Research findings on chronic mental illness: A model for continuing care in the health maintenance organization

Research findings on chronic mental illness: A model for continuing care in the health maintenance organization

Research Findings on Chronic Mental Illness: A Model for Continuing Care in the Health Maintenance Organization James E. Sabin T 0 DATE, health main...

1MB Sizes 1 Downloads 27 Views

Research Findings on Chronic Mental Illness: A Model for Continuing Care in the Health Maintenance Organization James E. Sabin

T

0 DATE, health maintenance organizations (HMO) have largely avoided the problem of chronic psychiatric illness, as the cost of providing coverage might threaten the economic viability of a prepaid practice.’ Recent developments in Massachusetts law, however, have required an HMO, the Harvard Community Health Plan (HCHP), to develop an approach to chronic illness under conditions that provide a fair test of the HMO as a setting for effective yet economical treatment in this important area. The challenge of developing a treatment program tailored to the HMO setting has also provided an opportunity to reexamine the scientific basis for what is generatly called “continuing care.” The HCHP, a university-affiliated prepaid group practice serving the Greater Boston area, began operation in October, 1969. By 1975 it had grown to a membership of approximately 50,000. As has been characteristic of prepaid group practices nationally,’ mental h,ealth benefits were limited to patients with “psychiatric conditions of recent onset which are judged subject to significant improvement through relatively short term therapy.” Defining benefits this way limited expenses by excluding a group of disorders likely to require especially costly care--- longstanding psychiatric illnesses that are either slow to improve, fixed, or deteriorating---and especially costly modes of’ treutment---long-term psychotherapy or psychoanalysis and long-term hospitalization. As of January I, 1976, it became illegal to omit coverage of chronic psychiatric illness in Massachusetts. Chapter 1174 required any health insurance issued in the state to provide a certain level of benefit for “expense arising from mental or nervous condition.” The actual benefit level required, five hundred dollars of outpatient treatment and sixty days of psychiatric hospitalization, does not go significantly beyond the coverage already offered by the HCHP. But the law extends the range of conditions covered to include chronic illnesses that may require continuing treatment. Since Blue Cross--Blue Shield and private insurance companies are required to provide the same benefits. the law does not put an HMO at a competitive disadvantage in setting premiums. As part of the process of developing a treatment program for chronic mental illness, I reviewed research literature pertaining to the outpatient treatment of schizophrenia, the major affective disorders, and the less easily classifiable conditions that may require hospitalization and long-term care to see what practical guidelines can be derived from scientific studies of treatment effects. The disFrom

the Harvard

James

E. Sabin,

Medical

School,

Communit_v

Heaith

M.D.:Harvard Junior

Plan. Boston,

Community

Associate

in Medicine

Xeafth

Ma.rs. Plan. CIinicai

(Psychiatry),

Peter

fnstruc~or Bent

in Pswhiatry,

Brigham

Hospital,

Harvard Boston.

Mass. Address Street.

c

reprint

Boston.

requests

to: James

E. Sabin,

M.D.,

Harvard

C~rn~~~~~t~ Health

Pictn. 690 Beacon

Mass. 022/5.

197X h! Grune

d Stratton.

Inc. ISSN

0010-440X

00l0~440~~~7~~l90l-oo1o%o1.~N)/o

Comprehensive Psychiatry. Vol. 19. No. 1 (January/February),

1978

83

84

JAMES

E. SABIN

cussion that follows moves from an analytical review of relevant research to a series of conclusions presented in the form of a treatment model. The model takes into account research findings on treatment effectiveness and the kinds of opportunities and constraints posed by the structure of the HMO. This model has given direction to the continuing-care program now being implemented at the HCHP. Subsequent studies will report on the evolution of the program, staffing requirements, the impact on hospitalization, and evaluations of treatment outcome. PHARMACOTHERAPY

The research literature in psychopharmacology is especially useful to a continuing care program, as it bears on the prevention of relapse and the maintenance of health. This section centers on studies of the effectiveness of pharmacotherapy and of the equally important problem of how to identify the optimum regimen for a particular patient and to elicit the patient’s cooperation. In an excellent review of 24 controlled studies that compare relapse rates of schizophrenic patients treated with maintenance antipsychotic medication or placebo, Davis finds “overwhelming statistical evidence that the antipsychotic drugs prevent relapse in schizophrenia.“” Of 1068 patients treated with placebo, 698 (65%) relapsed, while only 639 (30%) of 2127 patients maintained on antipsychotics relapsed. Among the 24 studies, three are especially pertinent to a continuing care program. Hogarty and his collaborators followed 374 recently discharged schizophrenics in aftercare, combining more and less active sociotherapy with chlorpromazine and placebo.-‘-fi By 24 months, 80% of the placebo-treated patients relapsed, compared to 48% of the drug-treated patients. Leff and Wing compared chlorpromazine or trifluoperazine to placebo for 35 discharged schizophrenic patients whose illness was in the middle range of severity.’ Within one year, 80% of the placebo-treated patients relapsed, while only 35% of the phenothiazine-treated patients relapsed. Finally, Hirsch et al.x compared fluphenazine decanoate to placebo in a group of 81 chronic schizophrenic outpatients who had been maintained on fluphenazine decanoate for at least eight weeks. In the subsequent nine months, only 8% of the drug group relapsed, while 66% of the placebo group did! Given the unequivocal conclusion that antipsychotic m~ication has a powerful prophylactic effect in schizophrenia, and that no other established form of treatment is comparably effective in preventing relapse, the reciprocals of Davis’s statistical summary become crucially im~rtant. Sixty-five per cent of the placebo-treated patients relapsed, but 35% did not. While there is currently no way of predicting which schizophrenic patients will not require maintenance antipsychotic medication, the danger of serious side effects, especially the tardive dyskinesias, make it imperative to consider judicious clinical trials of decreased antipsychotic dosage. Apart from the physical side effects, for some patients the idea of needing medication has a deleterious impact on self-esteem. Finding that reduced levels suffice or that no medication at all is necessary may help their morale. And for those patients whose symptoms worsen with decreased dosage, the experience of collaborating with the physician in studying their own illness may enhance rapport and cooperationg+‘” In principle, the hospital phase of treatment could be used to identify which

MENTAL

ILLNESS

AND THE HMO

85

schizophrenic patients require no medication. However, May’s” comparative study of five treatment approaches to schizophrenia makes clear that diminished use of drugs will prolong hospital stay. Because of the expense incurred by hospitalization, a trial of several weeks without antipsychotics to identify s~ntaneous remitters is financially undesirable for both the patient and the HMO. For the patient, a prolonged drug-free trial may entail the expense of hospitalization beyond his insurance coverage. For the HMO, this approach will increase overall expenses and thereby require increased premiums or decreased benefits in other areas. Furthermore, in a follow up of his original study, May demonstrated that patients initially treated with drugs (or ECT), whose first admissions were appreciably shorter than those of the psychotherapy group, spent less time in the hospital after their release. Thus, by the criterion of hospitalization, drug-free treatment not only prolonged the initial hospitahzation, but also led to increased rehospitalization in the 2-5 years following the initial discharge. In the absence ol clear positive indications for extending a patient’s hospital care, an HMO should strive for the briefest possible hospitalization in order to concentrate its resources on vigorous precare and aftercare. While 70% of schizophrenics treated with antipsychotics did not retapse, 30%. did. Controlled studies show that a wide range of antipsychotic medications have essentially equivalent antipsychotic effect.‘” But clinical experience reveals that individual patients respond better or worse to different agents. Some response variation can be attributed to individual differences in absorption and metabolism. “.I,’ In addition, observations from a ten-year study of 92 chronic schizophrenics suggests there may be inherent variability in a patient’s response to the same drug at different times.“’ In the future, plasma levels of various metabolites may aid antipsychotic prescription, much as plasma lithium measurement assists the monitoring of lithium carbonate. For now, however, a continuing care program must be structured to facilitate the hallowed clinical method of trial and error in pharmacotherapy.” In a companion review of maintenance therapy for the affective disorders, Davis” examines eight well-controlled studies of lithium in bipolar aftectivr disease. These reveal an overall relapse rate of 79% with placebo (262 of 330), compared to 36% with lithium (117 of 329). Davis concludes that “the evidence that lithium prevents recurrence of manic-depressive illness has, in my opinion, been established beyond the shadow of a doubt.” Although fewer patients with unipolar affective disorder have been studied, evidence suggests lithium is approximately as effective for unipolar as for bipolar patients. Controlted studies of the preventive effect of maintenance tricyclics indicate considerable prophylactic value. essentially comparable to lithium. Because tricyclics do not prevent and may precipitate manic attacks, they are not indicated in bipolar illnesses. fn remitted unipolar patients, the relapse rate is approximately halved by tricyclics, with the relapse rate on placebo being “directly proportional to the severity of the patients’ illness prior to treatment.“‘!’ The policy of using brief hospitalization, in order to concentrate resources on a correspondingly enriched outpatient program aimed at preventing relapse and maintaining the best possible ambulatory function, applies to the affective disorders as well as to schizophrenia. The need for a setting that promotes careful clinical observation and rational trial of pharmacotherapy is even greater in the

86

JAMES

E. SABIN

large number of diagnostically obscure cases of significant chronic psychopathology that are not easy to classify. The amount of attention devoted to nosology in psychiatry today testifies to a pervasive concern with defining clinical entities, in order to provide a basis for prognosis and treatment planning. A few examples will suffice. Phobic anxiety syndrome with panic attacks, which may present a wide range of severe symptomatoiogy including alcoholism and drug abuse, has proved responsive to imipramine. 20Lithium carbonate helps reduce the emotional lability seen in what Rifkin and coworkers*’ call “emotionally unstable character disorder.” In a discussion subtitled “PsychopharmacoIogic Treatment Approaches to the Undiagnosed Case,” Klein and Shader” delineate a monoamine oxidase inhibitor responsive syndrome-“rejection sensitive dysphoria.” For purposes of planning a continuing care program, it suffices to recognize the extent to which efforts are being made to define syndromes responsive to specific pharmacologic treatments, and the consequent need for the kind of setting that encourages the necessary form of observation and collaborative inquiry between doctor and patient. The efficacy of a drug therapy program cannot be evaluated unless the physician knows what agents (prescribed and unprescribed) the patient is actually taking. But everything we learn about what is pejoratively referred to as “drug defaulting” or “patient compliance” suggests that we should not be complacent on this point, For a multitude of studies (well summarized by Blackwell”“) indicate that 25%-50% of outpatients follow a regimen significantly different than the one prescribed by their physicians. While no class of medication has privileged status in the realm of compliance, Hulka and colleaguesXa compared a broad group of “CNS medications” with cardiac and diabetic medications. All were prescribed for chronic conditions. Patients taking the CNS drugs showed the most self-initiated variance of dosage and addition of nonprescribed agents. Patients with chronic conditions of all kinds treated with long-term medication show decreased compliance.“” To add to the burden of uncertainty in a continuing care program, at least two studies show that among chronically ill psychiatric patients, the less the supervision, the less the compliance. In Great Britain, Wilcox, Gillan, and Hare2” found that inpatients, day patients, and outpatients had major deviation rates of 19%, 37% and 48%. And at the Walter Reed Hospital there was 7% deviation on a closed ward compared to 32% on an open ward and 35% in the outpatient clinic.2’ A physician’s sense of whether the patient is following “doctor’s orders” is dismayingly inaccurate. When Caron and Roth2s asked medical residents to estimate how well their hospitalized ulcer patients adhered to an antacid program, 22 of 27 overestimated antacid intake--the median estimate being 70% compared to an actual use of 46%. More im~rtant, confidence in the accuracy ofjudgment was in no way a reliable guide, in that the feeling of certitude did not correlate with actual accuracy. In another setting-a medical clinic, using tape recordings of doctorpatient interactions-Davison found that friendly rapport did not correlate with compliance. Taken together, these findings suggest that a clinician cannot rely on subjective conviction in judging whether or not a patient is following his prescription. The simplest way to assess compliance is to ask the patient if he is taking the medicine as prescribed. Studies show that patients are like physicians; both

MENTAL

ILLNESS

AND

87

THE HMO

overestimate compliance, Park and Lipman”” asked a group of outpatients to return unused pills at each visit and at the same time to report on their medication intake. Only half of the major deviatians revealed by pill count were reported verbally. And the propensity to report an intake closer to the prescribed regimen is seen to the same degree in both middle class private patients and working class clinic patients.“L An inescapable conclusion follows from the extensive literature on compliance: patient and doctor both prefer not to acknowledge their differences on how the illness should be treated. Physicians with more seniority hald more tenaciously to the masquerade of “doctor’s orders” and “patient compliance,” suggesting that as the physician becomes accfimatized to his professional roIe, he expects the patient to adhere to a reciprocal role.“” This tacit agreement to avoid confronting differences between doctor’s orders and patient’s actions often gives a haphazard quality to medical practice: “. . primary emphasis is placed on diagnosis and then prescribing a certain course of therapy for the patients” What happens from this point on has Largely been a matter of chance.““” Recent studies of the process by which patients enter treatment systems may help us transform the dismaI findings about compliance into a therapeutic opportunity. Levinson et al.‘j4 at the Massachusetts Mental Health Center and Lazare et al.“” at the Massachusetts General Hospital demonstrated that when clinicians regard the walk-in patient as a “customer” whose requests must be elicited as a prelude to “negotiated consensus” about the treatment plan, deftnable improvement in patient care and staff morate ensue. The cruciat aspect of the model delineated in these studies is a view of the patient as an active agent with ideas of his own about the objectives and process of treatment. If we approach our patients in a spirit of negotiation, we may elicit some of the concerns and attitudes that cause diminished compliance. Most prominent among these are: deniai of illness” reluctance to give symptoms of the illness:‘“-““*‘”fear of being controlled bv the physician or poisoned by the ntedication;“.“‘*“X subtIe extrapyramidal side effects that may be hard for the patient to describe;““~“’ worry about the expense of the medication, and confusion about what has been prescribed.” These issues are not always solvable, and the treatment may founder on the shoals of denial or paranoid suspicions. But the rate of noncompliance is so high that we have more to fear from not inquiring and negotiating than from relinquishing the assumption, perhaps reassuring to some patients as well as to ourselves, that what the doctor prescribes, the patient perforce follows. STRUCTURING

THE THERAPEUTK

PROGRAM

When we turn to the practical question of how an HMO should organize its continuing care program for major mental iliness there are fewer guidelines to be derived from research. The previous section documents the crucial content of such a program; the effort to identify and implement a regimen of effective pharmacotherapy. But what form should the program take? Optimal drug treatment requires an environment designed to encourage careful evaluation of the patient, negotiated consensus about the treatment plan, and monitored triais of medication. Beyond the basic objective of promoting these activities is the question of whether other modes of therapy will supplement whatever benefit pharmacotherapy provides to an individual patient. For purposes of analysis it is useful to

88

JAWfES E. SAEfN

pose questions about program design in simple terms. Should patients be seen individually~ with families, or in groups? How often should they be seen, and for how long? In an HMO, patients can be seen by generalists (primary care), speciahsts (mental health) or some combination, so it is especially important to ask who should see them. And most importantly, what processes are implied in the vague locution--“see a patient”? I have been abfe to locate only one w~li-controlled study comparing group to individual therapy in the outpatient treatment of major mental illness. O’Brien and colleagues*2 randomly assigned IO0 schizophrenic patients newly discharged from a Louisiana hospital to group or individual ~sy~hotherapy~ Rehospitalization rates did not differ at a statistically significant level for the two modes, but the trend favors group therapy (24% versus 40% for individual psychotherapy). There were, however, significant differences in social function and psychiatric symptomatology at 12 and 24 months that consistently favored the group approach. A survey of outcome in a naturalistic (nonexperimental) setting, conducted in New Haven, lends some support to the controlled study: “group treatment after discharge was associated with a good outcome (social adjustment)~ while individual treatment was not.“‘” Shattan et al_44 dissatisfied with their experience “seeing the patient in brief, stereotyped drug renewal interviews” randomly assigned 90 conditionally discharged schizophrenic patients to group and individual follow-up care. They found a decreased rehospitalization rate and increased unconditional discharge over 12 months in the group-treated patients. In only one study is individual superior to group treatment for outpatient schizophrenics. With 31 chronic schizophrenics followed for a year, Levine et al. -15found an identical rehospitalization rate (42%) for both types of psychotherapy, but a greater trend toward symptomatic improvement in individual therapy. Two studies of mixed or nonschizophrenic populations are useful for what they do not demonstrate. Herz et aL,‘6 in a treatment environment that favored individual therapy, randomiy assigned 144 aftercare patients to group or individual treatment. The program was structured so that an equal amount of therapist time per patient was provided in each mode. Covi et al.,” studying drugs and group psychotherapy in neurotic depression, imply that they anticipated group therapy would be superior to brief, individual supportive contact. By my calculation, therapist time per patient was essentially the same for both modes. Despite the potential impact of frankly acknowledged expectations, neither study found an advantage to either approach over the other. Neither was comparing psychotherapy to no psychotherapy. The comparison was between seeing patients individually and in groups. I believe that research findings to date mildly support the conclusion that group forms of treatment are associated with better patient outcome in outpatient treatment of major mental illness, especially schizophrenia. The research gives stronger support to the conclusion that group approaches are not inferior to individual treatment. While there is not a firmly established indication for group treatment, a program director can be relatively confident that a group centered approach is not contraindicated in comparison to an approach centered around one-to-one treatment. Because there may be other kinds of benefit associated with group treatment, it is important to document that the available studies of patient outcome reveal no advantage to individual treatment.

MENTAL

ILLNESS

AND THE HMO

89

The impact of group treatment on medication compliance has not been studied systematically, but several sources of evidence suggest group settings may enhance cooperation. Group support clearly has an effect on other forms of drug use: most notably the impact of Alcoholics Anonymous on alcoholism, According to Klein and Davis, “The most important reason for resistance (to ~sy~hotro~~c m~djcation) is that accepting the medication forces the patient to admit that he is sick.“‘: Groups appear to make acknowledgment of illness more tolerable. A study by Claghorn et al.“x randomly assigned 49 schizophrenic patients (on chlorpromazine or thiothixene) to group or individual drug maintenance, and over six months found that group patients shifted their self-perception toward heightened recognition of their own illness. This decrease in denial does not seem to have been d~moraIi~in~ or associated with increased resistance to treatment. Donfan, Rada, and KnighP demonstrated that chronic schizophrenics who were “refractory” and “poorly motivated’Q in a large aftercare clinic res~nded well to a nurturant. friendly group. Those assigned to group attended much more regularly than those assigned to supportive ~nd~v~dua~ therapy. Since Ruphenazine enanthate was given at the meetings, increased attendance resulted in increased medication use. Anecdotal evidence from two lithium clinics utilizing group education and waiting room discussion suggests that with manic depressives. as with schizophrenics, group participation eases the sting of admitting to an illness and facilitates use ot medication All.,~l With the exception of pharmacotherapy, what to do when seeing patients appears to be fess im~rtant than how fo f3~. A wide body of evidence suggests that the emotional climate of the therapeutic setting and the quality of human relationships in it correlate better with outcome than do the specificities of therapeutic techniques. Whitehorn and Betz‘P famous studies of what kind of therapist is most helpful to the schizophrenic patient led to the delineation of A and B configurations. For their clinical program, they established that therapists with characteristics (defined by the Strong Vocational Aptitude Test) similar to certified public accountants and lawyers were much more successful than therapists with characteristics akin to printers and mathematics/physical science teachers. However. this work, although elegant in its clear focus on characteristics of the therapist as a person, has not been consistently supported in other settings.“” Why this may be is suggested by cross-cultural studies of psychologic heating, which make e\ear that while characteristics of the healer obviously matter, the social context of the healing and the transactions occurring in the healing process are at least as important. Jerome Frank”’ identifies four features shared by alt successful psychotherapies. The patient has cantact with a concerned, caring, socially sanctioned help-giver. The treatment occurs in a place known for healing: the setting itself is therapeutic, A rationale, which helps to explain the patient’s suffering, is provided. Finally, therapeutic tasks or procedures are prescribed. This constellation of relationship, setting, rationale, and task, if successful, “relieve dysphoric feelings, rekindle the patient’s hopes, increase his sense of mastery over himself and his environment. and in general restore his morale. As a result, he becomes able to tackfe the problems he had been avoiding and to experiment with new, hetter ways of handling them.““’

90

JAMES

E. SABIN

Frank’s analysis of the common features of psychotherapeutic healing in different societies is supported by extensive research on the psychotherapeutic process itself reported by Truax and Mitchell. 55In individual and group treatment of patients ranging from relatively healthy to severely ill, three operationally defined “therapeutic conditions” appear to be associated with good outcome: genuine, nondefensive authenticity; nonpossessive warmth; accurate empathic understanding. Further research will probably lead to revision of these specific “conditions,” but the broad conclusion stated by Hans Strupp seems well supported: “To the extent that the patient is capable of resonating to the ingredients of a good human relationship, and to the extent that the therapist is able to supply these in terms meaningful to the patient, to that extent therapeutic change may be predicted to occur.“56 The exceptionally careful studies done by Hogarty and colleagues1-6 illuminate the dimension of time in continuing care. They found that sociotherapy (major role therapy) can have a beneficial impact on the community adjustment of schizophrenic patients, but only if provided for enough time. Sociotherapy did not forestall relapse in the first 6 months, but between 7 and 24 months did have a small but significant effect (44.3% receiving MRT relapsed versus 57.8% not receiving MRT). The beneficial effect of MRT on quality of life in the community (apart from relapse) was seen in the drug-treated patients. These effects were not demonstrable until 18 months and were seen still more clearly when the study ended at 24 months. Hogarty concludes: “If treatment objectives extend beyond the prevention of relapse (to maintenance of health), then our results suggest that schizophrenic patients must be continued in treatment well beyond a single year following hospital discharge (parentheses added).“6 The need to maintain treatment for an extended time is given further support by Raskin’s57 finding that 20% of a group of readmitted schizophrenics appeared to be reacting to premature discharge from aftercare or loss of the aftercare therapist. Finally, in a population of moderately depressed female outpatients who responded to amitriptyline, Weissman et alsx demonstrated that while psychotherapy did not prevent relapse, it improved overall adjustment. However, this effect was not evident until the eighth month, supporting the trend implied in Hogarty’s work. A final detail, extremely im~rtant to the organization of a continuing care program, is how frequently the patients should be seen. In Hogarty’s study of schizophrenics, the MRT patients were seen an average of twice monthly, while in Weissman’s work with depressed women the psychotherapy patients were seen weekly. Claghorn and Kinross-Wright’” randomly assigned schizophrenic patients discharged from Austin State Hospital to intermittent follow-up (approximately monthly) versus no follow-up. The control group (not followed at the clinic) had a hospitalization rate more than twice as high, suggesting the value of low-intensity aftercare. In Covi’s study,‘fi weekly group therapy and every-other-week individual therapy were comparably effective. While a dose-response curve for psychotherapy cannot be drawn, evidence suggests that weekly appointments are probably not necessary for all or even most patients. The reader familiar with clinical practice in continuing care has probably recognized that the guidelines derived from research correspond to the basic eiements of programs often described as “coffee and . . .” in the clinical literature. Beginning in 1962 with MacLeod and Middelman’s description of the Wednesday afternoon clinic at the University of Cincinnati HospitaLGo several programs have

MENTAL

ILLNESS

91

AND THE HMO

been described in adequate detail to give a picture of the clinical process.“-“” These programs vary in how much emphasis is placed on individua1 meetings with patients, In some, ail patients are seen alone at least briefly, while in others there is deliberate minimizing of one-to-one contact outside the group. AIl include an informal atmosphere in which patients and staff gather, usually with refreshments. In most, patients come on individualized schedules; there is no consistent group membership from week to week. According to these reports, the morale and pleasure of patients and staff improves with institution of the “coffee and . . .” format. A reasonable hypothesis is that the heightened benefit apparently associated with a group approach is partly a function of the improved human qualities of the treatment environment. Given Hogarty’s observation that “the helplessness experienced by patients and their families is too often equalled or surpassed by the despair of the ‘caretakers’ themselves,“~j~ the improvement in staff morale regularly seen in “coffee and. .‘% settings may allow staff to be better therapeutic persons. Whatever the specific therapeutic techniques and rationales involved, heightened morale in the therapist(s) may lead to provision of ‘“therapeutic conditions” to the patients. TREATMENT

MODEL

This overview leads to the conclusion that the single most important therapeutic variable affecting the course and outcome in major mental illness is pharmacotherapy. Not all patients with major mental illness benefit from somatic therapy, and some appear to do better without medication. But the primary requirement for success in a continuing care program is ong&zg evaluation as the foundation for rational pharmacotherapy. Recommendation of an agent is simply the first step. Monitoring its effectiveness, watching for and treating side effects, and ascertaining the degree to which the patient follows the regimen or substitutes agents of his own (alcohol, vitamins, marijuana, meditation, etc.) are all part of pharmacotherapy. Doing this in a group format appears to increase compliance. Since patients vary widely in the degree to which “psychiatry” has a positive or negative connotation, ideally both mental health and primary care should be available as bases in the HMO for the kind of group-centered care described here.“* (To avoid the expense of duplicated services, the HCHP is currently experimenting with joint sponsorship by both departments.) Broad but definable characteristics of the therapeutic setting (including the aura or reputation of the treatment program), the patient’s expectations of being helped, and the human qualities of the therapist(s), in addition to pharmacotherapy, are conducive to ciinical improvement. Insofar as the HMO has a positive valence for its enrollees, a mantle of charisma may be available to the continuing care program.“” Within the HMO, the continuing care staff should take steps to enhance the program’s reputation as a place of healing for chronic psychiatric illness. Encouraging an optimistic attitude regarding chronic psychiatric illness and respect for the continuing care program increases the overall therapeutic potency of the program. Organizing the treatment program around a group format adds to the likelihood of therapeutic gain, by at least three pathways: compliance with pharmacotherapy is probably increased in the group setting; the patient has op~rtunities for supportive interactions with other patients,“’ and the supportive aspects of groups may also improve the therapist’s morale, thereby enriching the human nourish-

92

JAMES

E. SABIN

ment available to the patients. For major mental illness, these ingredients need to be available over a long period of time (years, not months) for maximum impact. Duration of therapeutic contact over time seems to be more important than frequency of visits. Within a framework designed to be helpful in the manner common to all effective psychologic therapies, an effort should be made to identify more specific therapeutic interventions pertinent to the problems of each patient. While pharmacotherapy will be the most important component of treatment for many patients, including those who are less benefited by drug treatment, there may be focal probtems amenable to specific treatment approaches.” Phobic symptoms, sexual dysfunction, marital conflict, and psychologic crisis provoked by specific stresses all have more or less definable therapies, which may be prescribed on an individual basis. In the area of rehabilitation, welfare, and employment-related services, the HMO is at a disadvantage compared to a community mental health center (CMHC) serving a geographically limited catchment area. The CMHC is designed to draw on a network of affiliations with other agencies in working with its chronically ill population. In the HMO, indirect services, such as consultation with welfare and school departments, are hard put to compete with direct services for financial allocations. Just how much of a comprehensive treatment program for major mental illness can be provided within the HMO system, and what kinds of linkages can be established with other public and private agencies remains to be explored. ’ In an HMO, considerations of cost make it desirable to do as much continuing care as possible on an outpatient basis. The HMO operates as a comprehensive system, so unanticipated indirect effects of the continuing care program should show up fairly rapidly. If a patient is simply being maintained out of the hospital in deteriorated condition, other forms of medical utilization may increase in compensation. Neither the patient nor the system is served by shifting the cost of his care from mental health to medicine or the emergency room. A highly symptomatic patient may put excessive strain on his family, leading to increased utilization by other family members. Finally, in many HMOs there is a strong consumer role in policy formation. Modes of practice highly unacceptabIe to the clientele will lead to complaints. Because the HMO is a care system especially susceptible to marketplace influences, the continuing care program is likely to receive a high level of feedback about its direct and indirect impact. Because treatment of chronic psychiatric illness is expensive, an HMO is unlikely to offer extended benefits unless other insurers do the same. To act differently would be to price itself out of the market. iy Since Massachusetts law now requires all insurers to offer a certain minimum level of coverage for continuing care, the effectiveness and cost of a treatment model derived from clinical research and adapted to the structure of an HMO may now be tested in clinical practice. REFERENCES I. Gibson RW: Can mental health be included in the health maintenance organization? Am J Psychiatry 128:9 19, 1972 2. Reed LS, Myers ES, Scheidemandel PL:

Insurance Plans and Psychiatric Care: Utilization and Cost. Washington, D.C., American Psychiatric Association, 1972, pp 289-352 3. Davis JM: Overview: Maintenance therapy

MENTAL

ILLNESS

AND THE HMO

in psychiatry: I. Schizophrenia. Am J Psychiatry 132:1237, 1975 4. Hogarty GE, Goldberg SC, et al: Drug and sociotherapy in the aftercare of schizophrenic patients: One-year relapse rates. Arch Gen Psychiatry 2X:54. 1973 5. Hogarty GE. Schooler NR. Ulrich RF: Drug and sociotherapy in the aftercare of schizophrenic patients: II. Two-year relapse rates. Arch Gen Psychiatry 31:603, 1974 6. Hogarty GE, Goldberg SC, Schooler NR: Drug and sociotherapy in the aftercare of schizophrenic patients: 111. Adjustment of nonrelapsed patrents. Arch Gen Psychiatry 31:609. I914 7. Leff JP. Wing JK: Trial of maintenance therapy in schizophrenia, in Klein DF, GittelmanKlein R (eds): Progress in Psychiatric Drug Treatment. New York. Bruner/Mazel. 1975, pp 431 443 X. Hirsch SR. Gaind R. Rohde PD. et al: Outpatient maintenance of chronic schizophrenic patients with long-acting fluphenazine: Doublehlind placebo trial. Br Med J 1:633, 1973 9. Havens LL: Problems with the use of drugs in the psychotherapy of psychotic patients. Psychiatry 26:289, 1963 IO. Havens LL: Some difficulties in giving schizophrenic and borderline patients medication. Psychiatry 31:44. 1968 I I. May PRA: Treatment of Schizophrenia: A Comparative Study of Five Treatment Methods. New York. Science House, 1968 II. May PRA, Tuma H, Yale C, et al: Schizophrenia--a follow up study of results of treatment: II. Hospital stay over two to five years. Arch Gen Psychiatry 33:48 I, 1976 13. Greenblatt DJ. Shader RI: Psychotropic drugs in the general hospital, in Shader RI (ed): Manual of Psychiatric Therapeutics. Boston. Little. Brown, 1975. pp I I I4 14. Curry SH, Davis JM, Janowsky DS, et al: Factors alfecting chlorpromazine plasma levels in psychiatric patients. Arch Gen Psychiatry ‘2209, 1970 15, Curry SH, Marshall JHL, Davis JM. et al: Chlorpromazine plasma levels and effects. Arch Gen Psychiatry 22:289, 1970 16. Van der Velde CD: Variability in schizophrenia: Reflection of a regulatory disease. Arch Gen Psychiatry 33:489, 1976 17. Klein DF, Davis JM: Diagnosis and Drug Treatment of Psychiatric Disorders. Baltimore, Williams & Wilkins, 1969. pp I7 77 18. Davis JM: Overview: Maintenance therapy in psychiatry: II. Affective disorders. Am J Psychiatry l33:1, 1976 19. Klerman CL, DiMascio A, Weissman

93

MM. et al: Treatment of depression by drugs and psychotherapy. Am J Psychiatry I3 I: 186, 1974 20. Quitkin FM, Rifkin A. Kaplan J. et al: Phobic anxiety syndrome complicated by drug dependence and addition: A treatable form of drug abuse. Arch Gen Psychiatry 27:159. I972 21. Rifkin A, Quitkin F, Carillo C, et al: Lithium Carbonate in emotionally unstable character disorder. Arch Gen Psychiatry 77:5 19. 1972 22. Klein DF. Shader RI: The borderline state: Psychopharmacological treatment approaches to the undiagnosed care, in Shader RI (ed): Manual of Psychiatric Therapeutics. Boston, Little, Brown, 197.5, pp 284 289 23. Blackwell B: Patient compliance. New Engl J Med 289:249. 1973 24. Hulka BS. Kupper LL, Cassell JC. et al. Medication use and misuse: Physician-patient discrepancies. J Chronic Dis 78:7. I975 25. Luntz GRWN, Austin R: New stick test for P.A.S. in urine: Report on use of “phenistix” and problems of long term chemotherapy for tuberculosis. Br Med J l:l679. 1960 26. Wilcox DRC, Gillan R, Hare EH: Do psychiatric outpatients take their drugs‘? Br Med J 2:790, 1965 27. Irwin DS, Weitzel WD, Morgan DW. Phenothiazine intake and staff attitudes. .4m .I Psychiatry 127:1631. 1971 28. Caron HS, Roth HP: Patients’ cooperation with a medical regimen: Difficulties in identifying the non-cooperator. JAMA 203:922, 196X 29. Davis MS: Variations in patients’ compliance with doctors’ advice: an empirical analysis of patterns of communication. .Am .I Public Health 58:274, 1968 30. Park LC, Lipman RS: A comparison 01 patient dosage deviation reports with pill counts. Psychopharmacologia 6299, 1964 31. Rickels K. Briscoe E: Assessment of dosage deviation in outpatient research. J C’lin Pharmacol lO:l53. 1970 32. Davis MS: Variations in patients’ compliance with doctors’ orders: Analysis of congruence between survey responses and results ol empirical investigations. J Med Educ 41: 1037. 1966 33. Stewart RB. Cluff LE: A review of medication errors and compliance in ambulatory patients. Clin Pharmacol Ther 13:463, 1972 34. Levinson DJ, Merrifield J. Berg K: Becoming a patient. Arch Gen Psychiatry 17:385. 1967 35. Larare A, Eisenthal S, Wasserman L: The customer approach to patienthood: Attending to patient requests in a walk-in clinic. Arch Gen Psychiatry 32:553. 1975 36. Polatin P. Fieve RR: Patient rejection of

94

lithium carbonate prophylaxis. JAMA 218:864, 1971 37. Van Putten T: Why do patients with manic-depressive illness stop their lithium? Compr Psychiatry 16:179, 1975 38. Richards AD: Attitude and drug acceptance. Br J Psychiatry 110:46, 1964 39. Van Putten T: Why do schizophrenic patients refuse to take their drugs? Arch Gen Psychiatry 3 167, 1974 40. Rifkin A, Quitkin F, Klein DF: Akinesia: A poorly recognized drug-induced extrapyramidal behavioral disorder. Arch Gen Psychiatry 32:672, 1975 41. Mazullo J: The nonpharmacological basis of therapeutics. Clin Pharmacol Ther 13:157. 1972 42. O’Brien CP, Hamm KB, Ray BA, et al: Group versus individual treatment with schizophrenics: A controlled outcome study. Arch Gen Psychiatry 27:474, 1972 43. Schwartz CC, Myers JK, Astrachan BM: The outcome study in psychiatric evaluation research. Arch Gen Psychiatry 29:98,’ 1973 44. Shattan SP, Dcamp L, Fujii E, et al: Group psychotherapy of conditionally discharged patients in a mental health clinic. Am J Psychiatry 122:798, 1966 45. Levine H, Patterson V, Murphy BG, et al: The aftercare of schizophrenics: an evaluation of group and individual approaches. Psychiatr Q 441296, 1970 46. Herz MI, Spitzer RL, Gibbon M, et al: Individual versus group aftercare treatment. Am J Psychiatry 131:808, 1974 47. Covi L, Lipman RS, DeRogatis LR, et al: Drugs and group psychotherapy in neurotic depression. Am J Psychiatry 131:191, 1974 48. Claghorn JL, Johnstone EE, Cook TH, et al: Group therapy and maintenance treatment of schizophrenics. Arch Gen Psychiatry 31:361, 1974 49. Donlon PT, Rada RT, Knight SW: A therapeutic aftercare setting for “refractory” chronic schizophrenic patients. Am J Psychiatry 130:682, 1973 50. Bey DR, Chapman RE, Tornquist KL: A lithium clinic. Am J Psychiatry 129:468, 1972 51. Fieve RR: The lithium clinic: A new model for the delivery of psychiatric services. Am J Psychiatry 132:1018, 1975 52. Whitehorn JC: Studies of the doctor as a crucial factor for the prognosis of schizophrenic patients. Int J Sot Psychiatry 6:7l, 1960 53. May PRA: A brave new world revisited: Alphas, betas and treatment outcome. Compr Psychiatry 15:1. 1974

JAMES

E SABIN

54. Frank J: Persuasion and Healing: A Comparative Study of Psychotherapy. Baltimore. Johns Hopkins University, 1973, pp 3255330 55. Truax CB, Mitchell KM: Research on certain therapist interpersonal skills in relation to process and outcome, in Bergin AE, Garfield SL (eds): Handbook of Psychotherapy and Behavior Change. New York, John Wiley and Sons, 1971, pp 299-344 56. Strupp HH: On the technology of psychotherapy. Arch Gen Psychiatry 26~270, 1972 (Quotation, p 273) 57. Raskin M, Dyson WL: Treatment problems leading to readmissions of schizophrenic patients, Arch Gen Psychiatry 19:3X, 1968 58. Weissman MM, Klerman FL, Paykel ES: Treatment effects on the social adjustment of depressed patients. Arch Gen Psychiatry 30:771, 1974 59. Claghorn JL, Kinross-Wright J: Reduction of hospitalization of schizophrenics. Am J Psychiatry 128:344, 1971 60. MacLeod J, Middelman F: Wednesday afternoon clinic: A supportive aftercare clinic. Arch Gen Psychiatry 6:56, 1962 6 I. Draper E, Daniels RS, Rada RT: Adaptive psychotherapy: An approach toward greater precision in the treatment of the chronically disturbed. Compr Psychiatry 9:372, 1968 62. Daniels RS, Draper E, Rada RT: Training in the adaptive psychotherapies. Compr Psychiatry 9:383, 1968 63. Rada RT, Daniels RS, Draper E: An outpatient setting for treating chronically ill patients. Am J Psychiatry 126:789, 1969 64. Masnik R, Bucci L, Isenberg D, et al: “Coffee and . . .“: A way to treat the untreatable. Am J Psychiatry 128:164, I97 I 65. Fox RP: Therapeutic environments: A view of nondyadic treatment situations. Arch Gen Psychiatry 295 14, 1973 66. Isenberg PL, Mahnke MW, Shields WE: Medication groups for continuing care. Hosp Community Psychiatry 25:5 17, 1974 67. Hogarty GE, Goldberg SC, Schooler NR: Drugs and sociotherapy in the aftercare of schizophrenia: A review, in Greenblatt M (ed): Drugs in Combination With Other Therapies. New York, Grune & Stratton, 1975, pp I- I2 68. Schwartz CC, Myers JK, Astrachan BM: Psychiatric labeling and the rehabilitation of the mental patient: Implications of research findings for mental health policy. Arch Gen Psychiatry 31:329, 1974 69. Wilmer H: Transference to a medical center: A cultural dimension in healing. California Med 96: 173, 1962

RENTAL

ILLNESS

70. Caplan munity

Development. tions. 71.

G:

Mental

Support Health:

New

1974. pp Bergin

outcomes.

AND THE HMO

in

I AE:

95

Systems Lectures

York,

and on

Behavioral

Com-

Change.

Publica-

~~217-270

40

72.

The evaluation

Bergin

Handbook

Concept

AE,

of therapeutic

Garfield

SL

(eds):

nization

of

Psychotherapy

New York,

Dorsey

JL: The health

act of

group practice

and

John Wiley

1973 (P.L. plans. Medical

Behavior

and Sons.

maintenance 93-222) Care

1971, orga-

and prepaid 13: I. 1975