“Through a glass, darkly”: the clinical and ethical implications of Munchausen syndrome

The case discussed by DeWitt and colleagues in this issue of the Journal1 brings a number of contentious ethical dilemmas in general hospital medicine into sharp focus. In their account of an older man diagnosed with Munchausen syndrome following recurrent presentations with medically unexplained symptoms, the discussion is oriented, quite rightly, around concerns about the potential for harm to both the patient and the health system. The authors call for a system of notification of patient information between health care providers to address these two different imperatives. Such a mechanism may also help to address the lack of knowledge of mental illness and its relationship to physical health in acute care settings. Although discussion of the challenges associated with introducing electronic medical records (EMRs) for this purpose and the need for hard decisions about resource allocation is relevant, it fails to consider the more mundane, but just as critical, core issues in psychiatric ethics.

In the first instance, the acute care physician is confronted with the question of how this kind of patient’s problems could be best managed. The fundamental challenge of this man’s psychopathological behaviour is his mode of help-seeking. A patient whose psychological distress is communicated through physical symptoms implores a set of clinically determined responses, which are ultimately incapable of relieving the distress. In emergency departments, the oft-used epithet “frequent flyer” also invites a problematic relaxation of clinical vigilance, increasing the likelihood that an actual life-threatening medical problem will not be identified. In light of the tension between these two responses, the management paradigm of such a patient requires the availability of a skilled mental health clinician — either a consultation–liaison psychiatrist or, increasingly, a nurse practitioner with skills in psychosomatic medicine. The involvement of mental health care in this patient’s clinical management is based on regular, structured therapeutic contact that is independent of the patient’s distress level. The focus of intervention is to help the patient develop the capacity to cope with unexplained medical symptoms. A sophisticated understanding of such a patient’s situation would facilitate many important clinical processes, including identification and treatment of depression or anxiety, reduction of the patient’s intrapsychic tension, and development of more adaptive patterns of help-seeking. All of these interventions have the net benefit of alleviating the patient’s suffering and reducing the patient’s problematic presentations to acute care settings.

The discussion by DeWitt et al of the clinical management of their patient provides an apt framework to consider the ethical problems raised by the case. One of the background ethical themes highlighted in this case concerns the dilemmas that exist around the diagnostic act in psychiatry. In essence, the patient’s behaviour is reduced to the diagnostic labels of somatoform disorder or the factitious disorder usually referred to as Munchausen syndrome. In the setting of general hospital psychiatry, diagnoses are often made as a means of reframing a patient’s difficult behaviour in order to mediate a dispute with medical unit staff. Such “situational diagnoses”2 are intrinsic to the work of consultation–liaison psychiatrists, whose role is to minimise the deleterious impact of a patient’s psychopathological behaviour on his or her medical care. Psychiatric disorder carries a history of stigmatisation, and psychiatric diagnoses are often based on the patient’s problematic behaviour, rather than their suffering.3 The process of stigmatisation experienced by people with mental illness has broad implications. As Sartorius has argued:

The ethical quandary in this theme emerges from the manner in which psychiatric disorders discredit the patient. As Goffman noted, psychiatric diagnosis serves as a means of discrediting the discreditable.5 Psychiatric diagnoses place the patient in a social role as being in a perpetual state of violating social norms.6 Although this has an aesthetic benefit of reducing complicated human behaviour into tidy diagnostic categories, the process reinforces the generally negative attitudes of health professionals towards psychiatric patients.7 A diagnosis of Munchausen syndrome effectively constrains any health service contact into an unwelcome contact between a psychiatric problem and a medical system.

DeWitt and colleagues rightly identify the dilemmas associated with using EMRs as a means of identifying such patients more promptly. These systems are being given serious consideration in Australia as a measure to reduce errors in health care. Any EMR system would presumably have an element of patient control as a concession to the profound privacy concerns raised by its introduction. As such, the kind of sensitive mental health information that DeWitt et al see as critical to the systemic management of the kind of patient described in their case would require a means of communication within the health system that overrode the patient’s choice to withhold it. EMR systems are costly, complex and invite concerns about standards permitting exchange of clinical data, privacy and legal barriers.8 Indeed, there is anxiety about the potential harm posed by such systems to patient privacy.9 There are few empirical data on the benefits of EMRs in health care, particularly mental health care. In the specific instance of patients presenting with medically unexplained symptoms, their instantaneous identification as “somatisers” by an EMR system is likely to automatically generate a clinical impression that their presentation is a psychiatric problem inconveniently appearing in a busy emergency department. The irony here is that there are cost-effective interventions for these patients’ problems, such as structured clinical intervention,10 antidepressant medication11 and cognitive behaviour therapy.12 These treatments represent a tiny fraction of the cost of EMR systems, the main effect of which is likely to be diversion of these patients from care.

DeWitt et al’s other main concern is the possibility of iatrogenic harm to the patient from “unnecessary” investigation. Iatrogenic harm is a classic manifestation of the “doctrine of double effect” (DDE), first defined by Thomas Aquinas.13 The DDE provides exculpation for moral agents whose well intentioned acts occasion harm. Beauchamp and Childress’ landmark work in medical ethics placed the DDE in a clinical context.14 Iatrogenic harm arising from a well intentioned, beneficent act or intervention, which is proportionate to the clinical problem, serves as grounds to excuse the physician for unintended harm, even if such harm is foreseeable. If, however, a harmful effect is a means to a good effect (ie, the harmful effect is intentional), then such harm cannot be excused.14 In the presence of informed consent from a patient, post-traumatic dissection of a carotid artery during angiography is an instance that satisfies the conditions of the DDE. Refusal to thoroughly investigate the patient’s complaints, based on the intention of not reinforcing problematic help-seeking or illness behaviour, is not grounds for the DDE if that patient suffers a stroke. Indeed, it has been found that patients with mental health problems are less likely to be investigated or referred for more specialist care.15

DeWitt and colleagues’ main concern is for the resource implications of the pattern of help-seeking behaviour exhibited by their patient. Clearly the cost of excessive medical investigation with little expected benefit is anathema to clinicians and the taxpayer, but this misses an important point. Questions of resource allocation are invariably answered on a utilitarian basis and, in general terms, such determinations have face validity. In a world with no resource limitations, this patient would have access to psychiatric care in close liaison with medical teams, with a view to alleviating the distress that motivates the pattern of problematic help-seeking behaviour. Such measures are costly and present difficulties in measuring benefit. This latter point is of critical importance in questions of distributive justice in relation to mental health. Utilitarian considerations involving mental health care present a clear manifestation of the “quantification problem” of utilitarianism (ie, how to accurately measure how preferences are gratified by a particular measure). Ethicist Peter Singer has argued that different groups’ interests should be considered in a differential fashion, based on a “journey model” of life. In essence, preferences of groups who are better able to enact a fulfilled life journey should be placed above the preferences of those who cannot — this is Singer’s concept of “diminishing marginal utility”.16 An apposite example of the latter group is people with mental illness. Severe mental illness diminishes a person’s capacity to enact a fulfilled life journey and thus, in this approach, reduces their entitlement to have their preferences considered. Utilitarian-based choices, particularly about health care resources, always convey harm to someone.17 Given that there is currently no reliable means of measuring benefit in mental health settings,18 it is patients like the man in DeWitt et al’s case study who will always lose out in questions of resource allocation. In returning to the argument put forward in the case study, to cite cost savings as a justification for the kind of violation of the privacy of mental health patients that DeWitt et al argue for compounds the manifest injustice these patients already face in the health system.

The interface between physical and mental health is a philosophical problem dating from Descartes’ work in the 17th century. The legacy of Cartesian dualism has been to see psychiatric problems and medical problems as two distinct categories, with little relation to each other. DeWitt and colleagues’ case study highlights this issue in the context of a health system under pressure. In our estimation, questions of EMRs and resource allocation are only part of the ethical implications of the case. As we know, ignoring or failing to integrate mental health care in future health care planning is to invite a higher burden of morbidity, mortality and cost.19 To view the challenges of this case as being oriented around the most effective means of constraining the patient’s interaction with the health system misses this point entirely.