Treatment Resistance in Anorexia Nervosa and the Pervasiveness of

Brief Communication

Treatment Resistance in Anorexia Nervosa and the Pervasiveness of Ethics in Clinical Decision making Chris MacDonald, PhD1 Clinical efforts to treat anorexia nervosa (AN) are constantly resisted by patients. Although the primacy of patient autonomy is a cornerstone of modern medical ethics, clinicians will nonetheless often be justified in pursuing particular interventions despite such resistance, given the reduced competency of patients suffering from this multifactorial psychiatric illness. While a literature exists on the ethical justification for imposing treatment, that literature has focused exclusively on situations in which patients refuse treatment outright. When patients resist rather than refuse treatment, clinicians are faced with the ethical challenge of deciding whether particular interventions constitute justified infringements upon patient autonomy. Given the fact that treatment resistance is endemic to AN, we see that ethical decision making must also be a continual part of the disorder’s treatment. This paper argues that the treatment of AN merely constitutes a particularly clear example of what is in fact a general phenomenon: ethical decision making pervades all clinical practice. (Can J Psychiatry 2002;47:267–270) Clinical Implications • Clinicians should be particularly careful about infringing upon the already-limited autonomy of patients with anorexia nervosa (AN). • Ethical issues are pervasive in clinical settings, and even daily, nonemergent ethical issues deserve our careful ethical consideration. • Psychiatric practice provides fertile territory for examining ethical challenges faced in all clinical settings. Limitations • The considerations presented here are preliminary; the ethics of dealing with treatment resistance has received insufficient attention in the bioethics and eating disorder literature. • More critical attention needs to be focused on the less obvious ways in which patient autonomy is limited in clinical settings. • This discussion draws upon the literature on eating disorders and bioethics. Its scope could be broadened through an examination of treatment resistance in other psychiatric illnesses.

Key Words: ethics, treatment resistance, patient autonomy, eating disorders, clinical decision making norexia nervosa (AN) is a multifactorial psychiatric disorder (Note 1) characterized by significantly diminished body weight and distorted body image that is often accompanied by denial and general cognitive impairment. More than 90% of cases occur in girls and women, and it is thought to affect 0.5% to 1.0% of females in late adolescence and early adulthood (1). Long-term mortality for AN is over 10%, with death most often resulting from starvation, suicide, or electrolyte imbalance (1).

A

W Can J Psychiatry, Vol 47, No 3, April 2002

Treatment programs for AN generally combine nutritional therapy aimed at increasing body weight and restoring electrolyte levels with psychiatric therapy or family therapy, or both, aimed at remedying underlying sources of the problem (which are still not well understood). Patients suffering from AN, however, do not want to gain weight. Indeed, they typically manifest an intense fear of weight gain, even when they are bordering on physical collapse from malnutrition. They thus typically resist treatment, in one way or another. Such 267

The Canadian Journal of Psychiatry—Brief Communication

resistance may manifest itself in several ways. These include, but certainly are not limited to, refusal to eat the medically prescribed quantity of food, removal of nasogastric (NG) tubes, and covert exercise or consumption of laxatives or emetics in an attempt to counter the effects of therapeutic nutrition. This sort of behaviour and the fact that a high percentage of patients continue to be treatment-refractory contribute to the perception that AN patients are difficult. (2). Little attention has been given to the ethical aspects of dealing with treatment resistance in the clinical management of AN. What little attention has been given to the ethics of treating this illness has focused largely on the issue of treatment refusal, rather than treatment resistance. That is, attention has focused on the question of what circumstances may justify clinicians’ forcing treatment upon a patient who categorically refuses to cooperate (3–8). Treatment refusal only leads to a genuine dilemma when the patient’s physical condition has deteriorated to the point where her life is in grave danger. The dilemma in such cases is typically whether to respect the patient’s autonomy or to impose tube-feeding aided by physical and pharmacologic restraints. Such cases are indeed tragic and worthy of ethical analysis. But such cases are also quite rare. Treatment resistance, on the other hand, is quite common. Such being the case, it seems clear that the ethics of dealing with treatment resistance warrants increased discussion. Faced with a patient who resists treatment, clinicians may find themselves engaging in various forms of coercion, persuasion, and manipulation. Some of these practices will be subtle, and others will be overt. Even voluntary patients will have their actions limited or modified in a range of ways. For example, even when in hospital voluntarily, patients still have their lives controlled to a remarkable degree by the health care professionals responsible for their care. Among the practices in which clinicians might engage are the following: • restricting the patient’s movement within the hospital (for example, ordering her to stay in her room or on the ward) • threatening outpatients with hospitalization • demanding that an inpatient eat 100% of her meal • cajoling the patient into doing volunteer work, with the

hope that forging such social connections will benefit the patient • forbidding patients to exercise (as opposed to merely pre-

scribing bedrest) Consider, for example, the ethical issues faced by clinicians in the following 2 fictionalized cases.

Case 1 Debbie is an 18-year-old woman suffering from severe, chronic AN. She has been in and out of hospital over the last 4 years. At 164 cm, she weighs just 33 kg. Debbie is in the 268

hospital voluntarily. She asserts that the goal of this admission is merely to stabilize her electrolytes and steadfastly refuses any suggestion that she should try to gain weight while in hospital. Debbie has very reluctantly accepted feeding via an NG tube. Any weight gains she might make, however, are offset by her constant pacing in the hallways of the ward. As a result, the attending physician has prescribed total bedrest. Debbie says she gets lonely in her room, however, and is often seen walking up and down the halls, chatting with other patients. When challenged, she returns to her room, only to emerge again as soon as the nurse’s back is turned.

Case 2 Alison is a 32-year-old woman with chronic AN. Over the last 10 years, Alison has been hospitalized many times for her eating disorder, but she is currently being treated as an outpatient. Measurements taken last week indicated that Alison’s body fat was at 12%. This week it is at 11%. Her community care worker tells the attending physician that Alison has shown signs of suicidal ideation. Alison desperately wants to avoid another hospitalization. The clinic’s nurse thinks that Alison’s cognitive state seems to be deteriorating, but the team psychiatrist doubts that her state has deteriorated sufficiently to warrant committing Alison under the Mental Health Act. The team decides to “make clear to Alison” that their worries about her might lead to forced hospitalization “if she doesn’t soon show signs that outpatient treatment is working.” Dealing with cases such as these requires that clinicians balance the desire to do good for the patient against the desire to respect the patient’s autonomy. The primacy of respect for patient autonomy is a cornerstone tenet of modern medical ethics. Autonomy is understood to be the ability to direct one’s own life to make one’s own decisions. It is generally seen as having 2 components: control of one’s actions (that is, the absence of constraint) and the capacity for rational deliberation. The fully autonomous agent is both free to act and has the capacity to deliberate rationally about those actions. Individuals suffering from (severe) AN have diminished capacity in both of these regards. Given the compulsive nature of their disorder, they tend to lack control over certain aspects of their behaviour; namely, their behaviour with regard to food and exercise. Secondly, individuals suffering from AN typically lack the capacity for rational deliberation concerning the effect of their caloric intake. Such individuals thus may lack both of the characteristics necessary for autonomous action with regard to food and exercise (Note 2). Infringement of autonomy is always ethically worrisome, but it is particularly so in certain kinds of situations. First, because of the notorious power imbalance between clinicians and their patients, it is a special worry in clinical settings. Second, infringement of autonomy is of particular concern with regard to W Can J Psychiatry, Vol 47, No 3, April 2002

Treatment Resistance in Anorexia Nervosa and the Pervasiveness of Ethics in Clinical Decision making

those whose autonomy is already seriously compromised. Individuals being treated for AN fall into both of these categories. Thus, special care is warranted when taking any action that abridges the liberties of a patient with an eating disorder (ED). When are we justified in interfering with a patient’s autonomy, and how far are we justified in doing so? Let us look now at what might justify infringement upon a patient’s autonomy in particular situations. In some cases, treating a patient despite resistance might be justified on the basis of consent. Most patients in ED programs are there more or less voluntarily. That is, most patients want clinical intervention, even though they may disagree with clinicians over the specific goals of that intervention (for example, weight gain vs palliation, or body fat increase vs electrolyte rebalance). When a patient has voluntarily entered a treatment program and knows what such treatment will entail, she is in effect consenting to a certain amount of infringement upon her autonomy. Given this general argument, some questions must be raised in particular situations. First, just how voluntary is the patient’s participation in treatment? Has her family (perhaps feeling exhausted and helpless) exerted undue pressure? Has she been threatened with civil commitment if she does not enter treatment voluntarily? The amount that can be justified based on consent is proportional to the freedom with which that consent was given. Treating a patient despite resistance might also be justified simply on the basis of the good of the patient. The duty to do good—the duty of beneficence—is central to the ethos of all the healing professions. For patients suffering from AN, treatment promises both physical and psychological benefits. If the treatment succeeds in inducing weight gain, the patient will be healthier physically (for example, she will have less risk of heart attack) and mentally (she will have less starvation-induced cognitive impairment). Further, the patient stands to benefit emotionally, both as a result of improved cognition and as a result of ameliorated relationships with friends and family. It is also quite likely that the patient’s autonomy will itself be enhanced by treatment. If psychiatric treatment is successful, the patient may come to be governed less by her compulsions. Similarly, if nutritional therapy succeeds in reducing cognitive impairment, she will be better able to think rationally and make her own decisions. Yet, because of the primacy of respect for autonomy in modern health care ethics, the simple fact that a proposed treatment stands to benefit the patient is not by itself sufficient to justify imposing that treatment. The desire to do good must be balanced against the need to respect autonomy. In general, the principle of respect for autonomy says that autonomous actions should not be subjected to controlling constraints by


others (9). However, to the extent to which a particular patient’s actions are not autonomous in the sense described above, we may be justified in interfering with her actions—for example, imposing treatment for her own good. According to the principle of weak paternalism, an agent may intervene “on grounds of beneficence or nonmaleficence only to prevent substantially nonvoluntary conducts that is, to protect persons against their own substantially non-autonomous actions” (9). Again, however, a principle of proportionality must be applied. Is intervention justified only as far as we know what will help? Or is our intention to help sufficient justification? AN is a poorly understood illness: its cause and cure are unknown. Thus, clinicians are often less than certain that a particular treatment or intervention will be of benefit. It might fairly be argued that we can impose treatment only when we have good reason to believe that a particular treatment will benefit a particular patient. Further, it should go without saying that when several equally effective options are available we should, other things being equal, seek to use the option that involves the least infringement upon the patient’s autonomy. Moreover, there are contextual features that must be taken into account in deciding whether to infringe upon a particular patient’s autonomy in a particular situation. First, what is the nature of the relationship between this patient and this clinician? Have they just begun working together, or has this patient been in this clinician’s care for years? Some long-term clinical relationships are such that the patient may actually expect the clinician to cajole and badger when the need arises. It is crucial to ask what the effect of a particular restriction or imposition will be upon the therapeutic alliance between the patient and the clinician. In a particular case, will a proposed action, or the patient’s awareness of it, erode the professional–patient relationship (10)? Given that treatment resistance (as opposed to outright treatment refusal) is endemic to EDs, and given that dealing with treatment resistance requires that ethical decisions be made about whether, or to what extent, to infringe upon the patient’s autonomy, it clearly follows that ethical decision making pervades the treatment of EDs. In treating individuals with EDs, clinicians must decide 100 times a day whether a certain treatment, limitation, or requirement constitutes a justified infringement upon the patient’s right to self-determination. More generally, clinicians must decide the extent to which they are justified in infringing upon the patient’s autonomy, given the particular patient’s degree of illness and readiness to accept treatment, and given careful consideration of the ethical factors discussed above. Because all patients suffering from EDs generally resist treatment at some level, we must think of all clinical interaction with such patients as ethically significant.

269

The Canadian Journal of Psychiatry—Brief Communication

It will be clear to many by now that what I have said about the centrality of ethics in clinical practice is by no means limited to the treatment of EDs. The treatment of AN merely constitutes a particularly clear example of what is in fact a general phenomenon: ethical decision making pervades all clinical practice. Patients of all kinds in inpatient settings have their actions restricted justifiably or unjustifiably in all sorts of ways. Every action that infringes upon another person’s autonomy is ethically significant, though not always ethically wrong. Thus, every instance in which a clinician takes some action that affects a patient’s ability to act autonomously must be thought of as requiring an ethical decision. More generally, any action that stands to benefit or harm somebody is ethically significant and requires an ethical decision. This description covers quite literally every action taken by health care professionals in clinical settings. Ethical decision making is not an occasional activity in clinical settings: it goes on constantly, if only implicitly. Of course, this is not to suggest that any clinician does, should, or could engage in active ethical decision making about every single action taken. Making every decision from first principles is neither desirable nor necessary. For one thing, the professional standards we so often take for granted guide much of our everyday decision making (11). Similarly, institutional policies and norms settle many ethical issues for us before they even arise. (Of course, both professional standards and institutional policies can and should be questioned from time to time.) But it is useful to remind ourselves that an ethical issue is not something that arises every few weeks in clinical settings, in those regrettable moments of crisis in which clinicians feel the need to seek advice from ethics consultants or committees. Ethical issues of an acute nature may (we hope) be rare, but ethics—the making of value judgements, of

weighing our actions against shared standards—is a task in herent to clinical life.

Notes 1. Some have disputed that AN is indeed a psychiatric illness. For the purposes of this paper, I will be following the weight of professional opinion in assuming that it is. 2. Patients with anorexia nervosa serve as a reminder that autonomy and competency, properly understood, are issue-specific rather than global. That is, patients ought not to be characterized as either “competent and autonomous” or “incompetent and non-autonomous,” but rather as more or less competent (or autonomous) with regard to particular kinds of choices.

References 1. Eating Disorders. In: Diagnostic and statistical manual of mental disorders. 4th ed. Washington (DC): American Psychiatric Assoc; 1994. 2. Kaplan AS, Garfinkel PE. difficulties in treating patients with eating disorders: a review of patient and clinical variables. Can J Psychiatry 1999;44:665–70. 3. Goldner EM, Birmingham CL, Smye V. addressing treatment refusal in anorexia nervosa: clinical, ethical, and legal considerations. In: Garner DM, Garfinkel PE, editors. Handbook of treatment for eating disorders, 2nd ed. New York: Guilford Publications; 1997. 4. Hébert PC, Weingarten MA. The ethics of forced feeding in anorexia nervosa. CMAJ 1991;144:141– 4. 5. Kluge E-H. The ethics of forced feeding in anorexia nervosa: a response to Hébert and Weingarten. CMAJ 1991;144:1121–4. 6. Goldner E. Treatment refusal in anorexia nervosa. Int J Eat Disord 1989;8:297–306. 7. Tiller J, Schmidt U, Treasure J. Compulsory treatment for anorexia nervosa: compassion or coercion? Br J Psychiatry 1993;162:679–80. 8. Draper H. Treating anorexics without consent: some reservations [editorial]. J Med Ethics 1998;24(1): 5–7. 9. Beauchamp TL and Childress JF. Principles of Biomedical Ethics 4th ed. New York: Oxford University Press; 1994. 10. Burgess MM, Hayden MR. “Patients’ rights to laboratory data: trinucleotide repeat length in Huntington disease” American Journal of Medical Genetics 1996;62: 6–9. 11. MacDonald C. Clinical Standards and the structure of professional obligation. Professional Ethics 2000;8(1): 2000. 11–7.

Manuscript received May 2001, revised, and accepted February 2002. 1 Lecturer, Dept of Philosophy, Dalhousie University, Halifax, Nova Scotia. Address for correspondence: Dr C MacDonald, Dept of Philosophy, Dalhousie University, Halifax NS B3H 4H7 e-mail: [email protected]

Résumé : La résistance au traitement dans l’anorexie mentale et l’omniprésence de l’éthique dans la prise de décisions cliniques Les efforts cliniques pour traiter l’anorexie mentale (AM) font constamment l’objet de la résistance des patients. Bien que la primauté de l’autonomie du patient soit la pierre angulaire de l’éthique médicale moderne, les cliniciens seront néanmoins souvent justifiés de poursuivre certaines interventions en particulier malgré cette résistance, étant donné la compétence réduite des patients souffrant de cette maladie psychiatrique multifactorielle. Même s’il existe une documentation sur la justification éthique de l’imposition d’un traitement, cette documentation aborde exclusivement les cas où les patients refusent carrément le traitement. Lorsque les patients résistent au traitement plutôt que de le refuser, les cliniciens font face au problème éthique de décider si des interventions particulières constituent un empiétement justifié sur l’autonomie du patient. Étant donné le fait que la résistance au traitement est endémique de l’AM, nous croyons que les prises de décisions éthiques doivent également faire partie en permanence du traitement de ce trouble. Cet article fait valoir que le traitement de l’AM ne constitue qu’un exemple particulièrement flagrant de ce qui est en fait un phénomène général : la prise de décisions éthiques est présente dans toute la pratique clinique. 270
