The issue of noncompJiance raises legal and ethical questions that extend from the one-to-one relationship of doctor and patient and reach to matters of public policy. Consent, by definition, is central to compliance. Patients who knowingly comply are consenting, or at least assenting, to treatment. Those who do not comply are withholding consent. The informed consent doctrine increasingly has shaped the process of obtaining consent and hence provides a useful framework for examining the legal and ethical questions raised by compliance and noncompliance.
The informed consent doctrine requires that a patient's consent to treatment be informed, voluntary, and competent. These requirements have been reviewed in detail elsewhere.1 This discussion will be confined to a review of the elements of informed consent as they interface with specific compliance issues.
That a patient must be informed about what he is expected to consent to is a long-established legal principie. A series of court cases in the last quarter century have outlined the necessary extent of information that must be provided to patients.2 In general, patients must be informed of lhe risks and benefits of any proposed treatment or alternatives and the consequences of no treatment. Though this would seem to be a simple enough principle, the necessity to inform patients and the role of information in the ongoing treatment of chronic patients can raise difficult clinical issues.
The issue first arises around the medication of the acutely psychotic pafieni. A fully informed consent for neuroleptics requires discussion of both acute side effects and long-term ones, ic, tardive dyskinesia. The difficulties inherent in informing grossly psychotic patients are obvious to anyone who has ever attempted it. Most authorities agree that discussions about tardive dyskinesia can be postponed until the acute disorganization has resolved.3,4 Beyond this point, however, physicians are obligated to discuss the risk of tardive dyskinesia with their patients. Failure to do so seriously disadvantages the patient and creates a liability risk for the clinician.'5 Telling patients about tardive dyskinesia flies in the face of that clinical intuition which holds that informing patients of such a serious risk will decrease compliance. This concern is very likely largely responsible for the fact that many patients, especially those viewed as poor compilers, are not being informed of the risk of tardive dyskinesia. This is ethically unacceptable and clinically unsound. There is no empirical evidence that informing patients of the risk of tardive dyskinesia decreases compliance. To the contrary, at least one study has recently demonstrated that informed consent procedures, including education about tardive dyskinesia, did not decrease neuroleptic compliance.6
The clinician's concern about this is more properly directed at that subgroup of noncompliance patients who are chronically noncompliant and who may, in fact, use information about tardive dyskincsia to rationalize their medication refusal. This group has been characterized as chronically ill, lacking in insight, consistently denying illness, and tending towards an ego syntonic grandiosity when ill, which they may actually prefer to normalcy.3,7 It is unlikely that information can explain or be responsible for noncompliance in this group since logical reasoning processes seem to have little to do with the decision to comply or not to comply.
For other potential noncompliers, discussions of side effects may actually enhance compliance in several ways. Through such discussions, patients may experience the concern of the therapist and the therapeutic alliance may grow. As patients feel more a part of the decision-making process, their own sense of responsibility may increase. Lastly, such discussions may actually heighten the clinician's awareness of his patient's side effects, leading to appropriate dosage adjustment, increasing the patient's comfort, and ultimately improving compliance. This point must be emphasized. The inadequate assessment of side effects contributes to noncompliance.
Informed consent discussions should be viewed as a two-way street in which patient and doctor receive information in a dialogue that is part of the ongoing therapeutic process. The patient learns about the medication he is being asked to take and the physician learns about the patients' concerns, including the side effects he is experiencing that may interfere with compliance. Many side effects may go unreported by the patient and unrecognized by the physician. One recent study found that tardive dyskinesia was unrecognized in 70% of cases and inadequately documented an additional 20% of the lime.8 This points to a massive omission in clinical practice. Since the discomfort of various extrapyramidal side effects may lead many patients to noncompliance,9 it is clear that the failure to address adequately the issue of side effects with patients contributes to noneompliancc.
A valid informed consent must be given voluntarily, free of coercion. The line between persuasion and coercion can be difficult to discern. The informed consent doctrine, reduced to absurdity, would have the physician merely presenting the facts, much like a weather report,10 to a patient who makes a decision by rationally weighing the evidence. This model has the physician abandoning persuasion for the sake of absolute patient autonomy. It has little to do with clinical realities and is useful mainly for dramatizing one end of the voluntariness spectrum.
At the other end of the spectrum lies coercion, the application of undue influence which robs the patient of freedom of choice. In the 1972 Kaimowitz case, a Michigan court found that a patient committed to a state hospital for an indeterminate period could not voluntarily consent to an experimental psychosurgery procedure.11 This case highlighted the degree of coercion inherent in institutionalization.
One may well ask how voluntariness can be a problem in noncompliance. If anything, noncompliant patients seem to flaunt their freedom of choice in the faces of clinicians and family members. To address this question, we must reverse the lens through which we usually look at the noncompliance problem. Our concern usually focuses on the chronic patients who do not comply; but what about the vast numbers of patients who do comply, even in the face of clear cut tardive dyskinesia? Are these patients overly compliant? Do elements of the therapeutic alliance, especially positive transference reactions, leave these patients inordinately vulnerable to persuasion and, if so, should we think of such persons as coerced?
The evidence is accumulating that psychiatrists have been prescribing higher doses of neuroleptics than is now felt to be necessary for anti-psychotic efficacy.12 The practice of chronic maintenance medication has been challenged by studies demonstrating the efficacy of time-limited courses of medication begun as soon as the patient shows the first prodromal signs of a relapse and discontinued when (he patient is stabilized.'* Enacting such a plan can be problematic and further investigation of the efficacy of such strategies is required. But they should give us pause and lead us to reexamine the way many of our chronic patients are treated. Patients treated in clinics, passed from one therapist to the next over the years, may never have their medication regimen questioned and changed. Even with patients followed more consistently, it takes a considerable effort to observe patients closely enough to target early prodromal signs and symptoms, Jt is certainly easier to simply maintain the medication if ihe patient is compliant and many patients are so treated. We may wel) be colluding in unnecessary overmedication, creating an increased risk of tardive dyskinesia to the patient and an increased risk of liability for the clinician. This should lead us to reconsider the role of our influence with the patient and to consider that, especially when doubts exist as to the proper treatment, the consistent application of persuasion with many patients threatens to cross the line to coercion.
It would be inappropriate, on the other hand, for therapists to abandon persuasion. In fact, it would be virtually impossible to act as a physician without engaging in persuasion. We must be careful to observe our persuasive efforts, avoid becoming coercive for the sake of expethence, and be aware of the power inherent in the doctor-patient relationship.
No element of informed consent has created more confusion than competency. The criteria by which competency is tested become increasingly stringent as the risks inherent in accepting or refusing treatment increase.14 In general, to demonstrate competence a patient must demonstrate understanding of his condition, of the risks and benefits of the proposed treatment, and of alternatives to treatment. The right to refuse treatment cases (Ren nie i' Klien, Rogers, etc.) which are well reviewed elsewhere,15 have clearly established that an involuntarily hospitalized patient can noi be presumed incompetent with regard to treatment decisions. Involuntary patients are presumed competent until proven otherwise. Of course, the right to refuse is a relative one, easily overriden in some jurisdictions and overriden with difficulty in others.
The clinician's response to treatment refusal is bound by legal requirements that vary from jurisdiction to jurisdiction. A variety of developments have led to an increasing tendency to respond to treatment refusals legalisticaly, for instance, by lumping all refusals into emergencies (patients who may be medicated over their objections) and non-emergencies (which require court intervention prior to medication in some jurisdictions). Many, possibly most, treatment refusals have clinical meanings that can be addressed by the staff. Often the refusal constitutes a communication by the patient that can be addressed psychotherapeutically. In addition, clinicians may fai/ to discriminate that group of refusers who arc marginally competent and whose refusal can be addressed successfully by vigorous efforts to improve the patient's decisional capacity. It is becoming increasingly clear that a large group of treatment-refusing patients who seem dc facto or clinically incompetent may respond to significant psychoeducational efforts.
Reports by Stanley10 and others indicate that efforts to teach patients that are directed at the patients' cognitive or emotional incapacities can enhance decisional capacity (and perhaps thereby turn incompetent refusers into competent consentors). Other studies show little gain from such efforts and this issue remains controversial.17 Of course, psychoeducation requires effort and this may challenge the resources of the treating institution. The balance between the time and energy devoted to addressing therapeutically a treatment refusal on the one hand and overriding it rapidly for the sake of institutional resources on the other, is another ethical challenge to our practice.
The trend toward viewing treatment refusals legalistically, that is, as based on a legal right lhat calls primarily for an evaluation of legal criteria for overriding the refusal, is simplistic and overlooks the many meanings refusais may have and the clinical strategies for dealing with them. There is growing empirical evidence that psychoeducational efforts that include the family as well as the patient are effective. The uninformed family is, in a sense, incompetent to assist in compliance efforts. The failure to address the role of the family constellation may contribute to the patient's ongoing incompetence.
The role of families raises the issue of proxy decision-making. While the role of families is often critical in compliance, it must also be recognized that the interests of cooperative family members do not always coincide with the best interests of the patient. Family members' collusion with overmedication may be as serious a problem as their collusion, in other cases, with the patients resistance to medication (Case Report, page 591).
The parents depicled in the Case Report are so disturbed by their daughter's psychosis that they are incapable of considering other issues in balancing the risks and benefits of treatment.
The final element of informed consent is the patient's actual consent. In California, written informed consent documents have been required since 1980 for the administration of antipsychotic medication. The document to be signed by the patient must include the nature of the patient's condition, the reason for taking the medication, including the likelihood of improving or not, alternative treatments, the type of dosage, likely side effects and, in particular, a statement about the risk of tardive dyskinesia.
California is presently the only jurisdiction that requires written informed consent documents for neuroleptic administration. The value of such documents either as a tool for informing patients or as a protection for clinicians is dubious at best.
There are many arguments to be made against the use of such forms. While aimed at ensuring that some informed consent procedure actually takes place, one can be too easily lulled into using the acquisition of a signature on a piece of paper as a substitute for a true informed consent procedure. The courts recognize this risk and a significant body of case law argues that such written consent forms, if not supported by additional documentation, provide little protection against liability. The unfortunate emphasis on form signing deemphasizes the importance of the informed consent discussion and the impact of this discussion on compliance.
Consent procedures should be dialogues, aimed at increasing the patient's autonomy by addressing the patients' concerns realistically, by addressing the risks and benefits realistically, and by avoiding coercion. Furthermore, consent procedures should be ongoing. Think of the countless patients in clinics who may or may not have engaged in truly informed consent before beginning neuroleptics, but who certainly have not had that process updated in years. Numerous studies point to the rapid decay of memory regarding treatment information. Both overly compliant and resistant patients should be engaged in an ongoing dialogue with regular periodic updates of informed consent (which, of course, should all be documented).
There is, of course, a large group of chronic noncompliers for whom discussions about medication seem futile. These patients, described by Appelbaum and Gutheil as "symptomatic refusers"1* often have delusional ideas about their illnesses which interfere with treatment. They demonstrate extreme negativism and denial and may prefer an ego syntonic grandiosity to the realities of the non-psychotic state.7 At times, such patients must be medicated against their will.
This brings us to the subject of the exceptions to informed consent. The first exception is the medical emergency. In virtually every jurisdiction, patients can be medicated involuntarily in emergencies. They can be medicated without their consent if they are found not competent (though with widely varying requirements in different jurisdictions). Patients can waive their right to informed consent and clinicians can sometimes act on the grounds of therapeutic priviledge. These latter exceptions are discussed below.
The involuntary administration of medication, whether during a medical emergency or otherwise, must have an impact on compliance. On the one hand, such episodes of extreme coercion may contribute to an adversarial stance between patient and clinician. For some patients it undoubtedly reinforces a "me against them" viewpoint. On the other hand, clinical experience shows that a significant subgroup of patients have a positive attitude about being involuntarily medicated by the time they are ready for discharge from the hospital. These patients are more likely to be the anxious and depressed, ego dystonic psychotics who are refusing in a moment of ambivalence or fear or as a communication to the therapist. Support for this observation can be found in a few studies that have found a positive attitude toward involuntary hospitalization by the time a patient has responded to treatment.19·20 Further research on the attitude of patients toward involuntary treatment is important in order to provide empirically-based arguments in the ongoing debate about the right to refuse treatment.
Two other exceptions to informed consent frequently raise ethical concerns. Compliant patients may seem to waive their right to informed eonsent by failing to ask questions. Often they seem to assent rather than actually consent to treatment. The clinician, only too happy to have the patient accept treatment, may be unwilling to "cause trouble" by pressing unwanted information on the patient and family. This is a trap for patient and physician. A patient does not waive his right to information merely by failing to ask for it. Rather, he must make an explicit statement such as "1 don't want to decide myself, doctor. I trust you, you decide for me."
If a clinician suspects a patient is truly waiving his right, he must inquire about Jt and clarify the waiver. If not, patients must be informed. Here again, some clinicians may be hesitant to provide information which the patient has not asked for out of concern that providing it will contribute to noncompliance - a supposition that runs contrary to the fledging empirical evidence, Ln the same vein, physicians can not invoke therapeutic privilege - the right to withhold information if it will do substantial harm to the patient to provide it - if that harm will evolve solely through the patients' refusal of the treatment under discussion.
PUBLIC POLICY AND COMPLIANCE
Leaving the informed consent model aside, a number of public policy issues interplay significantly with compliance. One such issue is that of outpatient commitment. Twenty-six states and the District of Columbia explicitly allow outpatient commitment, and bills are pending or being developed in other jurisdictions.21 Outpatient commitment is aimed primarily at the noncompliant patient and compliance remains the central problem in outpatient commitment. While it may be that a subgroup of carefully selected patients in need of involuntary treatment may comply with outpatient commitment - and for these patients such treatment may constitute the least restrictive involuntary alternative - many patients will not comply.
The practice of outpatient civil commitment raises a number of questions. What if a patient does not comply? Will hospitalization then be mandatory? If hospita/izatron is a required response to noncompliance then hospitalization can be seen as punishment or, at best, as behavioral modification. After all, if hospitalization were required clinically the patient would have been hospitalized to begin with. But if hospitalization is not a required outcome for noncompliance to outpatient commitment, how can we enforce treatment plans? Will noncompliance become a misdemeanor? At the same time, psychiatrists may be left holding the bag (of responsibility) for patients under their supervision (by virtue of commitment) over whom they have no effective control.
The most critical public policy issue that touches on compliance is the distribution of resources. While medication compliance is very much an issue between individuals - between patients, doctors, or other staff members, and between patients, families, or friends - it is also very much an issue influenced by this public policy question. Deinstitutionalization has left a generation of chronic mentally ill patients bereft of basic supports. Homelessness and transient connections to caretakers are inevitable concomitants of poor compliance. Deinstitutionalized homeless mentally ill individuals receive so little in the way of service or treatment that it is an irony to think of them as noncompliant. How can you be noncompliant if no significant treatment is being offered? Chronic patients who are most prone to noncompliance are almost certainly over-represented in this group. To consider the interpersonal and intrapsychic issues involved in non-compliance in this group without addressing the larger public policy issues can not help but miss the mark. Here, psychiatrists can work on the compliance issue by advocating for more public resources for the chronically mentally ill.
Another public policy issue is the inadequacy of third party psychiatric benefits. Do 21 or 30-day limits on inpatient coverage expose patients to higher neuroleptic doses in shorter periods of lime, thereby increasing the incidence of side effects and ultimately decreasing compliance? Does the pressure to discharge patients before their benefits expire decrease the time and energy available for psychoeducation with patient and family? Does this same pressure put limits on the development of the therapeutic alliance between patient and doctor? In each of these ways, limited inpatient coverage acts to discourage compliance. DRGs or any other prospective payment plan that is imposed on inpatient psychiatry will surely make matters worse. Limits on outpatient coverage, of course, have the same impact. The $250.00 annual cap that medicare imposes on outpatient psychiatric care leaves the clinician hard pressed to provide reasonable care.
Compliance issues are all the more likely to be addressed inadequately under these conditions. To address these questions of public policy, mental health clinicians will have to step outside the clinical arena and enter the political debate over the distribution of resources. The ethical obligation to promote the patient's interests requires an active advocacy best accomplished through professional organizations such as the American Psychiatric Association. Lobbying and public relations efforts that are geared toward increasing the quality and quantity of services available to patients, when successful, can only serve to diminish the external pressures on the doctor-patient relationship that contribute to noncompliance.
1. Meisel A. Rolh LH, Lidz CW: Toward a model of the legal doctrine of informed consent. Am J Psychiatry 1977; 134:285-289.
2. Lidz CW, Meise) A, Zerubavel E, et al: informed Consent: A Study of Decisionmaking in Psychiatry. New York, The Guilford Press, 1984.
3. Marder SR, Mebane A. Chicn C. et al: A comparison of patients who refuse and conseni to neuroleptic treatment. Am J Psychiatry 1983: 140:470-472.
4. Hal leek SL: Law und lhe Practice of Psychiatry, New York, Plenum Press, 1980.
5. Wellstein RM: Tardive dyskinesia and malpractice. Behavioral Sciences and the Law 19B3; 1:85-107.
6. Munetz MR. Roth LH: Informing patients about tardive dyskinesia. Arch Gen Psychiatry 1985; 42:866-871.
7. Van Putten T. Crumpton E. YaIeC: Drug refusal in schizophrenia and the wish to be crazy. Arch Gen Psychiatry 1976; 33:1443-1446.
8. Weiden PJ, Haas G. Mann JJ , et al : Underrecognition of neuroleptic movement disorders. New Research #195. Presented at the American Psychiatric Association annual meeting, Washington, DC. 1986.
9. Van Putten T: Why do schizophrenic patients refuse to take their drugs? Arch Gen Psychiatry 1974; 31:67-72.
10. Lidz CW: The weather report model of informed consent: Problem in preserving patient voluntariness. Bull Am Acad Psychiatry Law 1980; 8:152-160.
11. Kaimowiiz v. Michigan Department of Public Health. Div. No. 73-19434-AW, Circuit Court of Waync County, Michigan, 1973.
12. Baldessarini RJ: Chemotherapy in Psychiatry: Principles and Practice. Cambridge, Massachusetts, Harvard University Press, 1985.
13. Carpenter WT. Stephens JH. Rey AC, et al: Early intervention vs continuous pharmacotherapy of schizophrenia. Psychopharmacot Bull 1982; 18:21-23.
14. Drane IF: The many faces of competency. The Hastings Center Repori I985; 15:17-21.
15. Mills MJ. Yesavage )A, Gutheil TG: Continuing case law development in the right to refuse treatment. Am J Psychiatry 1983; 140:715-719.
16. Stanley B: Senile dementia and informed consent. Behavioral Sciences and the Law 1983; 1:57-71.
17. Beck JC, Staffin RD: Patient's competency to give informed consent to medication. Hosp Community Psychiatry 1986; 37:400-402.
18. Appelbaum PS, Gulheil TG: Drug refusal : A study of psychiatric inpatients. Am J Psychiatry 1980; 137:340-346.
19. Gove WR, Fain T: A comparison of voluntary and committed psychiatric patients. Arch Gen Psychiatry 1977; 34:669-676.
20. Kaiman TP: An overview of patient satisfaction with psychiatric treatment. Hosp Community Psychiatry 1983; 34:48-53.
21 . Keilitz I, Hall T: State statutes governing involuntary outpatient civil commitment. Mental and Physical Disability Law Reporter 1985; 9:378-397.