“Poor Insight”: A Capacity Perspective on Treatment Refusal in Serious Mental Illness

Consultation psychiatrists are often asked to assess the capacity of patients with serious mental illness to make decisions about the treatment of advanced somatic illness. In these situations, it becomes obvious that there are stark differences in the handling of treatment refusal for somatic versus psychiatric illness—differences that reflect and perpetuate distinctions between “medical” and “mental” illnesses. One result of these distinctions is that incapacitated patients, supported by surrogate deciders, are likely to receive the standard of care for serious somatic disorders. Meanwhile, similarly impaired patients with serious mental illness are tragically often deprived of care that might lead to lasting benefits.

This Open Forum scrutinizes this discrepancy and seeks to resolve it by applying a general medical capacity perspective to decisions made by persons with serious mental illness about their psychopathology. It begins with the presentation of a prototypical case of a patient with schizophrenia who refused treatment for a severe somatic illness.

The Case of Mr. C

Mr. C, a 55-year-old homeless man with schizophrenia and many past psychiatric hospitalizations, was “found down” and brought to an emergency room. Advanced cellulitis/osteomyelitis was discovered in his left leg. The infection had clearly been ignored for a long time. Surgeons determined that above-the-knee amputation was unavoidable and lifesaving. Mr. C refused, perseverating on the idea that his leg was “inflamed” and that “I’ll just ice it.” A psychiatric consultant concluded that Mr. C lacked the capacity to refuse amputation. Medical guardianship was obtained; the guardian consented to amputation.

Following surgical recovery, Mr. C was psychiatrically hospitalized in hopes that treatment of schizophrenia would facilitate functional recovery. Aside from slight attenuation of positive symptoms, minimal improvement was achieved. Despite repeated falls, Mr. C denied problems ambulating and was uncooperative with physical therapists. He was discharged to a locked skilled nursing facility. Some staff were pleased that he was “rescued” from homelessness (and death). Others lamented that decades of social neglect might now be compounded by fear and resentment in Mr. C over his confinement.

A Capacity Perspective on the Status Quo

Experienced inpatient and community-embedded mental health professionals might react to the late and contentious intercession of “the system” in Mr. C’s life with sadness, outrage, or demoralization. None would find it newsworthy. But the commonplace nature of Mr. C’s plight belies a long chain of complex dilemmas that were resolved reflexively or, worse, went unnoticed. Addressing those dilemmas calls for clinical and ethical engagement instead of the current risk-reduction strategy, which is more redolent of social control.

We pose four questions to analyze the assumptions, goals, and constraints with which clinicians contend when contemplating involuntary intervention for somatic versus mental illness. We recognize that mental health clinicians attempt to provide the best treatment possible under difficult circumstances. Nevertheless, we believe that a problematic philosophy generates these circumstances.

Is there a double standard?

The “four abilities” model of assessing medical decision-making capacity involves determining if the patient expresses a clear choice, understands pertinent information, appreciates the personal implications of that information, and manipulates it rationally (1). Assessments occur when a sufficiently important decision is at hand and the patient is suspected of lacking capacity to make the decision, almost always because of psychopathology (2). For Mr. C, an operative intervention for infection was imperative, and his judgment was compromised by schizophrenia (via cognitive symptoms, as is typical [3]). But with serious mental illness, the focus of treatment decisions and the source of potential incapacity are one and the same. What then?

Most patients with schizophrenia can capably make general medical decisions (4), but roughly 60% are unaware of their psychopathology (5). “Poor insight” is a standard signifier for this lack of awareness (6). It is also a euphemism for deficits in understanding and appreciation. At major decision points in somatic illnesses, inadequate understanding and appreciation suggest incapacity and a need for surrogate deciders. Capacity assessment requests in these situations are ever-rising (2).

When treatment for a serious mental illness is refused, poor insight is handled quite differently. These patients often experience a series of involuntary hospitalizations, typical of the “revolving door” that failed Mr. C. In this way, civil liberties are temporarily suspended, and considerable resources are expended for serial pauses in the ongoing deterioration of lives.

What is the purpose of involuntary intervention?

Conceptions of capacity evolved differently in psychiatry compared with the rest of medicine. Starting in the 1970s, U.S. courts decoupled commitability and incompetence in mental illness. The state’s interest when constraining personal liberty shifted from treatment to protection from immediate harm. With this change, patients are involuntarily hospitalized only when their psychopathology produces imminent danger. However, committed patients retain the right to refuse treatment (partially stemming from legal concerns over the free speech–suppressing potential of “mind-altering” medications) (7).

Distinguishing between treatment and protection as a justification for commitment accentuates the somatic-psychiatric double standard. In acute exacerbations of serious mental illness, the object of concern is not disease, as it would be in any other medical situation; it is danger. Danger can be managed through confinement without a sustainable plan for treating its cause. Thus involuntary psychiatric hospitalizations sometimes resemble protective custody more than they do clinical care. In this scenario, treatment rejection is not considered a problem because psychopathology is the object of neither medical decision making nor the state’s interest.

The right to reject treatment of an unacknowledged, harmful illness was “respected” throughout Mr. C’s adult life. But when the focus of treatment decisions shifted from psychiatric to serious somatic illness, incapacity was invoked. If a treatment-based standard held for both classes of malady, involuntary interventions that were more therapeutically meaningful than serial hospitalizations might have been attempted earlier. This type of preventive beneficence might forestall the psychosocial toxicities of impaired decision making. By contrast, a focus on safeguarding civil rights may serve to cloak the abandonment of persons with serious mental illnesses.

How do we think about “harm”?

A “dare-to-care” (6) stance, such as the one described above, forces us to reconsider what sorts of harm justify violating an individual’s civil rights. Treatment implies relieving or avoiding harm. Defining harm is thorny because the definition can hinge on one’s values, but the concept of harm bears heavily on this discussion. In general medical assessments of decision-making capacity, immediacy of harm is a key factor. In other words, fast harm (for example, orthopedic injury and myocardial infarction) has greater currency than slow harm (for example, complications of diabetes or hypertension). Furthermore, harm is considered in the context of the risk-benefit profile of a patient’s decision (1).

Apart from suicide and violence, there are few fast harms associated with chronic mental illness. Persisting homelessness, incarceration, and disenfranchisement do not occur suddenly but instead corrode lives insidiously. Even “inability to care for self,” although often handled as if it were an immediate and circumscribed event, occurs on a gradual downhill slope.

Slow harm happens largely outside the clinic’s walls. However, once severe somatic illness sets in, persons with serious mental illness cease to be invisible. Clinicians’ abilities and desire to help are activated, as in Mr. C’s case. Cynically, his “rescue” seems like false heroism—convenient and overdue. Clinically, it may reflect the providers’ relief that “finally we are permitted to act.”

Must a bias toward fast harm be the case? After all, the risk-benefit analysis of psychiatric treatment incorporates slow harm. Tardive dyskinesia and metabolic side effects, for example, are legitimate reasons for rejecting antipsychotic medications. If the slow harm of treatment is fair game, perhaps the slow harm of psychopathology can permissibly influence the risk-benefit calculus in capacity evaluations of patients with poor insight (whose own decision making erroneously omits benefits). We may need to focus more intently on trajectory than velocity of harm when thinking about capacity in serious mental illness.

Should there be a reciprocity requirement?

These considerations ought not—and, as history suggests, would not (7)—lead to wholesale incapacitation of all seriously mentally ill patients who lack insight. Aside from “feasibility constraints” (8) stemming from social resources and will, other factors, if examined honestly and with ethical clarity, could prohibit intervention.

The risk-benefit analysis used in capacity assessment presumes that there is something beneficial on offer. Minus that benefit, intervention is unjustified, and incapacity is irrelevant. Notwithstanding the state’s protection orientation, some ethicists maintain that medical violations of personal liberty must uphold a principle of reciprocity. Eastman (9) asserted that “patients’ civil liberties may not be removed for the purposes of treatment if resources for that treatment are inadequate.” If effective treatment for a patient’s psychopathology does not exist, or if societal limits are insurmountable, what medical argument supports restricting that person’s liberty?

That is not to suggest that the pursuit of those resources should be abandoned. Rather, one must be wary of applying clinical tools to social and scientific problems better addressed through activism and research. Even the use of hospitalization as social protection operates on a quarantine principle (for example, in the case of nonadherent tuberculosis patients), which is the province of public health, not clinical medicine (10).

Conclusions

Thirty-six years ago, Appelbaum and Gutheil’s classic article, “Rotting With Their Rights On,” cast treatment refusal in serious mental illness as dependent on capacity and dismantled the “doublethink” that allows simultaneous commitment and treatment refusal (11). Yet “doublethink” persists and contributes to the protracted social and physical degradation illustrated by the case of Mr. C. By the time persons with serious mental illness get “really sick”—that is, experience serious somatic illness—the psychopathology of alienation and “autistic retreat into themselves” (12) are entrenched. Earlier intervention guided by capacity-oriented principles warrants consideration.

Intervention cannot be limited to injecting antipsychotics but must entail high-quality psychosocial services (such as assertive outpatient treatment) that patients benefit from and might actually want, even under conditions of involuntary outpatient commitment (13). Consistent treatment engagement might require initial coercion. Recently revived attention to the idea of an “asylum” (14) is compatible with efforts to develop alternatives to today’s minimal and intermittent interventions. Patients in longer-term hospitalizations or those provided with assisted outpatient treatment would receive involuntary psychiatric treatment of sufficient duration to provide chances at recovery or lasting stabilization (15).

A “medicalized” capacity perspective does not relax the tension between medical beneficence and the fundamental human right to be left alone. It reveals, however, incongruities in handling psychiatric versus somatic illness decision making, hidden reasoning in philosophies of involuntary intervention and harm, and the possible use of the reciprocity principle to protect the civil rights of patients with serious mental illness. This mode of thinking might deemphasize dangerousness in favor of a renewed and revised focus on treatment that is commensurate with our role as healers for a vulnerable patient population.