Ethics Commentary: Ethical Use of Power in High-Risk Situations

Provision of mental health care involves ethical use of power – power that exists within the healing relationship and power that is entrusted to clinicians by society. This power is directly proportionate to the vulnerability and the emotional and physical risk present in a situation. The ethical use of power is a fairly elusive concept characterized by the intent (i.e. seeking to do good and to minimize harm to individuals and affected others) and the outcome (i.e. whether it has in fact minimized suffering, preserved life, ensured safety, or enhanced well-being) of the clinician’s actions. Ethical use of power is expressed in diverse ways, such as in a subtle interpretation offered in the course of intensive psychotherapy with a person who has been traumatized or, alternatively, as in administering emergency medications and involuntarily committing a person with serious mental illness to treatment on a locked psychiatric ward. Both actions may be undertaken with integrity and faithful intent to help a suffering person and to protect others from harm—or not. And both actions may result in good outcomes for individuals and for affected members of society—or not—irrespective of original intent.

Power and Vulnerability in Therapeutic Work

At the heart of mental health care is the power to heal. This power derives from the strengths of the patient and the expertise of the clinician. With special training, knowledge, and experience, the clinician is able to alleviate, or at times, lift completely, the burden of suffering associated with mental illness. Patients and families know that, for example, the psychiatrist is the one who can prescribe potent medications to manage symptoms, offer reassurance, arrange hospitalization for safety and stabilization, and mobilize beneficial services. The power to heal in therapeutic work also derives, interestingly, from the interpersonal process between the patient and clinician, which can be among the most intimate and deep of all human relationships. In the context of psychotherapeutic work, patients share their life stories, innermost concerns, disquieting fantasies and fears, and loves and losses with their caregiver. This openness and transparency makes patients vulnerable, testing the limits of trust and interpersonal reliance.

This pairing of strength and vulnerability, trust and dependence allows for a unique form of therapeutic healing in the face of tragedy and serious suffering. Nevertheless, this same vulnerability gives clinicians the power to harm, reject, misunderstand, or exploit patients who struggle with the experience of mental illness, which itself may generate helplessness, despair, distress, and exceptional dependence on the clinician. For these reasons, it is incumbent upon the clinician to treat every interaction as having the potential to help or to harm and as being very significant in the life of the patient.

Ethical Considerations in High-Risk Situations

High-risk situations heighten the obligations of clinicians to use power responsibly to help ill individuals and to keep them and others from harm. These are situations in which dangerous behavior, threats of suicide or homicide, or grave passive neglect due to mental illness become evident. Often these situations are characterized by insufficient information, such as when a patient with a serious mental illness takes a bus into town, manifests severe symptoms and erratic, impulsive behavior, and is brought to a community mental health center for evaluation. Under such circumstances, it is possible for the clinician to make mistakes with important ramifications. On one hand, clinicians may underestimate the seriousness of the situation, for instance, thereby failing to intervene to ensure the well-being and safety of the individual and others who are affected. On the other hand, clinicians may also overinterpret and overreact to a situation, moving quickly to more aggressive interventions than may be necessary to fulfill obligations to the patient and to society faithfully. Similarly, in addition to challenges in the clinical assessment in high-risk situations, judgment errors may occur in which clinicians overvalue independence and autonomy to the point where decisionally compromised patients are permitted to take dangerous steps. Overvaluing of safety, however, may cause clinicians to usurp the rights of individuals who might be cared for adequately under less-restrictive means. In all of these cases, clinicians find themselves in binds, vulnerable to the risk of not adhering to appropriate standards for clinically and ethically sound care just as their patients are at risk for harms that range from having their rights violated to losing their lives.

Suicide, Violent Behavior, and Mental Illness

Suicide represents a serious public health burden in the United States, disproportionately affecting persons with mental illness. Suicide is consistently among the top ten causes of death in the United States; hundreds of thousands of individuals have committed suicide, and literally millions have received emergency treatment for serious attempts. About 90% of those who commit suicide have a diagnosable mental disorder, most commonly depression, often complicated by comorbid substance abuse (1). Mentally ill ethnic minority youth, elderly white men, and other special subgroups such as indebted farmers are at particularly high risk.

Violent behavior and mental illness, unlike suicidality and mental illness, are not tightly linked (2, 3). Findings of the Epidemiologic Catchment Area Study indicate, for instance, that 90% of persons with mental illness are nonviolent (4). This study found that among violent individuals with mental illness, a feeling of being threatened or of losing internal control, agitation, substance abuse, and lack of treatment were all related to violent actions (2). Other empirical work has revealed that it is not the presence of delusions or hallucinations per se but command voices and beliefs that one is being controlled or threatened that precipitate violence in people with psychotic disorders (5).

Unfortunately, much of the public and, sadly, some mental health professionals are convinced that most patients with serious mental illness are likely to be violent. In a 1999 survey of 1,444 people by Link et al. (6), 87% of respondents believed that violence was likely in a person who showed symptoms of illegal drug abuse, 61% thought it likely in someone with schizophrenia, and 33% in a person with depression. Two-thirds of respondents said they would use legal means to force people with substance abuse into treatment, and half would use similar interventions for treating people with schizophrenia. Ninety percent responded that those who were dangerous to self or others should be forcibly treated (6).

Ironically, people with mental illness are far more likely to be the victims of violent crime than the perpetrators. Hiday et al. (7) looked at 331 patients who were involuntarily committed to psychiatric care and court ordered to outpatient commitment after discharge. The rate of criminal victimization of these individuals with more serious mental illness was two-and-a-half times that of the general population. Interestingly, the patients’ recognition of being vulnerable to crime was low–only 16% were concerned about their personal safety. Factors that contributed to victimization were substance use, urban dwelling, unstable housing, and personality disorder (7). A subsequent study demonstrated that outpatient commitment reduced the rate of criminal victimization, substance abuse, and violent incidents, primarily through medication adherence (8).

Ethical Use of Power in Situations Involving Potential for Self-Harm and Harm to Others

Four central ethical issues surround the ethical use of power in relation to high-risk situations involving the potential for suicidal and violent behavior: prediction, duty to intervene, duty to warn and duty to protect, and strengths and accountability of persons with mental illness. Each will be discussed below, and Table 1 provides several issues to think through in these contexts.

Prediction

The first consideration pertains to the challenges in predicting suicidal and violent behavior. Accurate prediction of self-harm and violent behavior is extraordinarily difficult. With respect to suicide and parasuicidal behavior, one can be guided by past patterns of behavior and by a constellation of traditional risk factors (e.g. male gender, being unmarried and without children) and newly recognized risk factors (e.g. agitation and hopelessness) (9). Hall and Platt (10) reviewed risk factors for 100 patients who made serious suicide attempts, for example, finding that severe anxiety; panic attacks; depressed mood; major affective disorder; loss of a relationship; recent substance use; feelings of helplessness, hopelessness, or worthlessness; insomnia; anhedonia; inability to hold a job; and recent onset of impulsive behavior to be predictive of suicidal behavior. Presence of suicide notes were not accurate indicators in this study (10). In a comprehensive review of risk appraisal and management of violent behavior, Harris and Rice (5) found that the factors most consistently associated with violence are male gender, youth, past antisocial and violent conduct, psychopathy, substance abuse, and aggression as a child. Major mental disorder and other psychiatric distress were poor actuarial predictors of violence (5).

Thus, prediction of these high-risk behaviors is partly an issue of clinical acumen and awareness of patients’ risk factors and, certainly, also a matter of curiosity, intuition, and diligence in evaluating patients, gathering additional data, and reviewing collateral materials. However, risk assessment is inherently probabilistic, which means that expertise will never fully eliminate uncertainty. This fact is cold comfort after an at-risk patient takes a self-harmful step or acts violently. For many reasons, ranging from ethical ideals to pragmatic parameters, one simply cannot hospitalize all individuals solely on the basis of risk of possible suicide or violence at some future time (11). Consequently, the clinician has the difficult task of balancing many complex factors in fulfilling the duty of caring for people with mental illness who may enact self-harmful and violent wishes and behaviors. Building appropriate safeguards is therefore critically important.

Duty to Intervene

The second consideration surrounding the ethical use of power in relation to high-risk situations pertains to the professional obligation to intervene therapeutically in the context of severe illness. The duty to intervene represents the confluence of a medical ethical duty to help and a legal duty to act to protect vulnerable or endangered members of society. The ethics concept related to this duty is beneficence. The legal concept is Parens Patriae, literally translated as the “parental” responsibility to seek to keep an individual from harming himself or herself through active or passive means, invoking the power of the state to act.

The duty to intervene therapeutically becomes ethically complex only when the ill individual wishes to decline care. In such situations, an intentional process to preserve the rights of the ill individual is enacted, which usually involves placing a patient on an “involuntary hold” for a time period specified by state law (e.g. 24 hours in some states, up to 7 days in others), during which the physician assumes responsibility for keeping the patient safe and administering only those treatments to which the patient consents or that are absolutely necessary. This process seeks to ensure that the patient’s autonomy is encroached upon only as much as is necessary to keep him or her safe and to allow for a formal determination of appropriate treatment in the context of a legal hearing.

Criteria for keeping someone against his or her wishes for reasons of mental illness fall under the jurisdiction of each state and, accordingly, may vary considerably. The criteria ordinarily relate to several core elements that must coexist: that the presence of mental illness causes an individual to be at risk for imminent harm to self and/or others, by either active or passive means, and that the proposed intervention is believed to be beneficial and effective and is the least restrictive means of keeping the individual, or others, safe from harm. This set of criteria helps to prevent abuses of power, such as detaining a nonmentally ill person for inappropriate reasons or placing mentally ill individuals in more restrictive settings than are absolutely necessary. The emphasis on the “least restrictive means” to enhance individual liberty has led to the creation of mandatory or involuntary commitment to outpatient treatment for some individuals, an approach that has met with initial success in the treatment of addiction and comorbidity. The criteria also help to distinguish duties to intervene for reasons of mental illness from other causes. For example, if a person is purposely violent in the absence of mental illness, society mandates that he or she should not be shielded by a mental health code but, rather, should fall under the purview of the laws governing criminal behavior. The same is true if a person happens to be mentally ill but this illness is not the specific cause of the violent behavior.

Duty to Warn and Duty to Protect

The third consideration in the ethical use of power in relation to high-risk situations pertains to confidentiality and the obligation to help others who may be in danger, issues which tragically collided in the Tarasoff case. The 1974 and 1976 Tarasoff rulings in California changed the climate of psychiatric practice, mandating a duty both to warn and to protect individuals who are endangered by a potentially violent person with mental illness. In such situations, the patient’s privilege of confidentiality is overridden by the imperative to seek to preserve others’ safety (3). The standard of care in these emergency situations is to inform the endangered individual of a threat and to try to ensure his or her safety and to obtain collateral information from police officers, family members, friends, or staff of health care and social service agencies. Although clinicians clearly should make every effort to obtain the patient’s permission for these contacts, if such permission is not granted, clinicians must comply with their ethical and legal obligations.

A post-Tarasoff study found that 14% of U.S. psychiatrists had warned a third-party victim in the year preceding the survey. Forty-five percent of those who reported did so against their best clinical judgment, a figure that was much higher than that from those who had reported for other reasons, such as child abuse (12). The few studies that have investigated the effect of reporting on the therapeutic alliance have not substantiated the widespread concern that the ruling would have a detrimental effect on the therapeutic relationship (3). For example, of the 3,000 mental health professionals Givelber et al. (12) studied, 70%–80% believed that an ethical duty to override confidentiality and take action to protect a potential victim from a dangerous patient existed before the Tarasoff rulings.

Strengths and Accountability of Mentally Ill Persons

In discussions of power in the therapeutic relationship, emphasizing the potential vulnerability of the ill person is natural but may leave the impression that people with mental illness are so powerless and dependent that they have no responsibility for their actions or treatment. On one hand, psychiatric patients often possess several overlapping vulnerabilities, such as minority status, poverty, gender, homelessness, lack of education, and medical illnesses, that expand and augment the power of psychiatrists in ways that are subtle, complex, and often culturally determined. On the other hand, persons with mental illness have equal rights and responsibilities in society, although they have some additional protections as well. Furthermore, clinicians will attest to the heroic and virtuous individuals who daily live out the reality of the most severe and devastating forms of mental illness. These individuals fully understand what it means to be responsible, to be good citizens, to be compassionate, to endure a very unfair “deal” in life with great dignity. A paternalistic approach that further stigmatizes people with mental illness and inadvertently denies them of equal human and moral standing is fundamentally unjust and certainly unkind.

Ethical Use of Coercive Pressure in Mental Health Care

Intervening on the thoughts, feelings, relationships, and sometimes the liberties of persons with mental illness is, of necessity, part of their treatment because of the very nature of psychiatric disease. Ethical principles govern such intervention, such as seeking to help, to avoid harm, and to minimize the encroachment of a person’s rights. Intervention should never occur to gratify or convenience the clinician (13).

It is important to acknowledge differing perspectives on the issue of coercive pressure in mental health care. On one side of the debate are civil rights advocates, some consumer movements, and a number of psychiatrists, such as Thomas Szasz (14), who claim that any effort to treat a patient against his or her will is coercion and inherently unethical. Proponents of this position disagree among themselves about whether violence toward oneself or others is a valid criterion for involuntary psychiatric admission or whether community sanctions and the criminal justice system should deal with these threats. Many people who are against commitment and forced treatment are protesting the very real excesses of the past, when patients were warehoused for decades without due process and of the present, where in some countries individuals are institutionalized on the “grounds” of mental illness due to different political beliefs. On the other side of the debate are patient organizations like the National Alliance on Mental Illness and the majority of psychiatrists, who believe that schizophrenia, bipolar affective disorder, and depression are real neurobiological disorders that affect cognition and the expressed preferences of ill individuals. From this perspective, mental illness merits intensive treatment, as matters of beneficence and justice (15). Most proponents of commitment and forced treatment acknowledge the violation of rights in the past and the corresponding duty to protect the liberty of patients to the full extent possible. In all cases, these proponents affirm that ill individuals must be treated with dignity and respect.

Unfortunately, a darker side to the use of commitment, forced medication, restraint, and seclusion exists that is ethically unacceptable to all involved in this discussion: the abuse of power in treatment settings to demean or punish mentally ill individuals. In the rarest of cases, sociopathic clinicians may seek out roles that place them in control of vulnerable individuals (16). More commonly, poorly screened and trained staff or exhausted and demoralized clinicians may inappropriately use their power against patients who are chronically suicidal, personality disordered, psychopathic, or cognitively impaired. Clinicians who view treatment refusal as a challenge to their power may be more apt to react with anger. Realizing that patients are expressing themselves in one of the only ways left open to them in a virtually powerless situation can go a long way to eliminating the physician’s wish to punish or abandon them. In sum, the very best of clinicians will manage their occasional antipathic feelings toward patients through vigorous and honest self-scrutiny, teamwork, consultation, and proper self-care if they are to engage in the ethical treatment of vulnerable patients who are mentally ill.

Endeavoring to Use Power Ethically in Mental Health Care

The ethical use of power in high-risk situations rests on several pillars. First, the principles of Respect for Persons, Autonomy, Beneficence, Nonmaleficence, and Justice together suggest the importance of economical and judicious use of power in high-risk situations. The clinician must act in a manner that involves the minimal exertion of power in achieving a necessary aim such as safety so that the mentally ill individual’s rights are minimally encroached upon.

Second, mental health clinicians have complex obligations, therapeutically, ethically, and legally. Given these high stakes, clinicians in high-risk situations should never be completely “alone” in making tough decisions. They should seek consultation, gather advice from multidisciplinary colleagues, and intensively pursue additional information from multiple sources. They must be extraordinarily attentive to feelings of countertransference and extraordinarily diligent in seeking, synthesizing, and documenting information and making clinical judgments. Knowledge of legal and policy requirements of the setting and state are absolutely imperative. Legal and economic considerations are too often the determinants of clinical care. Clinicians who place the safety and well-being of patients and the community as their highest priority and exemplify this advocacy in their therapeutic relationships are actually less likely to be the objects of legal actions or institutional censure (17). This must be tempered by the humbling realities of the difficulty in predicting harmful behavior. These recommendations obviously cannot guarantee a beneficial clinical outcome, but they can help physicians come away from even high-risk encounters with the conviction that they have exercised power ethically in the service of the patient and the community.

Third, making every effort to work with the patient therapeutically is essential (Table 2). Treating individuals with respect, compassion, and dignity; helping patients to identify the need for care and finding, together, acceptable and safe options; and integrating duties-to-report and duties-to-warn into treatment interactions are all important strategies in this process. Many clinicians and ethicists have been concerned that they must assume a police-like or judicial role that is contrary to their purpose as patient advocates and healers. Psychotherapists in particular may feel that the trust and confidentiality crucial to effective personal change may not be possible under legal mandates and political pressure. For these reasons, individual clinicians must search their hearts and know their societally-mandated and professionally-affirmed duties to arrive at acceptable approaches to dealing with these complicated, multifaceted issues with their patients.

Suggested Readings

Anfang SA, Appelbaum PS: Twenty years after Tarasoff: reviewing the duty to protect. Harv Rev Psychiatry 1996; 4:67–76

Backlar P, Cutler DE: Ethics in Community Mental Health Care: Commonplace Concerns. New York, Springer, 2002

Pope KS, Vasquez MJ: Ethics in Psychotherapy and Counseling. San Francisco, Josey-Bass, 1998

Roberts LW, Hoop JG: Professionalism and Ethics: Q & A Self-Study Guide for Mental Health Professionals. Arlington, VA, American Psychiatric Press, Inc., 2008

Simon RI: Clinical Psychiatry and the Law, 2nd ed. Washington, DC, American Psychiatric Press, 1992

Szasz TS: Involuntary psychiatry. Univ Cincinnati Law Rev 1976; 45:347–365