trists' practices, independent insured patients were over-represented -- 50% of psychiatrists' insured patients in comparison to 25% of the insured membership. Socially assisted patients had fewer services from psychiatrists than "independent" insured patients. (Table 4~.

198 TABLE 4. ONTARIO MEDICAL SERVICES INSURANCE PLAN SUBSCRIBERS CONTINUOUS ENROLLMENT Socially Assisted Low Income Independent Population Size 316 543 312 (thousands) Percent 28% 46% 26% Distribution Persons with diagnosis of mental disorder Number (thousands) 17 22 16 Percentage of subscribers 5.3% 4.0% 5.1% Single Service Claims For Mental Disorders Distribution of patients among general practitioners 34% 42% 24% psychiatrists 19% 31% 50% Average number of services by: - general practitioners 3 4 4.5 - psychiatrists 5.5 6.5 8 Multi Service Claims for Mental Disorders (single claims~for number of services for a patient during a single month) Distribution of patients among: - general practitioners 30% 44% 26% - psychiatrists 19% 31% 50% Average number of services by: - general practitioners 15.5 23.5 26.5 - psychiatrists 12 15 18 From Hanly

199 A long-term study of the redistributive impact of national health insurance on the utilization of physicians' services is being conducted in Nova Scotia by Murray G. Brown in a project funded by the National Center for Health Services Research. This project is comparing the pre- and post~edicare utilization of two groups ~ ~ ~~~ . ~ . . . , ~ . . a group of about Zl,OOO pUDllCly Insured welfare recipients and another group of 145,000 pri- vately insured non-poor persons. The benefits were similar for both groups. Figure X shows the pre medicare utilization of psychiatric services for each group: publicly insured welfare recipients received about five psychiatric services per 1,000; privately insured non-poor received 50 services per 1,000. The universal insurance program~started with a rate of utilization of psychiatric services less than the privately insured nonpoor were getting four years earlier. Did the poor continue to receive smaller numbers of services, and did the privately insured persons get a larger number of services? A sub-project is now assess- ing the post~edicare use of psychiatric services by the two groups. FIGURE X 70 60 REDISTRIBUTIVE IMPACT OF NHI ON D I SADVANTAGED POOR ( BROWN-DALHOUS IE) SUB STUDY ON PSYCHIATRIC SERVICES to on — 50- - cat , 40- A: 30 - 20 10- P R I VA TE LY ? NON-POOR N-145, 000} P U B L I C LY I N S U RE D WELFARE ( N - 21, 300) .~ - . ~ ? o . . . 1967 1968 1970 1971 1972 1973

200 Agenda for Future Research We suggest that the topic of this conference, Mental Health Services in Primary Care Settings, be the focus for state-of-the- art conferences on each of the three concerns outlined above: (a) The epidemiology of mental morbidity in primary practice; (b) Quality assurance and cost containment methods for mental health services in primary practice; and (c) Health systems research on mental health services in primary practice. These proposed state-of-the-art conferences are not meant to produce simplistic answers but are seen as opportunities for defining crucial questions for which our current methodology is adequate, and identifying the questions for which we need better methodology (including data bases). In focussing on the epidemiology of mental morbidity in primary practice we need to consider the current approaches to case identifica- - tion; describe the kinds of "problems" they are identifying; and discuss the theoretical rationale of these approaches and their relevance to current patterns of practice and concepts of "medical need." In reviewing our understanding of the natural history of these conditions we should consider the future development of measures of stage, severity and complications. Are we now ready for collaborative or international studies (with NIMH's Diagnostic Interview Schedule and the Research Diagnostic Criteria) or are we still at the stage of the Diagnostic Exercises (WHO; the Anglo-American bilateral Project) where video- taped interviews were used to elicit and assess diagnostic usage and identify the symptoms, signs, and historical features associated with the use of a particular diagnosis? What factors affect whether a primary practitioner defines a patient as having a mental disorder; what determines the need for treatment and what are the practitioner's expectations of course and outcome? What circumstances affect whether the mental symptoms, social context and past history are ignored, diagnosed as a mental disorder, or given an ICD-9 V-code? When are psychosocial circumstances, or "problems which influence the person's health status," considered a current illness and when will they be given a V-code within the ICD-9 classification? How have current procedures for quality assurance and cost cortainmer.t been applied to ambulatory mental health care? Do they work? What are the conceptual blocks or missing resources which have made psychiatry falter in applying these methods? How can quality assurance and cost containment approaches be incorporated in teaching and training programs for primary practitioners? How can we

201 develop definitions and procedures for assessing medical necessity, level of care, and discharge planning for mental morbidity in primary practice? Can attention be re-routed from diagnosis to review problems of management? The third topic is that of ~ on mental health services. A recent ADAMHA report _/ identified health systems research as suffering most from diffuseness of purpose and structure and urged immediate, intense attention on developing better methods of identifying populations in need of service, defining the extent of underservice and acquiring data on utilization and cost across many service settings and financing arrangements. The report urged that immediate priority be given to studies on the utilization and cost of services under current health insurance and projected National Health Insurance plans. Although there is much overlap in content and approach between epidemiology, quality assurance and cost containment, and health systems research, by and large, there has been sparse collaboration by the few researchers in each of these fields. This third conference should consider the reasons for the insularity and seek short-term solutions for enhancing collabora- tive efforts. An initial focus for such collaborative efforts might be the redistributive effect in prepaid mental health services - is Matthew 25:29 applicable? CONCLUSIONS This paper has described the Canadian experience with National Health Insurance prepayment of psychotherapy by primary practitioners, and the implications of these data for epidemiology, quality assurance and cost containment efforts, and for clinical care. The epidemiologic dimensions of the problem are clearly visible. However, there is far less information on the clinical and economic consequences of prepaid care, and the extent to which utilization can be predicted, demand satisfied, and needs fulfilled. The 1964 Royal Commission project on the extent and results of psychiatric care in Canada 8/ correctly predicted mental hospital use in 1971, but its high-level projection of psychotherapy (0.2 hours per person per year) has now been exceeded in many areas and the divergence is continuing to widen. Both improved understanding and better data are needed to project the needs for medical psychotherapy. We need to bring together, in a more systematic manner, the skills and resources of epidemiologists and health services researchers in primary practice, as well as in the specialized mental health services. In this way we can bring epidemiology and health services research into the community and clinical settings - out of the closet of research journals.

202 REFERENCES 1. ADAMHA Workgroup on Epidemiology, Health Services Research and Statistics (Data Systems), Epidemiology, Health Systems Research Statistics/Data Systems. U.S. Department of Health, Education and Welfare; Alcohol, Drug Abuse and Mental Health Administration, 1978. 2. G.W. Albee, Does Including Psychotherapy in Health Insurance Represent a Subsidy to the Rich from the Poor? American Psychologist, 32:719-721, 1977. 3. American Psychological Association, Health Benefits Task Force, Committee on Health Insurance. The Integration of Mental Health - Services in a Comprehensive Nat~onal Health Plan. Unpublished document, ed. R. Bent and A.E. Shapiro, Washington, D.C., 1974, cited in Bent. 5. R.J. Bent, Impact of Peer Review on Future Health Practice. Chapter 13 in The Professional Psychologist Today, H. Dorken and Associates. 6. W.B. Clark and D. Cahalan, Changes in Problem Drinking Over a Four-Year Span, Addictive Behaviors, 1: 251-259, 1976. 7. Canada, Information Canada, Canada Year Book, 1973. 8. Canada, Royal Commission on Health Services Report, Volume I, Ottawa: Queen's Printer, 1964. 9. E. Crowell, Redistributive Aspects of Psychotherapy's Inclusion in National Health Insurance. A Summary. American Psychologist, 32: 731-737, 1977. 10. C. d'Arcy, Patterns in the Delivery of Psychiatric Care in Saskatchewan, 1971-1972, Service Interface Study Interim Report. Applied Research Unit, Psychiatric Research Division, University Hospital, Saskatoon, Saskatchewan, August 1976. C. d'Arcy, Patterns in the Delivery of Psychiatric Care in Saskatchewan, 1971-1972: An Overview of Service Sectors and Patient Volumes. Can. Psychiatr. Assoc. J., 21:91-100, 1976 12. C. d'Arcy, (II) Types of Contacts and Same Patient Career Characteristics, Can. Psychiatr. Assoc. J., 22:31-36, 1977.

203 13. C. d'Arcy, (III) Patient Socio-demographic and Medical Characteristics, Can. Psychiatr. Assoc. J., 22:215-223, 1977. 14. C. d'Arcy and J.A. Schmitz, Sex Differences in the Utilization of Health Services for Psychiatric Problems in Saskatchewan. Can. J. PsYchiatrY. 24:19-27. L979. 15. C. d'Arcy, M. Vanden Ham, S.D. Goldie, The Development of a Comprehensive Psychiatric Service Utilization Data File, Can. J. Pub. Health, 67:237-248, 1976. 16. C.A.R. Dennis, R. Drape and D. Matz, Medicare Data: Its Use in Defining the Effects of the Environment on Health, Ottawa, National Research Council (NRCC15387), 1977. 17. H. Darken and Associates, The Professional Psychologist Today, San . . Francisco: Jossey-Bass, Inc., 1976. 18. L. Eisenberg, The Future of Psychiatry, Lancet, ii, 1371-1375, 1975. 19. G.A. Goldberg and D.C. Holloway, Emphasizing "Level of Care" over "Length of Stay" in Hospital Utilization Review, Medical Care, 13:474-485, 1975. 20. J.S. Gonnella and M.J. Goran, Quality of Patient Care - A Measure- ment of Change: the Staging Concept, Medical Care, 13:467-473, 1975. 21. M. Greenblatt, Psychiatry: The Battered Child of Medicine, New England Journal of Medicine, 292:246-250, 1975. . 22. C. Hanly, Mental Health in Ontario. A Study for the Committee on the Healing Arts, Toronto: Queen's Printer, 1970. 23. M.R. Hanson and others, Five County Cost-Effectiveness Study, Unpublished document prepared for the California Department of Health, Health Planning and Intergovernmental Relations, Sacramento, California, October 11, 1974, cited in Darken. 24. G. H. Hatcher, Canadian Approaches to Health Policy Decisions National Health Insurance, American Journal of Public Health, 68: 881-889, 1978. 25. V.A. Hicks, Changes in Physicians Patterns of Practice Associated with Changes in Relative Prices, Unpublished M.A. Thesis, Dalhousie University, 1977. 26. R.E. Kendell, The Role of Diagnosis in Psychiatry, Oxford: Blackwell Scientific Publications, 1975.

204 27. E. Kennedy in Group for the Advancement of Psychiatry, Committee on Therapy; Psychotherapy and its Financial Feasibility within the National Health Care System, Volume X, No. 100, Group for the Advancement of Psychiatry, 1978. 28. A. Lewis, The State of Psychiatry, Essays and Addresses. New York: Science House, Inc., 1967, p. 187. 29. A.M. Ludwig and E. Othmer, The Medical Basis of Psychiatry, ATTI. J. Psychiatry, 134:1087-1097, 1977. 30. A. J. McSweeny, Including Psychotherapy in National Health Insurance. Insurance Guidelines and Other Proposed Solutions. American Psychologist, 32: 722-730, 1977. 31. Mathematica Policy Research (Princeton, N.J.) A Study of the Responses of Canadian Physicians to the Introduction of Universal Medical Care Insurance: The First Five Years in Quebec. Project Report 78-09. 32. R. Mayou' Psychiatric Decision Making, British Journal of Psychiatry 130: 374-376, 1977. 33. Medical Post, January 2, 1979. 34. A.G. Motulsky, Biased Ascertainment and the Natural History of Diseases, New England Journal of Medicine, 298:1196-1197, 1978. 35. National Center for Health Statistics, Advance Data from Vital & Health Statistics, Office Visits to Psychiatrists: National Ambulatory Medical Care Survey, United States, 1975-76, Number 38, August 25, 1978. 36. S . H. Nelson, Current Issues in National Insurance for Mental Health Services, Am. J. Psychiatry, 133: 761-764, 1976. 37. J.P. Newhouse, C.E. Phelps, W.B. Schwartz, Policy Options and the Impact of National Health Insurance, New England Journal of Medicine, 290:1345-59, 1974. 38. New York County Health Services Review Organization, Screening Criteria for Psychiatric In-Patient Services of General Hospitals, New York, 1976. 39. A. Richman, Psychiatric Care and Prepaid Medical Insurance Plans, Chapters 16-19, in Psychiatric Care in Canada: Extent and Results, Royal Commission on Health Services, Queens Printer, 1966. 40. A. Richman, Assessing the Need for Psychiatric Care, Can. Psych. Ass. Ji., 11:179-188, 1966.

205 41. A. Richman, Psychotherapy, Social Class, and Medical Insurance Plans, Medical Care, 4:15-16, 1966. 42. A. Richman, Cost Containment and Quality Assurance Requirements for Third Party Coverage for Ambulatory Psychiatric Care, Presented at the Annual Meeting of the American Public Health Association, Washington, D.C., 1977. 43. A. Richman, The Application of Psychiatry of Methods for Reviewing the Quality of Patient Care, Presented at the Annual Meeting of the Canadian Psychiatric Association, Halifax, Nova Scotia, October, 1978. 44. A. Richman and H. Pinsker, Utilization Review of Psychiatric In- patient Care, Am. J. Psychiat. 130:900-903, Aug. 1973. 45. R.A.B. Robie, Psychiatry and General Practice in North America, Canada s Mental Health, Supplement No. 35, March 1963. 46. S.S. Sharfstein and H. J. Magnas, Insuring Intensive Psychotherapy, Am. J. Psychiat. 132:1252-1256, 1975. 47. M. Shepherd, B. Cooper, A.C. Brown and G. Kalton, Psychiatric Illness in General Practice, London: Oxford University Press, 1962. 48. R. M. Swanson, N.C. Wilson, J.L. Mumpower and R.H. Ellis, Dimensions of Mental Health Treatment Goals, Dupl. Fort Logan Mental Health Center, Colorado, n.d. 49. J.D. Thompson, AUTOGRP Model: Applications to Yale-New Haven Hospital Data Set, Milbank Memorial Fund Quarterly, Health & Society, 56: 268-273, 1978. 50. U. S. Dept. Health Education and Welfare. PSRO Program Manual. 1974, Washington, D.C. 1974. J.S. Werry, Psychotherapy - a Medical Procedure? Can. Psychiatric Assn. J., 10:278-282, 1965. - 52. World Health Organization, WHO Collaborative Project on Det'~rmi- nants of Outcome of Severe Mental Disorders, 1977-1979, MHNJ 78, 14, Rev. 2, August 1978. 53. World Psychiatric Association, The Declaration of Hawaii, Psychiatric News, Page 23, October 7, 1977.

207 MENTAL HEALTH AS AN INTEGRANT OF PRIMARY CARE Michael Shepherd, M.D. Professor of Epidemiological Psychiatry Institute of Psychiatry University of London It is now a little over 3 years ago since I visited the National Institute of Mental Health in 1975 as a WHO consultant to assess and report on the status of mental health care systems in the USA, with particular reference to primary care. To implement the various recommendations of my report I urged at that time the close involve- ment of the Division of Biometry and Epidemiology which, it seemed to me, "would be uniquely placed to provide basic data and a factual underpinning for any proposals to initiate studies as required and to help integrate the effort which may be required in the next phase of the mental health movement." The organization and content of this meeting shows clearly how much progress has already been made by the Division under the leadership first of Dr. Kramer and then of Dr. Regier and how it has already begun to come to grips with many of the issues raised by our own work on the extramural dimensions of mental disorder. From the beginning this was based on the primary care physicians or general practitioners who occupy a central posi- tion in the health service structure which differs radically from that prevailing in the two other models encountered in developed countries. 1/ (Figure 1) In the conditions of the National Health Service the British general practitioner is the physician of first contact, the profes- sional figure who is the gate-keeper to all medical facilities. Furthermore, as he keeps routine records of his consultations, it seemed reasonable to try and assess the amount and the nature of mental disorders with which he is concerned. Accordingly, we initiated a series of studies with this objective in mind. In quantitative terms the most striking result was that a large segment of morbidity, amounting to about one-seventh of all consultations, was attributable or closely related to mental ill-health. 2/ A number of workers have since confirmed the substance of our findings but at the time most of the few people who appreciated their significance appeared to be individual general practitioners with an interest in psychiatry. Here, for example, is an illustration pre- pared by a well-known British general practitioner, working indepen- dently, who analyzed the results of his own activities in the early 1960s. 3/ (Figure 2)

208 FIGURE 1 USSR UR USA ~ I~. family | FAMILY unit l . ~ ~ irst-co'~tact care Specialist care for the ambulatory Hospital in-patient care P()I.N (:I INI(: _ [A.MII.\' L 1 C,P 1 1 I I en SPF,(:IALOIDS | . / 7TV\ 1 1 1 t\~1 ~ H _ 1 1.11 4 comparison of t/:ciow of medical care in t/~ USSR, L'~ and USA 1.

209 FIGURE 2 BREAKDOWN OF PSYCHIATRIC ILLNESS IN AUTHOR'S PRACTICE FOR PATIENTS CONSULTING IN A COMPLETE YEAR Atcoh31~r addiction Psychopathy Compensation neurosis- Schizophrenia- - Amentia - . 1 IATRIC ~~N Chronic Anxiety and · ~ | I | I | Acu~ Hystero:~: il il tEDndo~=~1~'~ he Anxiety States Am;< `~/4TR~C \~0~/

210 It is less profitable to dwell on his finer diagnostic categoriza- tions than to underline, first, the relative proportions of acute and chronic illness and of major and minor disease. Secondly, the domi- nance of mood disorders, which were his particular concern, stands out in this highly schematic picture of depressive disorders. (Figure 3) While the details may be inaccurate to the point of caricature, the findings nonetheless serve to raise several relevant questions, some relating to the sphere of administration, and others to clinical practice and research. From an administrative standpoint we may look at the figure in terms of cost, of manpower requirements or of educational objectives, all of them necessary prerequisities to rational planning. It is, however, essential for planners to declare their interest if they are not to go astray. A strikingly relevant example of what may happen otherwise was provided in the United Kingdom during the last war when the Ministry of Health sponsored a survey of psychiatric facilities existing in the country. This inquiry was conducted by C.P. Blacker, who used the findings to put forward quantitative pro- posals for the reorganization of the whole of the national mental health services, including a detailed model plan for a population of a million persons. 4/ Much of the report is eminently sensible and anticipates later provisions of the National Health Service. In respect of the general practitioner service, however, Blacker's assessment led him to recognize the problems but not the angle from which he was observing them: "The so-called psychosomatic disability," he wrote, "has been much discussed of late, here and in America; it has even been suggested that as much as a third of all sickness has psychiatric features, the term psychiatric being used to include psychosomatic illness. We can picture to ourselves the effects of clinical attendances if practitioners as a whole came to believe that a third of their patients could be benefited by the attention of psychiatrists. The community contains, as it has always contained, a reservoir of psychosomatic and psychopathic cases; their descent in vast multitudes upon the psychia- tric clinics of this country might be caused by nothing more than an alteration of standpoint among general practitioners." This curious comment reflects an attitude which appears again in his vision of the future relationships between the lofty, seignorial psychiatric specialist and the lowly general practitioner: "It is therefore suggested that when the general level of psychiatric knowledge is raised throughout the medical profession by improved teaching methods - or even

211 FIGURE 3 O 12 per 1000 Suicides 0 1 3 per 1000~ ~ Compulsory admissions 077perlOOO ~ 1101011111 - - 1144~111I' 1-90 perlOOO Level of recognition _ Informal admissions ,_ Patients seen by psych istrist Patients seen by ~ ? ~ ~ ~ ~ ~ ~ ? ~ ~ ~ ~ ~ ~ ~ ~ ~ ! ~ 'general practitioner ellll~ Scale diagram of the ' iceberg of depression'. One small square odor persons per logo of the home population. Source: Watts, C.A.H. (1966~: Depressive Disorders in the Community. Briston: John Wright.

212 before this happy time is reached - it should be the aim of the clinic to send the patient back to his doctor, reporting improvement, at the earliest date reasonable, at the same time furnishing the practitioners with guidance as to how to handle the patient in future." I am particularly fond of the phrase, "reporting improvement," but the passage as a whole embodies a view of the general practi- tioner as the doctor who, in the phrase of the late Lord Moran, had fallen off the 'specialist ladder." That view is still echoed today. Setting out his views on the organization of psychiatric services, for example, a prominent British social psychiatrist not long ago estimated that three psychiatrists are needed for a population of 60,000 people, in order to care for about lOOO patients during the course of 1 year. He then went on to say: ''There will also be about 24 family doctors in the area. These doctors, however, cannot give psychiatrists much help for in our Health Service family doctors are already seeing the bulk of the patients with socioeconomic problems." _/ A similar perspective has been adopted all too often by clinical psychiatrists, though they peep at the matter through the practi- tioner's key-hole rather than the planners'. Thus if, in Figure 3, we summate the first four categories, the proportion of depressed patients who come to the attention of a psychiatrist is no more than 2.92/1000 of the general population and no more than 1.8% of all depressed people. In consequence, psychiatrists in their clinical practice are familiar with only a very small band of the depressive spectrum and one, furthermore, which differs in respect of presenting features and in severity from the larger part. None- theless, at a recent international conference on depression a well- known European psychiatrist made the following comment: ". . . in possibly as many as 40% to 50% of all patients consulting a general practitioner for any reason whatsoever no organic causes for their symptoms can be found . . . This raises the question as to whether all these patients should be regarded as psychiatric cases and there- fore treated by a psychiatrist. The answer is probably 'No'." 6/ It is worth pondering on the two principal reasons for this seemingly willful disregard of the evidence by mental health profes- sionals, as demonstrated by the views I have cited. The first of these reasons, I would suggest, has to do with an understandable reluctance to relinquish what may be termed the psycho-centric per- spective, whether this be identified with psychotherapy, psychiatric education, administration or clinical expertise. As such it exempli- fies an outlook which is common enough in science as well as other forms of less rational human activity. In all essentials this was described, and even named, as early as 1440 by Nicholas of Cusa who, 20 years before Copernicus' observations, attacked the assumptions of medieval scholarship for the geocentricity of its outlook on the physical universe in a famous tract which he entitled, 'Learned

213 Ignorance.' Though the present example is rather less momentous, there has been a medieval flavor to the pronouncements of many psychiatrists, not only in their unwillingness to modify their own perspective but also - and this is the second of my two reasons - in their attempt to impose their own conceptual apparatus on the material under study, regardless of its goodness of fit. The consequences of this variety of learned ignorance reveal them- selves most clearly through systems of classification which, au fond, reflect no more than underlying systems of thought. When we started work with general practitioners our first inclination was to employ or adopt the standard International Classification of Diseases. After a very a brief experience of the clinical problems encountered, however, it became apparent that neither the ICD, then in its 7th edition, nor any available alternatives did justice to the situation. Accordingly, we were compelled to construct a more relevant system of our own, designed to meet the needs of the 8p's by distinguishing between 'formal' psychiatric illness and what we called 'psychiatric-associated' disorders, thereby anticipat- ing the multiaxial systems which have since been widely compassed to do justice to the health-menta1 health interface (Figure 4~. Using this schema, with all its manifest imperfections, we found that about one-third of recorded psychiatric morbidity had to be classified in this way (Figure 5~. Attempting to identify in more detail the content of this heterogeneous category, we found that the 'associated' factors included a wide range of physical disorders on the one hand and of social pathology on the other (Figure 6~. - FIGURE 4 CLASSIFICATION OF pSYCHIATRIC CONDITIONS FORMAL PSYCHIATRIC ILLNESSES 1. PSYCHOSIS 2. MENTAL DEFICIENCY 3. DEMENTIA 4. NEUROSIS schizophrenia, manic depressive psychosis, organic psychosis. Or marked subnormal intelligence. deterioration of mental powers in excess of normal aging process. anxiety state; depressive, hysterical, phobic or asthenic reactions; others. 5. PERSONALITY-DISORDER PHYSICAL ILLNESSES, OR PHYSICAL SYMPTOMS WITH PSYCHOLOGICAL COMPONENT 6. PHYSICAL ILLNESS where psychological mechanisms have, in 7. PHYSICAL SYMPTOMS your opinion, been important in the development of the condition. which have, in your opinion, been elaborated or prolonged for psycho- loeical reasons. 8. PHYSICAL ILLNESSES 9. PHYSICAL SYMPTOMS OTHER PSYCHOLOGICAL OR SOCIAL PROBLEMS . X PSYCHOLOGICAL OR SOCIAL PROBLEMS - please describe in full.

214 FIGURE 5 PATIENT CONSULTING RATES PER 1,000 AT RISK FOR PSYCHIATRIC MORBIDITY, BY SEX AND DIAGNOSTIC GROUP DIAGNOSTIC GRO[nP MALE FEMALE BOTH SEXES Psychoses ... Mental subnormality ... Dementia ... Neuroses ... Personality disorder Formal psychiatric illness(~) ... Psychosomatic conditions Organic illness with psychiatric overlay Psychosocial problems ... · ~ ~ · ~ - 24~5 1361 466 _ 2~7 8-6 1~6 2~9 162 1~6 5567 ~ 116~6 762 4~0 67~2 131~9 .. 34~5 16~6 10~0 5~9 2~3 1 e4 88.S 5~5 102 1 29~9 15~0 7~5 Psycl~iatric-associated conditions(l) Total psychiatric morbidity(l) ... · · ~ 3896 97.9 57 2 175 0 48-6 139~4 Number of patients at risk 6,783 7~914 14~697 . (lj These totals cannot be obtained by adding the rates for the relevant diagnostic groups because while a patient may be included in more than one diagnostic group, he will be included only once in the total.

215 FIGU~ 6 MEDICAL, SOCIlL OTHER FACTORS NONE NENTI0#ED r _a [;.-.~ ~ 40 30 20 10 ~ ) PERCENTAGE OF PSYCHIATRIC PATI E NTS DlffEtENCE BETWEEN SEXIS X~2on I d.~. ASSOC I AT E D 4 . 5 4 * PHYSICAL ILLNESS SENILITY CONSTITUTION - ALCOHOLI SM MENOPAUSE ETC. PREGNANCY EARLY ENVIRONMENT ADOLESCENT STRESS S E X UAL PROBLE M S MARITAL PROBLEMS CH I LD MANAGEMENT DEPENDENT RELATIVES HOUSI NG ETC . BEREAVEMENT Wl DOWHOOD OCCUPATIONAL PROBLEMS OVERWORK ~ STUDY MALES ( N- 664 ) FEMALES (~- l38s) 0 02 NS 3 41 NS 8 06** 0 01 NS 3 63 NS 0 45 NS 7 74** 17 16** 14 03 $ $ 8 32 ** 15 58 * ~ 37 2 7 * * 0 01 NS 0 0I NS 8 73*,

216 The current interest being taken in the health-mental health inter- face may serve to justify a very brief summary of some of our subsequent work which, it is worth reiterating, grew out of our own data rather than from any theoretical preconceptions. We first attempted to examine the association with physical illness indirectly by identifying a con- trol group of cases reported as being free from any form of mental ill-health. These were compared with patients diagnosed as suffering from psychiatric disorders. Emotional disorder proved to be associated with a high demand for medical care; the patients attended more fre- quently, exhibiting higher rates of general morbidity and more categories of illness, especially chronic disease. It can, of course, be argued that these findings were largely manifestations of a high demand for medical care attributable to the patients' attitudes to health and that the patients were labelled as neurotic largely because of the frequency of their attendances and the multiplicity of their ailments. Independent data were therefore sought by estimating physical disease among groups scoring high and low on a screening questionnaire, a procedure which yielded similar findings. Evidence from several other studies pointed in the same direction, but as most of these findings were drawn from investigations directed at other objectives a large-scale independent inquiry was designed to determine directly whether individuals with psychiatric illness did or did not suffer from more physical illness than mentally healthy people. 7/ From a population undergoing a health-screening program subjects between the ages of 40 and 64 were randomly chosen and assessed in four stages: (1) by the completion of a self-administered questionnaire; (2) by a standardized psychiatric interview; (3) by physical screening-tests carried out by trained ancillary staff; and (4) by a physical examination by an independent physician. The subjects with psychiatric disorders were compared with a control group from the same population, matched for age, sex, marital status and social class. The results showed strong presumptive evidence of an association between physical and mental illness in this population, the links being most marked with subjects suffering from cardiovascular and respiratory disease. On the other side of the same coin a number of workers, including ourselves, have shown a number of discrete physical conditions to be associated significantly with mental ill-health and at least one large- scale American study has indicated that psychosocial factors must also be considered in this context. This is the work of Shaffer, et al, who investigated a population for disability benefits under the U.S. Social Security Administration's disability program. 8/ These workers made psychiatric assessments of such patients suffering from physical disorders and assessed the mental health of more than 1000 individuals, matched with 14,000 patients attending a medical clinic. The results demonstrated a marked difference between the two groups, furnishing an estimate of up to 44Z of individuals with moderate to severe psychoneuroses or personality disorders among the applicants for disability benefit.

217 Such findings serve to lead naturally from the physical to the environmental associations of psychiatric morbidity. Some of the links with social pathology have been clarified in recent years by studies concerned with hospitalized mental disorder and mental handi- cap in relation to its social consequences and social causation. The field has also been cultivated, however, by a host of social investi- gations of the general population which impinge on normal and abnormal mental states, often unwittingly, through the employments of concepts like 'social malaise' and the fashionable 'subjective social indicators, incorporating such feeling-states as dissatisfaction, ill-being, un- happiness and annoyance, all of them very close to what the psychiatrist would regard as mood-disorders. 9/ Unfortunately such notions, interesting as they are from the standpoint of theory, need to be operationalized if they are to be employed for empirical research. To our dismay we discovered more than 10 years ago that the necessary spade-work had not been carried out, and so we undertook a number of studies designed to remedy the deficiency. These, in brief, consisted first in constructing and standardizing a Social Schedule to be administered by trained investi- gators on community populations, focussing on the three broad areas of material conditions, social management or coping, and so-called satisfaction. _ / It then proved necessary to order and categorize the range of social problems encountered so as to achieve a management classification 11/ which lent itself to the study of various patient groups. With these tools we have been able to examine the social aspects of a variety of clinical conditions. The findings to date show that at the level of primary care social factors enter so closely into the matrix of what physicians call psychiatric disorders as to justify study both in their own right and on their role as potential pathways for intervention. Figure 7, for example, illustrates the findings on the health status of 300 consecutive patients referred by 8 general practitioners to their attached social workers in the course of their routine clinical practice. 12/ The ratings were made by a medical member of the research-team and demonstrate that the health of the population was generally poor; only seven percent of referrals were without a somatic or a psychiatric diagnosis and more than a quarter were sufferings from both mental and physical ill-health. A majority of cases with major psychiatric illness were in the senium, but the bulk of minor mental disorders were presented by married women with family problems and associated mood-disorders. It has also become clear that the health of those individuals whose primary contacts have been for designated social disorders calls for investigation, and in some of our more recent work we have been examining the health status of individuals who have been referred, or have referred themselves, directly to social agencies as a form of first contact. In one recently completed study, for t

218 FIGURE 7 ALAN - dims OF CLIMB as. No. Of CUENr5 us no, so to 0— _/ ~ ~— '=640D~ ~~ Sac lo= mu-' ~ Sac awl a~ PAL ANTI ~~ "tlSICAL ~ ~ or _ ~ PS~IU - IC a TAR PEAL ~ ~ ~~ FOR P - SK:AL ~ ~ He's n 11 ~ l 181111 11 11111 1 11 :::: llllllll · ·:: 1~11 1 1 11 · .:: 11111601 __ :::: 11~1,ll1 Source: Corney, R. and Briscoe, M. (1977~: Investigation into two different types of attachment schemes. octal Work Today, 9. example, information has been obtained about all individuals referred over a 3-month period to social workers in two settings -- a large health center run by a group of general practitioners and a local authority area team which is administered by the non-medical social services department. 13/ The information included medical data relating to physical and mental illness of handicap as well as the presenting social problems which led to referral. A substantially high proportion of clients turned out to be assessed as suffering from ill-health, physical or mental or both (Figure 8~. Whether they are termed clients or patients these people must evidently be classified in socio-medical or medico-social terms if justice is to be done to their status. The order of the words merely reflects the professional background of the observer. What, then, are the implications of such findings for the pro- vision of primary care services? In Britain the socio-medical approach to health has been officially recognized by the National Health Service Reorganization Act of 1973 which was followed by a Working Party Report entitled, significantly, "Social Work Support for the Health Service." 14/ In their report the Working Party paid particular attention to the de- velopment of social work in the context of general practice and entered a strong plea for experimentation in this field. Our own research, which long antedated the report, has been conducted very much in its spirit.

219 F INURE 8 SOCIAL WORKERS ' ASSESSMENTS OF CLIENTS ' FOR PROBLEMS Intake Attachment Category of problem No. % No. % 1) Relationship/Emotional/ 46 mental illness 2) Practical 3) Associated with physical 33 Disability/illness 38.7 42 51.1 40 33.6 23 28.1 27.7 17 20.8 119 100 82 100 By attaching a small group of research-oriented social workers to a primary care health-center and a local social service department we have been able to monitor the nature and extent of social pathology encountered in the community, and to examine the means of intervention adopted by the social workers and to evaluate their efficacy. One prospective study designed to evaluate the therapeutic role of a social worker attached to a metropolitan general practice in the management of chronic neurotic illness has already been reported. 15/ The psychiatric and social status of two matched groups of patients, one attending the practice with a social-worker attachment and the other attending neighboring practices without this facility, were ascertained independently at the beginning and end of a twelve-month period, using standardized interview techniques. A comparison between the outcome of the groups indicated some benefit to patients who had received the experimental service. Although both groups showed a reduction in psychiatric symptoms during the follow-up year, the fall was much more pronounced in the experimental group. At follow-up, 38.0% of the experi- mental group had been taken off psychotropic drugs, compared with 24.7% of the controls. Continuing medical care and supervision were deemed necessary for 59.8% of the experimental patients, compared with 77.3% of the controls. Similarly, the experimental patients were found to have improved in all main areas of social functioning, whereas the controls showed very little change in this respect at the end of twelve months. Changes in the psychiatric and social-adjustment scores for the two groups were positively correlated. These findings suggested that social-worker intervention has some therapeutic effect on chronic neurotic illness, at least in some cases, and hence that it is realistic in this context to speak

220 of social treatment. But what, precisely, is social treatment? Even a cursory glance at the vast social work literature reveals a major split which Dame Eileen Younghusband has identified as: ... the conflict between casework and the young radical school of community action. This refers to the knowledge mainly used by caseworkers and that mainly used by com- munity activists. To the latter the unpardonable sin is that casework method is largely based on psychoanalytic theory which causes the caseworker, so they allege, to be primarily concerned with a professional relationship, with the client's unconscious motivation, and with use of the transference in an essentially unequal situation, when what he really needs is help in getting means tested benefits to which he is entitled, or better housing or education or more pay. Conversely, community work draws draws largely on knowledge from sociology and political theory, both of which seem to be active, related to the real world, concerned with how to bring about social change. This is in sharp contrast to dynamic psychology which seems to them anything but dynamic in its social context because it implies that basically human nature is unchanging. 16/ The "casework" concept has tended to dominate the theory and prac- tice of social work in the United States, especially with the large corps of social workers in private practice. It has also been influential in Britain despite the many differences in organizational structure. Yet while its relevance to the needs of the general population have been challenged on theoretical grounds by the advocates of social change neither group has provided empirical evidence to confirm its own claims. As part of our study of chronic neurotic illness it was possible to undertake a detailed analysis of the social worker's activities and to relate these to the client response. 17/ In nearly two-thirds of this sample the social worker's contribution was restricted to helping the patient and his or her family in dealing with practical problems and difficulties, a function for which social workers are specially trained and in which their skills do not overlap to any large extent with those of the psychiatrist. In the remaining one-third she exercised what may be regarded as a quasi-psychotherapeutic function, although here also practical help and support were given in a proportion of cases. The prominence of what has also been called "social brokerage" 18/ rather than traditional "casework" has been detected in other studies which have analyzed social worker activities in general practice e 19/ ~ 20/ ~ 21/ ~ 22/ & 23/ ~ The specificity of such intervention, however, remains question- able. On the available evidence the most probable explanation of any benefits conferred by the social worker appears to reside in

221 the way in which her personal activities supplement the resources which she mobilizes and which facilitate a more positive approach by the general practitioner towards a greater awareness of the social orbit of morbidity. Stimson 24/ has pointed out that the global notion of the social element in general practice embraces several themes: the social relationships between doctors and their patients; the awareness of social factors in disease and in illness-behavior; the social causes of disease; the social consequences of disease; social welfare problems; and the socio-psychotherapeutic role of the-doctor. [he presence of a social worker as part of a primary care team may be expected to catalyze all these activities and so diffuse his or her influence at various points of professional contact. Finally, I should like to pull my argument together and devote a few words to its implications. On logistical grounds alone it is apparent that the mental health care of the community at large cannot be provided by psychiatric specialists. Our own alternative, advanced 15 years ago and based on epidemiological evidence, was that "the cardinal requirement for improvement of the mental health services . . . is not a large expansion and proliferation of psychiatric agencies, but rather a strengthening of the family doctor in his therapeutic role." 1/ In practice, of course, this emphasis on the primary care system must pay regard to the national variations in the structure of medical organ- ization, but the underlying principle has been endorsed and extended by the conclusion of a World Health Organization report that: "The primary medical care team is the cornerstone of community psychiatry." .75/ This statement is slowly finding favor with good family practition- ers, who see the core of their task in their own terms, as exemplified by the comment from a prominent British representative of the Royal College of General Practitioners some years ago: "The first thing a general practitioner has to decide is the relative importance of the emotional and physical factors in his patient's problems. Only the general practitioner approaches the matter quite in this way, and his ability to do so depends on his unique previous knowledge of the patient. Where this knowledge is denied to the doctor, assessment has to be made by more devious and less certain method of evaluating the emotional component by exclusion of the organic. This method of evaluating the emotional component is clumsy. For the 10-20% of selected problems which reach the hospital-based doctor, it is unsuitable and also waste- ful of medical resources. "The organic element is less definable in illness encountered by general practitioners than it is in the selected illness encountered in hospital practice. The emotional element, on the other hands is relatively more important in general practice." 26/ Today, I suspect this physician would substitute 'psychosocial' for 'emotional' and 'the primary care team' for the 'general pr`~cti- tioner'. He would not, however, modify his conclusions that the

222 assessment of mental ill-health in its broader sense is a central function of the primary care team, including non medical members of the caring professions -- the social worker, the health visitor, the mid-wife and the nurse. But what, then, is the role of the psychiatrists in this situa- tion? Not, I would suggest, necessarily that of other medical special- ists -- e.g., the dermatologist or the oto-rhino-laryngologist -- who can lay claim to authority in the diagnosis and treatment of not only the major conditions which come their way in hospital but also the vast mass of minor conditions which can be managed on an extramural basis. Here there is a continuity of expertise to which the psychia- trist cannot lay claim unless he sees himself in the false roles of specialist in psychological hermeneutics or in human engineering or in neurobiological manipulation, all of them self-images which have assumed some prominence in the past 30 years. In such roles his contribution is unlikely to be more than partially relevant to the problems posed by mental ill-health at the level of primary care. An altogether broader-based view of the discipline is required to encompass the presentation of mental ill-health not only as a series of particular clinical states but also as an integral component of much physical sickness on the one hand and much social dysfunction on the other. All too often the good general practitioner knows as much as may be required about his patient's background and domestic circumstances and about the community resources available to him. What he wants above all from his psychiatric colleagues are fact:s to help better diagnose and manage his own patients himself. Which drug should be administered to which patient, in what dosage and for how long? What are the diagnostic criteria he should employs? How efficacious are the available therapeutic measures, whether physical or psycho-social, for the population under his care? Such basic questions, and their numerous congeners, cannot be answered satisfactorily in the present state of knowledge. They are, however, eminently susceptible to investigation and, in many instances, to clinical investigation, although in a rather different setting from that to which the psychiatrist based in hospital, clinic of office is accustomed. Here, I would suggest, is a logical point of entry for the psychiatrist in search of a necessary if not sufficient role on the primary care scene. Furthermore, as a participant on these terms the mental health professional may become a beneficiary as well as a donor, for the task would surely help restore the holistic concept of the discipline which, though it has receded in recent years, underlay Adolf Meyer's notions of psychobiology which itself reached back to the earlier concept of psychological medicine and, still earlier, to the views of Andrew Wynter on psychiatry in relation to family medicine, expressed more than 100 years ago: "...we are convinced," he wrote in 1875, "that for the good of general medicine,

223 this particular study of psychological medicine, dealing as it does with so many complex problems should be merged in the general routine of medical practiced 27/ Such a process of integration, or rather of re-integration, would {6 for the good of not only general medicine but also general psychiatry.

224 REFERENCES Fry, J. (1969): Medicine in Three Societies. Aylesbury: Medicine and Technical Publications. 2. Shepherd, M., Cooper, A.B., Brown, A.C. and Kalton, G.W. (1966~: Psychiatric Illness in General Practice. Oxford University Press: London. 3. Watts, C.A.H. (1966~: Depressive Disorders in the Community. Briston: John Wright. Blacker, C.P. (1946~: Neurosis and the Mental Health Services. London: Oxford University Press. 5. Bennett, D.H. (1973~: Community mental health services in Britain. American Journal of Psychiatry, 130, 10, 1065-1070. - 6. Nijdam, S.J. (1973~: Discussion. In Depression in Everyday Practice, edited by Kielholz, P. Bern: Huber, p. 221. 7. Eastwood, M.R. (1975~: The Relation between Physical and Mental Illness. Monograph No.4. Toronto: University of Toronto Press. 8. Shaffer, J.W., Nussbaum, K. and Little, J.M. (1972~: MMPI profiles of disability insurance claimants. American Journal of Psychiatry, 129,403-408. 9. Shepherd, M. (1977~: Beyond the layman's madness: the extent of mental disease. In Developments in Psychiatric Research, edited by J.M. Tanner, London: Hodder and Stoughton. p. 178-198. 10. Clare, A.W. and Cairns, V.E. (1979~: Design, development and use of a standardized interview to assess social maladjustment in community studies. Psychological Medicine, 8, 589-604. Fitzgerald, R. (1978~: The classification and recording of 'Social Problems,' Social Science and Medicine, 12, 255-263. Corney, R. and Briscoe, M. (1977~: Investigation into two dif- ferent types of attachment schemes. Social Work Today, 9, 10-14. Corney, R. (1979~: The extent of mental and physical ill-health of clients referred to social workers in a local authority department and a general attachment scheme. Psychol. Med., 9 (In press).

225 14. Report of Working Party (1974): Social Work Support for Health Service. Department of Health and Social Security and the Welsh Office. London:HMSO. 15. Cooper, B., Harwin, B.G., Depla, C. and Shepherd, M. (1975~: Mental health care in the community: an evaluative study. Psychological Medicine, 5, 372-380. 16. Younghusband, E. (1974~: The future of social work. Social Work Today, 6, 33-37. 17. Shepherd, M., Harwin, B.~., Depla, G. and Cairns, V. (1979~: Social work and the primary care of mental disorder. Psychologi- cal Medicine (In the press). Baker, R. (1976~: The multi-role practitioner in the generic orientation to social work practice. British Journal of Social Work, 6, 327-352. 19. Collins, J. (1965~: Social Casework in a General Medical Practice. London: Pitman Medical. 20. Forman, J.A.S. and Fairbairn, E.M. (1968~: Social Casework in General Practice: a Report on an Experiment Carried Out in General Practice. London: Oxford University Press. 21. Ratoff, L. and Pearson, B. (1970~: Social case-work in general practice: an alternative approach. British Medical Journal, ii, 465-477. 22. Cooper, B. (1971~: Social work in general practice: the Derby scheme. Lancet, 1, 539-542. Goldberg, E. and Neill, J.E. (1972~: Social Work in General Practice. London: George Allen and Unwin. 24. Stimson, G. (1977~: Social care and the role of the general practitioner. Social Science and Medicine, 11, 485-490. 25. Report of Working Group (1973~: Psychiatry and Primary Medical Care. Copenhagen: WHO. 26. Crombie, D.L. (1972~: A model of the medical care system: a general systems approach. In The Economics of Medical Care, edited by Hauser, M.M. London: Allen and Unwin, pp.61-85. 27. Wynter, A. (1875~: The role of the general practitioner. In The Borderlands of Insanity. London: Robert Hardwicke.

227 MANAGEMENT OF EMOTIONALLY DISTURBED PATIENTS IN PRIMARY CARE SETTINGS: A REVIEW OF THE NORTH AMERICAN LITERATURE Janet Hankin, Ph.D. Johns Hopkins Health Services Research and Development Center Baltimore, Maryland In the Rome or one neaten practitioner or first contact, the primary care provider may be in an ideal position to recognize and treat the emotionally disturbed patient. The question emerges as to how the emo- tionally disturbed patient is managed in the primary care setting. This paper addresses the issue by reviewing a portion of the literature on the subject that was published from 1959 through 1978. North American primary care settings are included in this review. Mental disorder is broadly defined and includes disorders classifiable under Diagnostic and Statistical Manual (DSM-II), as well as problems in living, stress reactions, interpersonal crises, etc. 1= ^F Who health ~__~: I: ~c c: -~_ Before describing the management of emotionally disturbed patients by primary care practitioners, it is important to emphasize that in order to manage those patients, the providers must first identify patients with mental disorder. In their presentations to this confer- ence, Dr. Goldberg and Dr. Hoeper have raised serious questions about the ability of primary care providers to detect mental disorder. The studies reviewed here describe the management techniques for patients who have been identified. The fate of patients with hidden psychia- tric morbidity remains unknown. Primary care providers have several choices in the treatment of emotionally ill patients. They can pre- scribe psychoactive drugs, provide psychotherapy, and/~'r refer the patient to a mental health specialist. This discussion focuses mainly on the first two options. Published studies suggest that the prescription of psychotropic drugs is the modal management technique in the primary care setting. Nonpsychiatric physicians use drug threapy in 67 percent of-visits by patients with a diagnosis of mental disorder; psychotherapy is used in 22 percent of those visits. 1/ The proportion of primary care patients with emotional disorder receiving a prescription for psychotropic drugs ranges from 29 percent to 79 percent, depending upon the setting. 2/-5/ These rates of psychotropic drug prescribing are averages and thus disguise a wide variation among practitioners at each site. Several authors have reported the types of drugs prescribed. Hesbacher,

228 et al., studied seven family practices in Philadelphia and found that 55 percent of the patients on psychotherapeutics were receiving anti- anxiety drugs, 19 percent antipsychotic drugs, 16 percent antidepres- sants, 9 percent sedatives, 1 percent stimulants. 2/ These proportions are very similar to those reported by other investigators. 3/,6/,7/ Patient characteristics (such as age, sex, degree of psychological distress) and provider characteristics (year of medical school gradua- tion, isolation from colleagues) are correlated with the prescribing of psychotropic drugs. 2/,8/,11/ Two researchers have examined to what extent psychotropic drugs are used alone and when drugs and psychotherapy are combined. Fink, et al., report that in the Health Insurance Plan of Greater New York, 57 percent of patients with identified mental disorder were treated by drugs alone. 3/ Rosen, et al., studied five general hospital clinics in Monroe County, New York, and found that physicians used psycho- active drugs alone for 14 percent of patients, a combination of drugs and supportive therapy for 35 percent of those emotionally ill, and supportive therapy, drugs, and suggested environmental change for 10 percent of these patients. 5/ The efficacy of psychoactive drugs alone, compared with psychotherapeutic drugs combined with psychotherapy by the primary care physician, remains unstudied. 1/ In the course of the literature review, few studies were loca- ted that addressed the question of the appropriateness or quality of psychotropic drug prescribing by primary care physicians. Data indicate that not every patient prescribed a psychoactive drug suffers from mental disorder. Parry, et al., found that only one-third to one-half of psychotropic drugs were prescribed for a diagnosis of mental disorder. _ / Psychotropic drugs have also been prescribed for the treatment of obesity, insomnia, cardiovascular disease, musculoskeletal and gastrointestinal problems. The appropriateness of these prescribing patterns has been questioned. Careful monitoring of patients using psychotherapeutics has been urged, given the problems of patient noncompliance and the risk of overdosing or adverse drug reactions. _/,8/ The studies reviewed point to the need for additional research in the area of the management of emotionally ill patients by primary care physicians prescribing psychotropic drugs. Some of the unresolved issues include: First, when is it appropriate for primary care physi- cians to prescribe psychotropic drugs? Should every depressed patient receive an antidepressant? Can the primary care physician manage the patient on lithium? Second, are psychotropic drugs efficacious in and of themselves for the primary care patient, or should these drugs be combined with psychotherapy? Carefully controlled clinical trials are needed to answer this question. Third, is the management of psychiatric patients by psychotropic drugs a cost-effective method

229 compared with other alternatives? Fourth, what is the quality of psychotropic prescribing by the primary care physicians? Are they prescribing the correct type of drug for each type of mental disorder? Are the dosage levels high enough? Are they monitoring patients for compliance of the regimen and observing the potential for abuse? Are they concerned with adverse side effects and drug interactions? Finally, when should patients be referred to psychiatrists for drug therapy? The primary care provider also has the option of providing psychotherapy to the emotionally ill patient. The percent of patients with recognized mental disorder who receive some psychotherapy from primary care physicians ranges from 60 percent to 84 percent. 5/,13/,14/ Definition of psychotherapy varied with each setting. In some cases it was defined as "at least one discussion of his or her problem with the physician," while in other studies it was defined as counseling or supportive therapy. Psychotherapy, as it is vaguly defined, may be provided at some time to more than half of the mentally ill patients, but it occurs in only 22 percent of patient visits. 1/ Few data are available concerning the intensity and nature of these psychotherapy sessions. Two studies revealed that more than half of the patients receiving psychotherapy by primary care physicians are seen from one to four times for such therapy. 15/, 16/ Short term crisis therapy seems to be stressed by primary care practitioners. The content of the psychotherapy sessions is largely a mystery, although a few primary physicians provide anecdotal self-reports of how they treat their emotionally ill patients. 15/ Zabarenko and her colleagues at the Staunton Clinic in Pittsburgh observed eight general practitioners in 387 patient visits. They found that physicians were not always actively aware of psychological distress in the patient. _ / In a later study by these researchers, a psychia- trist observed a family physician for one year and noted that the following lessons could be learned about the management of a primary care patient with mental disorder: First, too many questions may impede the flow of patient information or prematurely terminate the transaction. The family physician should allow the patient to decide how much therapy he or she can tolerate and, finally, every behavior of the patient should be observed and evaluated. 18/ One of the recurring themes in the literature is that formal psychotherapy is difficult to conduct within the context of the primary care setting because of the limited time available to the provider. 4/,19/-21/ Figures from the National Ambulatory Medical Care Survey show that the nonpsychiatric physician averages only 13 minutes per visit with his patients. Several authors have sug- gested solutions to this, including the 20-minute hour or the 10- minute psychotherapy. 22/,23/

230 Studies on the outcome of psychotherapy by primary care physi- cians are practically nonexistent. Fink, et al., found that 36 percent of the patients receiving psychotherapy and 34 percent of their physi- cians rated the therapy as "very helpful." 3/ Controlled studies of the efficacy of primary care providers' psychotherapy could not be located. The dearth of research studies in the area of psychotherapy by primary care practitioners leaves many questions unanswered, including the following: What types of patients are best treated by primary care psychotherapy and what types of patients should be referred to psychiatrists? We need controlled, randomized trials to answer this question. What type of psychotherapeutic techniques are appropriate in the primary care setting? What technique is most successful for a given psychiatric diagnosis? Again, we need carefully designed outcome studies. Is it cost effective to provide psychotherapy in the primary care setting? What are the most successful collabora- tion models for psychiatrists and primary care providers? Experi- ments in liaison psychiatry and the team approach have been imple- mented in several delivery sites, but additional studies of these models are needed. In conclusion, the literature reviewed indicates that the primary care providers do manage emotionally disturbed patients with psychotropic drugs and psychotherapy. However, systematic data on the utilization of these techniques are lacking. One has the impression that drug prescribing and psychotherapy occur in a haphazard way. The works reviewed point to the need for future research and planning in order to effectively deliver mental health services within the primary care setting.

231 References Brown, B. S., Regier, D. A., and Baiter, M. B., Key interactions among psychiatric disorders, primary care, and the use of psycho- tropic drugs in Brown, B. S., ed. Clinical ~ ~,'~ Medicine. Princeton: Excerpta Medica, 1979. Anv1 REV/ l Pnc:1 ran ~ n lemma - `r Hesbacher, P.' Rickels, K., Rial, W. Y., Segal, A. and Zamostein, B. B., Psychotropic drug prescription in family practice. Compre- hensive Psychiatry, 17: 607-615, 1976. r 3. Fink, R., Goldensohn, S., Shapiro, S., and Daily, E., Changes in family doctors' services for emotional disorders after addition of psychiatric treatment to a prepaid group practice program. Medical Care 7:209-224, 1969. Locke, B. Z., Finucane, D.L. and Hassler, I., Emotionally disturbed patients under care of private nonpsychiatric physicians. American Psychiatric Association: Psychiatric Research Report 22:235-248, 1967. 3. Rosen, B. M., Locke, B. Z., Goldberg, I. D., Barbigian, H.M., Identification of emotional disturbance in patients seen in general medical clinics. Hospital and Community Psychiatry, 23: 364-370, 1972. 6. 7. 9. Baiter, M. B. and Levine, J., The nature and extent of psychotropic drug usage in the United States. Psychopharmacological Bulletin, 5: 373, 1965. Blackwell, B., Psychotropic drugs in use today. The role of diazepam in medical practice. Journal of the American Medical Association, 225:1637-1641, 1973. Baiter' M. B., An analysis of psychotherapeutic drug consumption in the United States. Anglo-American Conference on Drug Abuse: Etiology of Drug Abuse, I, 1973, pp. 58-65. Thomas, R. B. and Wessinger, W. N. A descriptive assessment of the psychiatric practice of nonpsychiatric physicians in South Carolina. Southern Medical Journal, 66:221-224, 1973. 10. Fisher, J. V., Mason, R. L., and Fisher' J. C. Emotional illness and the family physician. Part II. Management and Treatment. Psychosomatics, 16:107-111, 1975.

232 Linn, L. S., Physician characteristics and attitudes toward legitimate use of psychotherapeutic drugs. Journal of Health and Social Behavior, 12:132-140, June 1971. - 12. Parry, H.J., Balter, M.B., Mellinger, G.D., Cisin, I.H., and Manheimer, D.I., National patterns of psychotherapeutic drug use. Archives of General Psychiatry, 28:769-783, 1973. 13. Fink, R. and Shapiro, S., Patterns of medical care related to mental illness. Journal of Health and Human Behavior, 7:98-105, 1966. 14. Locke, B. Z., Krantz, G., and Kramer, M., Psychiatric need and demand in a prepaid group practice program. American Journal of Public Health, 56:895-904, 1966. 15. Glasser, M., Psychiatry in family practice. Canadian Psychiatric Association Journal, 21:483-488, 1976. - 16. Fink, R., Goldensohn, S., Shapiro, S., Daily, E., Treatment of patients designated by family doctors as having emotional problems. American Journal of Public Health, 57:1550-1564, 1967. 17. Zabarenko, L., Pittenger, R.A., and Zabarenko, R., Primary Medical Practice: A Psychiatric Evaluation. St. Louis: Warren Press, 1968. 18. Zabarenko, R.N., Merenstein, J., and Zabarenko, L., Teaching psychological medicine in the family practice office. Journal of the American Medical Association, 218: 392-396, 1971. . 19. Brodsky, C.M., The systemic incompatibility of medical practice and psychotherapy. Diseases of the Nervous System, 31: 579-604, 1970. 20. Ornstein, P.H., and Ornstein, A., Local psychotherapy: Its potential impact on psychotherapeutic practice in medicine. Psychiatry in Medicine, 3:311-325, 1972. 21. Brackway, B. S., Behavioral medicine in family practice: A unifying approach for the assessment and treatment of psychosocial problems. The Journal of Family Practice, 6:545-552, 1978. . Costelnuovo-Tedesco, P. The Twenty Minute Hour. Boston: Little, Brown & Co., 1965. 23. Ornstein, P.H., and Goldberg, A., Psychoanalysis and medicine II. Contributions to the psychology of medical practice. Diseases of the Nervous System, 34:277-283, 1973.

233 AN ANNOTATED B IBLIOGRAPHY ON MANAGEMENT OF EMOTIONALLY DISTURBED PATIENTS IN PRIMARY CARE SETTINGS: A REVIEW OF THE NORTH AMERICAN L ITERATURE Janet Hankin, Ph.D. The following annotations represent a selection of all articles on the treatment of mental disorder by primary care physicians which were published between 1959 and 1979. The literature search was restricted to articles relating to North American primary care prac- tices. The articles address the problems of psychotherapy and drug prescribing by primary care providers, as well as psychiatric referral. The articles which were chosen for annotation are representative of others published works. Because of space contraints, only a sample of the articles could be annotated. The omission of any given article should not be viewed as judgment about its quality or importance.

235 Baiter, M.B. An analysis of psychotherapeutic drug consumption in the United States. Anglo-American Conference on Drug Abuse: Etiology of Drum Abuse, I, 1973, pp. 58-65. In 1972, 16 percent of all prescriptions filled were for psycho- therapeutic drugs. These 215 million prescriptions were for anti- psychotics, anti-anxiety agents, antidepressants, sedatives, hypnotics and stimulants. Forty-four percent of the psychoactive drug prescriptions were for anti-anxiety drugs. Based on estimates from a national survey of 2,552 persons, 22 percent of American adults used a psychotherapeutic prescription drug. Women and persons over 60 were more likely to use these drugs. Data from the National Disease and Therapeutic Index indicate that general practitioners account for 50% of all new psychoactive drug therapy. Only one-third of all prescriptions for anti-anxiety drugs were given to patients with a diagnosis of mental disorder. If senility is added to this category of mental disorder, 50 per- cent of anti-anxiety drug prescriptions were for patients with emotional distress, and 50 percent for patients with physical illness. Brockway, B.S. Behavioral medicine in family practice: A unifying approach for the assessment and treatment of psychosocial problems. The Journal of Family Practice, 6:545-552, 1978. Dr. Brockway defines behavioral medicine as "any procedure or set of procedures based on learning theory which is used to assess and treat medical or medically related problems." The family practitioner who is oriented to behavioral medicine sees the patient's psychosocial problems "as a combination of behavior (or skill) deficits or excesses." Once the practitioner defines the problem, s/he must determine: What skills need to be learned to achieve the desired outcome? What factors in the environment maintain the skill deficit or excess? Next, the physician can use a variety of techniques to teach more appropriate behaviors (shaping and modeling, systematic desensitization, biofeedback, assertiveness training). Family practitioners are said to have used these methods to successfully treat persistent vomiting in an infant, headaches, encopresis, enuresis, severe temper tantrums, marital conflict, reactive depression, and primary orgasmic dysfunction. Initial data to determine the appropriate mode of management can be gathered within the constraints of a 20 minute office visit. Behavior treatments tend to succeed in one to six months. Family practitioners can also use behavioral medicine in a preventive way. For example, during prenatal visits, a couple can learn child management skills.

236 Brown, B.S., Regier, D.A., and Baiter, M.B. Key interactions among psychiatric disorders, primary care, and the use of psychoactive drugs. In Brown, B.S. (ed). Clinical Anxiety/Tension in Primary Medicine. Princeton: Excerpta Medica, 1979. The authors estimate that 60 percent of persons with mental disorder are treated by primary care providers. Of all persons with mental disorder, an estimated 32 million people in a given year, 15 percent see only mental health specialists, 54 percent see only primary care providers, 3 percent use general hospitals or nursing homes, 6 percent use both primary care and mental health providers, and 22 percent go untreated or use resources outside the health care system. About 5 percent of all patient visits to physicians in office-based practice result in a diagnosis of mental disorder. About half of these visits where a mental disorder diagnosis is received involve psychoneuroses. "Nonpsychiatric physicians account for more than 50 percent of all visits in which a diagnosis of psychiatric disorder is assigned." Nonpsychiatric physicians employ psychotherapy in 22 percent of visits with a diagnosis of mental disorder, and drug therapy is used at 67 percent of these visits. The authors call for research to determine which psychiatric disorders are best treated by psychotropic drugs, which disorders require psychotherapy, and when a mixed mode of treatment is needed. Additional research is needed to determine how to divide the responsibility between the primary care physician and the psychiatrist; what is the relative effectiveness of treatment by each type of provider, and the cost effectiveness of the treatment. Carey, K., and Kogan, W.~. Exploration of factors influencing physi- cian decisions to refer patients for mental health service. Medical Care, 9:55-66, 1971. The authors asked physicians in the specialties of general practice, medicine, and surgery to describe two patients who were referred to mental health services and two patients with emotional problems who were not referred. Seventy-eight of 92 physicians of the Group Health Cooperative of Puget Sound participated. Data were obtained on 140 patients who had been referred and 125 who were not referred. Patients with acute conditions and those who requested referral were more likely to be referred. The feeling of inability or lack of experience on the part of the physician resulted in a referral. The

237 "treatment" of mental disorder varied with the specialty: the medical group used diagnostic procedures; the surgical group, placebos; and the general practice group, psychoactive drugs. Coleman, J.V., and Patrick, D.L., Psychiatry and general health care. American Journal of Public Health, 68:451-457, 1978. The authors describe a five year experience with the Community Health Center Plan of Greater New Haven, Connecticut. This prepaid group practice integrates mental health services into primary care teams in internal medicine and pediatrics. The mental health clinician is a psychiatric social worker, a psychiatric nurse specialist, or a clinical psychologist. While the primary care providers have the major responsibility for total health care, the mental health clinician acts as a primary care extender. Psychiatrists are available for back- up support. During a two year period 15.7 percent of the patients seen were diagnosed as suffering from emotional problems (N=2,806~. Primary care clinicians alone handled 72 percent of these patients, and mental health clinicians (in addition to primary care providers) treated 28 percent. Mental health clinicians treated 55 percent of patients with chronic emotional problems. The proportion of the following diagnoses were treated by primary care clinicians alone: nonorganic psychoses (54%), anxiety (88%), depression (67%), personality disorders (47%), sexual problems (75%), alcohol pro- blems (80%), durg abuse problems (65%), situational disturbances (70%), social adjustment problems (58%), suicide ideation/attempt (75%~. The primary care physicians usually handled medication maintenance, although at times referral to mental health clini- cians was necessary to assist the primary care physician to establish the medication regimen. The authors conclude that this team approach has the major advantage of making the mental health clinician readily available to the primary care provider which can relieve him or her "of the undue, sometimes inordinate pressure of certain persistently demanding patients, usually patients with chronic characterological depressions and borderline states." Dressier, D.M. The management of emotional crisis by medical practi- tioners. Journal of the American Medical Women s Association, 28:654- 659, 1973. The person who is unable to handle stress utilizing customary modes of coping frequently consults the physician. The physician can reduce the distress and prevent further deterioration for

238 patients experiencing emotional crises such as loss of job, financial loss, serious illness, death of a family member. The physician should first help the patient feel that he "is not 'going crazy' or 'out of control'.'' The physician should adopt an "accepting, non-judgmental attitude" and be "calm but concerned, flexible but firm, and receptive but involved." The interview should focus on the current problem, rather than delving into the past, and conflicting feelings should be recognized. The physician should help the patient broaden his/her repertoire of coping skills. Family members should be involved in treatment. Medication may be use- ful in reducing symptomatic distress. The patient should be seen once or twice a week. When the patient has recovered, the physician should assess the need for follow-up specialized psychiatric care, especially if there is evidence of psychotic or neurotic symptoms, if the patient has a history of recurrent crises, and/or if the patient is interested in a deeper subjective examination. Fink, R., Goldensohn, S., Shapiro, S., and Dailey, E. Treatment of patients designated by family doctors as having emotional problems. American Journal of Public Health, 57: 1550-1564, 1967. The authors interviewed physicians at the Jamaica Medical Group of the Health Insurance Plan of Greater New York about 422 patients over 15 years old who were diagnosed as having a mental, psychoneurotic, or personality disorder. Twenty-six patients were referred for a psychiatric consultation. Patients were more likely to be referred if they had a chronic condition, a condition that greatly interfered with life activities, and/or a condition which was thought to improve with psychiatric treat- ment. Of all patients with mental disorder, 78 percent received a psychotropic drug and 92 percent had at least one lengthy doctor-patient discussion of the problem. Patients who did not improve with these treatments were more likely to receive a psychiatric referral. A total of 380 of the 420 patients were interviewed. Among patients with psychoactive drug prescriptions, 46 percent said they were very helpful, 34 percent somewhat helpful, and 20 percent little or no help. Nearly two-thirds of the patients reported that the doctor-patient discussions about their emo- tional problems were very helpful (34%) or somewhat helpful (30%~. Twenty-three percent reported that the discussions were of very little or no help, and 13 percent said they had not had any discussion.

239 Fisher, J.Ve ~ Mason, R.L., and Fisher, J.C. Emotional illness and the family physician. Part II: Management and Treatment. Psychosomatics, 16:107-111,1975. The authors surveyed 860 family physicians who were members of the Michigan Chapter of the American Academy of General Practice regarding detection and management of emotional illness. Physicians graduating from medical school after 1950 had a ten- dency to use psychoactive drugs with a lower percentage of their emotionally ill patients than physicians graduating prior to 1950. Physicians graduating before 1950 were more inclined to use tranquilizers and antidepressants with 50 percent or more of their emotionally ill patients than later medical school graduates. About 80 percent of all physicians used advice and reassurance for their mentally ill patients. Physicians graduating after 1950 were more likely than earlier graduates to use psychotherapy with their patients. Glasser, M. Psychiatry in family practice. Canadian Psychiatric Association Journal, 21:483-488, November, 1976. The author is a family practitioner who reviewed the charts of all patients he saw between September 1, 1964 and August 1, 1968 (N=4,801~. Of these patients, 394 were classified as psychiatrically ill. Beginning in September, 1967, he question- ed each returning psychiatric patient about the original com- plaint. A total of 287 patients identified as psychiatrically ill were evaluated. The remaining 107 psychiatric patients were seen only once and did not return for additional visits. The 287 patients involved in the follow-up study were evaluated in relation to change in symptoms' degree of functional impair- ment, mental status, level of severity of disease, and any new symptoms. The majority of patients were also rated by a psycho- logist. Nearly three quarters of the patients were diagnosed as neurotic; 11 percent had adjustment reactions; and the re- mainder were psychotics, drug and alcohol addicts, or suffered from character disorders. The author used therapy that was "eclectic, at times being supportive and fostering catharsis, while on other occasions being directive and offering interpretations of behaviour." Twenty-six patients were referred to mental health specialists. Most patients were seen by the family practitioner only five times, and none more than twenty-two. There were no signifi- cant differences in improvement rates for the patients treated by the mental health specialists versus those treated by the family practitioner. At the time of follow-up, 74 percent of the patients were judged to be better, 20 percent neutral, and 6 percent worse.

240 Greco, R.S. Psychiatry in everyday medical practice. Psychiatry in Medicine, 3:303-309, 1972. The author participated in Balint-type training sessions at the University of Pittsburg Medical School. He became convinced that "every doctor-patient transaction has a workable psycho- therapeutic aspect." He provides several examples in the dis- cussion. When the patient presents him/herself to the physician without specific complaints (e.g., for routine physical exams), the physician should take this opportunity to practice pre- ventive psychiatry. When the patients present an unorganized illness, symptoms are presented, and the physician can work to organize the illness and restore the patient's balance. In an organized illness, the patient and doctor agree on a diagnosis, and the physician can treat the problem directly. Hesbacher, P., Rickels, K., Rial, W.Y., Segal, A., and Zamostein, B.B. Psychotropic drug prescription in family practice. Comprehensive Psychiatry, 17:607-615, 1976. The authors surveyed 1,190 patients seen in seven family practices from March to September, 1970. Of these patients, 48.2 percent were experiencing emotional problems currently or had suffered from emotional problems during the past two years. Among patients with emotional disorder, 50.4 percent had received a psychotropic drug prescription within the past two years. Twenty-nine percent of the emotionally ill patients were taking psychotropic drugs at the time of the survey. All patients completed the Hopkins Symptom Checklist (HSCL) at the time of their visit. The authors found a general trend that the scores were highest for patients with emotional problems currently on drugs, lower for patients with emotional problems who were previously on drugs, followed by patients with emotional problems who were never on drugs, and the lowest for patients without emotional problems. The patients who had been prescribed drugs in the past two years were most likely to receive anti-anxiety drugs (55 percent of all patients prescribed psychoactive drugs) followed by anti-psychotic drugs (19%), antidepressants (16%), sedatives (9%), and stimulants (1%~.

241 Ketai, R. Family practitioners' knowledge about treatment of depressive illness. Journal of the American Medical Association, 235:2600-2603, - 1976. Ketai chose 227 family practitioners attending a seminar at the University of Michigan Medical Center as his subjects. Before the physicians heard a lecture on psychotropic drugs, they completed a multiple choice examination on the prescribing of psychotropic drugs. The answers of the physicians were compared with those of seventeen psychiatrists. The greatest discrepancy between the family practitioners and psychiatrists occurred for depressive illness. Nearly 27 percent of the family practitioners were unaware of how best to treat a depressed patient with severe anxiety and agitation. One-fourth of the family practitioners would begin treatment with tricyclics at too low a dose, while 9% at much too high a dose. Thirty-nine percent of the family practitioners would not raise the tricyclic dose to acceptable and proper levels. The author concludes that family practitioners should be taught how to use tricyclic antidepressants. He recommends a starting dose of imipramine hydrochloride or amitriptyline hydrochloride of 75 mg/day which should be raised to at least 150 mg/day within a few days. Some patients require dosage levels of 200-250 mg/day. A trial of two to three weeks at therapeutic levels is needed before determining that the drug is ineffective. Kiely, W.F. Psychotherapy for the family physician. American Family _ractice, 3:87-91, 1971 The author suggests that family practitioners practice suppressive and supportive psychotherapy, aiming toward intra- psychic equilibrium, rather than reorganizing the personality. Kiely suggests that while the initial visits may last from 30- 60 minutes, follow-up visits of 15-20 minutes are adequate. The family physician can begin the interview with simple questions like, "How have things been going generally?" or "I have the feeling that you've been working under a good deal of tension." However, the physician should avoid asking too many questions and using technical terms. Kiely argues that the physician should focus on current feelings and symptoms, rather than delving into the past. The family physician should avoid: provoking anxiety in the patient, showing judg- mental attitudes, creating a hostile reaction to the physician, and confronting the patient.

242 Locke, B.Z., Finucane, D.L., and Hassler, F. Emotionally disturbed patients under care of private nonpsychiatric physicians. American Psychiatric Association: Psychiatric Research Report, 22:235-248, 1967. Seventy-nine general practitioners out of 107 in Prince Georges County, Maryland, kept records on patients seen in one week during February-July, 1964. A total of 7,814 patients were included. Physicians identified 7 percent of the sample as emotionally ill (9 percent of those white and 15 years of age or older). Physicians provided the following types of care to patients with mental disorder: suggested psychiatric care or counseling (25% of patients), gave supportive therapy (59%), suggested environmental or social change (19%), prescribed drugs for the psychiatric problem (60%), suggested referral to other agencies or persons (6%), suggested other recommendations or other therapy (5%), none of the above (8%~.* Locke, B.Z., Krantz, G., and Kramer, M. Psychiatric need and demand in a prepaid group practice program. American Journal of Public Health. 56:895-904, 1966. All patients aged 15 and over seen at the Group Health Association (Washington, D.C.) in the Departments of Internal Medicine, Pediatrics, Allergy, and Dermatology during a 3 1/2 month period were included, N=6,104. Nearly 15 percent of the patients seen had a mental or emotional problem. Seventy-five percent of these patients were treated with psychoactive drugs, 63 percent received counseling, and 17 percent were referred for outside psychiatric help. For another 18 percent of patients who were not referred, the physician wanted to recommend addi- tional treatment, but was reluctant to do so because s/he felt that the patient would find it unacceptable or too costly. Ornstein, P.H., and Goldberg, A. Psychoanalysis and medicine. II. Contributions to the psychology of medical practice. Diseases of the Nervous System, 34:277-283, 1973. The authors describe two techniques of psychotherapy that can be used by the primary care physician. The long interview (or focal psychotherapy) focuses on the patient's life situation and personality. The physician spends most of the time listening in order to uncover: the problem that caused the illness, the effects of the patient's behavior on others, unconscious conflict- ing motives, and the conflict which is at the root of the problem. *Percentages add to more than 100% because patients could receive more than one type of care.

243 However, the long interview is time consuming, so an alter- native technique, the ten minute psychotherapy or "flash" may be used. The flash is empathetic understanding and "requires as a fine tuning-in, a briefly sustained intense identification with the patient that leads to a knowledge about him which doctor and patient silently share for the benefit of the patient." Raft, D. How to refer a reluctant patient to a psychiatrist. American Family Physician, 7:109-114, 1973. Psychiatric referral is more difficult when the family physi- cian has a negative attitude to psychiatry and transmits these feelings to the patient. Sometimes the referral is made because the physician is disappointed or angry with the patient who presents a physical symptom for which no organic base can be found. Even when the family physician does make a psychiatric refer- ral, the patient may be reluctant to cooperate. Many patients fear emotional illness or want to avoid exploring deep feelings. The family physician should exploit the doctor-patient relation- ship when s/he recommends psychiatric consultation. Some patients feel abandoned by the family physician when a psychiatric referral is made. The physician needs to reassure the patient that s/he will not be neglected. "The physician may simply have to refuse further investigation unless the patient will follow his advice to see a psychiatrist." Rosen, B.M., Locke, B.Z., Goldberg, I.D., and Babigian, H.M. Identi- fication of emotional disturbance in patients seen in general medical clinics. Hospital and Community Psychiatry, 23:364-370, 1972. - The authors studied patients seen in one month at four out- patient general medical clinics. Another clinic with a small case load reported on patients seen during two months. The clinics represented 5 of 6 outpatient general medical clinics in Monroe County, New York. A total of 1,413 patients aged 15 and older were studied. Twenty-two percent of the patients were diagnosed by their physicians as suffering from mental disorder. The types of treatment provided included: supportive therapy (31% of patients); drug prescriptions (14%~; environmental changes suggested (1%~; supportive therapy and drugs (35%~; supportive therapy and environmental changes (8%~; therapy, drugs, and environmental change (10%~.

244 Shortell, S.M., and Daniel, R.S. Referral relationships between internists and psychiatrists in fee-for-service practice: An empirical examination. Medical Care, 12:229-240, 1974. The authors interviewed 127 internists practicing in the northern suburbs of Chicago. During a one month period 0.9 percent of all patients seen were referred to psychiatrists. Internists over 50 years old, those in practice 20 years or more, solo practitioners, board certified, and those without a subspecialty had higher rates of psychiatric referral. Depression, followed by anxiety and neurosis were the most frequent reasons for psychiatric referral. Alcoholics were least likely to be referred to psychiatrists. The internists were generally satisfied with their patterns. Scaramella, T.J. Management of depression and anxiety in primary care practice. Primary Care, 4:67-77, 1977. The author argues that patients with anxiety syndromes "are more manageable and respond better to therapy when they are treated by their family doctor." These patients are reluctant to see a psychiatrist. The family doctor should work to alleviate the patient's fears by reassurance and under- standing. Specific case examples are presented which illustrate management techniques. Patients with anxiety states should be referred to psychiatrists when: l) neither patient nor physician can identify the source of stress or conflict, 2) patient fails to follow or benfit from primary care physician's treatment after three months, 3) patient expresses strong interest in psychotherapy, or 4) patient's personality makes it difficult for him/her to cooperate with primary care physician. The author presents techniques to overcome a patient's resistance to psychia- tric referral. The primary care physician can provide treatment for de- pressed patients in a majority of cases. Depressive disorder is usually characterized by low spirits, sleep disturbance, somatic complaints, and inability to function effectively. The primary care physician should: l) identify for the patient what is wrong; 2) explore the factors in the patient's life which may be contributing to the depression; 3) explain to the patient that somatic complaints, feelings of hopelessness and pessimism are part of the depression; 4) while acknowledging the symptoms, do not promote the use of symptoms to avoid life events; 5) measure the extent of depression by having the patient complete a depression inventory; 6) ask about self-destructive thoughts; and 7) Outline a specific treatment plan. Patients should receive a psychiatric consultation and/or referral if patient shows signs of psychosis, strong suicidal intentions,

245 previous episodes of mania, poor response to treatment after three months, patient has a depressive life style, or patient requests to see a psychiatrist. Smith, J.A. Office psychotherapy in family medicine. American Family Physician, 2:80-84, 1970. The author describes the symptoms, course, and treatment of anxiety by the family practitioner. Anxious patients may have acute episodes with autonomic symptoms such as cardiac palpita- tions, vertigo, dry mouth, and diffuse perspiration. The onset may be sudden and accompanied by intense fear and an urge to escape. After the acute episode, the patient may continue to complain about nausea, urinary frequency, vertigo, blurred vision, insomnia, palpitations, cardiac awareness, tinnitus, or cold hands and feet. The typical treatment of anxiety is to assure that patient that s/he does not have a dread disease. The physician should be careful about the content of both verbal and nonverbal communication. The physician should ask questions about the true cause of the patient's complaint, and should de-emphasize the importance of physical complaints. Patients with severe anxiety should receive an anti-anxiety agent. Zabarenko, R.N., Merenstein, J.; and Zabarenko, L. Teaching psychologi- cal medicine in the family practice office. Journal of the American Medical Association, 218:392-396, 1971. The authors describe an educational experiment where a psychiatrist served as a preceptor for a family physician. The physician learned not to make a rigid distinction between organic and psychiatric disease and realized the importance of seeing the total patient. It is important not to impede the flow of patient information by asking too many questions. Every be- havior of the patient should be observed (gestures, body language, banter). The physician learned to recognize major but hidden syndromes, especially depression. He learned that is was not wise to force the patient to accept the fact that no organic disease is present.