Psychiatrists routinely make the decision to provide involuntary treatment for certain patients by seeking a court-ordered inpatient commitment. However, for patients who are involuntarily committed in this manner, the ramifications of these decisions are tremendous, as they result in the loss of personal rights and liberty. Therefore, when making these decisions, psychiatrists generally weigh the risks and benefits of protecting society versus maximizing individual autonomy of these patients (1–5).

Though psychiatric symptoms (e.g. presence of psychotic features, suicidal ideation with a plan and intent, and/or previous suicide attempts) are the most important reasons for seeking involuntary treatment, sometimes nonpatient variables can also influence the decision to seek involuntary commitment. Perhaps the most important of these are the legal criteria in the jurisdiction of practice, which often dictate when involuntary commitment might be sought (6). In most jurisdictions, risk of harm to self or others, severity of mental illness, and ability to care for oneself provide reasonable criteria for involuntary hospitalization (7–8). In cases where the dangerousness is obvious, there is little disagreement about the need to seek involuntary commitment. However, as the assessment of dangerousness requires an interpretation of the presenting facts by the evaluating psychiatrists (9), in the absence of a clear risk of harm, psychiatrists can disagree about the need for involuntary commitment for a particular patient (10, 11).

Several factors can influence a psychiatrist’s ability to assess the dangerousness of his or her patients. While there is disagreement over the impact of the setting of practice on the decision to seek involuntary commitment (7, 12, 13), research does suggest that rather than actual patient behavior it is the perception of the dangerousness of a patient that will influence a psychiatrist to seek involuntary commitment (14). The psychiatrist’s perception of dangerousness might be influenced by the tendency to avoid, deny, or minimize a patient’s risk of violence (15). Some have suggested that a psychiatrist’s chances of seeking involuntary commitment might be influenced by the desire to control a patient’s behavior (16). On the other hand, a stable doctor-patient relationship has been suggested to minimize the chances of a psychiatrist seeking involuntary treatment for these patients (17). Some have even suggested that geographical location may influence a physician's decision to seek involuntary commitment, even if legal standards are similar (18).

Other possible reasons that might influence a psychiatrist’s decision to seek involuntary commitment may be prior exposure to a patient’s suicide and/or malpractice lawsuits. Previous exposure to a patient’s suicide might sensitize clinicians to a patient’s risk of self-harm, increasing the potential to seek involuntary commitment for other suicidal patients. Also, if some psychiatrists face malpractice lawsuits for not involuntarily hospitalizing patients who later committed suicide, they may be more likely to seek involuntary commitment when faced with a similar situation (19). On the other hand, psychiatrists who have been sued for unlawful confinement of involuntary patients may be more reluctant to seek involuntary commitment (20).

These predicaments lead some psychiatrists to practice “defensive psychiatry,” which has been described as the practice of decision-making that limits a psychiatrist’s liability while protecting patients and others from potential harm (21). However, “defensive psychiatry” does not ensure protection from lawsuits for wrongful confinement or negligent discharge. The decision to seek involuntary commitment continues to confer some professional risk to the practicing psychiatrist. One might, therefore, speculate that a psychiatrist’s attitude toward risk-taking influences the decision to seek involuntary commitment.

In this pilot study, we investigated the relationship between certain characteristics of the psychiatry residents, including their experiences with patient suicide and lawsuits along with their risk-taking behavior, and their decision to seek involuntary commitment, utilizing hypothetical decisions made in response to clinical vignette scenarios.

This pilot study was conducted in Massachusetts, where all residents (N = 208) enrolled in psychiatry training programs across the state during 1999–2000 were enlisted as subjects. Paper questionnaires were provided to these residents, inviting them to participate in this study. The questionnaire asked for demographic information, including subject age, gender, level of training, location of medical school (U.S. versus international medical graduate), and ethnic origin. The questionnaire also included five vignettes about patients who presented some risk of harm to themselves or others. Based on the information provided, residents were asked to decide whether they would seek involuntary commitment for patients in each of the five vignettes.

Further, the questionnaire asked about residents’ previous exposure to patient suicide and wrongful confinement lawsuits. Also, the residents’ risk-taking behavior was assessed using a reliable and validated scale, the Short Sensation-Seeking Scale (SSSS) (22). This scale is a shorter version of the original Sensation-Seeking Scale (19 questions) (23). Comprising 10 questions related to risk-taking behavior in social and recreational situations, the SSSS helps assess an individual’s temperamental attitudes toward taking risks. The University of Massachusetts Medical School Committee for the Protection of Human Subjects reviewed and approved this study.

Of the 208 residents contacted, 88 returned the completed questionnaires (42% response rate). Five residents returned blank surveys. Subject demographic and descriptive information is presented in Table 1. The majority of residents who responded were in their third or above post-graduate year (PGY-3) of training (60%). The majority of the respondents were Caucasian (83%) and had completed their medical education in the United States or Canada (66%). More than 90% of the respondents had never dealt with a malpractice lawsuit while 75% of the respondents had not experienced a patient’s suicide. About 13% of the respondents had experienced patient violence or homicide (Table 1).

Data were entered and analyzed on SPSS (ver. 11). Chi-square analysis between demographic variables and the residents’ prior exposure to suicides and lawsuits did not show any significant relationship with the rate of involuntary commitment sought in the five clinical vignettes. Scores from the risk-taking measure, the Short Sensation-Seeking Scale, were analyzed by t test for equality of means. The results are detailed in Table 1. A statistically significant relationship was found between:

1. Higher scores on the Sensation-Seeking Scale and fewer patients committed (p = 0.007)

2. Level of training and the total number of patients committed (p = 0.01). This relationship was rather complex, with the total number of patients committed in the five vignettes decreasing for residents in PGY-4, but increasing again for residents in PGY-5.

The results of this pilot study suggest a statistically significant association between the risk-taking potential of individual psychiatrists and their decisions not to seek involuntary commitment for the patients in the five clinical vignettes. The inherent risks that psychiatrists take when they decide not to seek involuntary commitment may be more tolerable for those psychiatrists with a greater propensity to assume risks elsewhere in their lives. We believe that this is the first study that links individual psychiatrists’ risk-taking behavior to their professional decisions regarding involuntary commitment for at-risk patients. This relationship may only be apparent when evaluating patients who do not have a clear risk of potential harm to themselves or others. Although our study showed a possible relationship between previous exposure to patient suicide and the decision to seek involuntary commitment (p = 0.060), this did not reach statistical significance. However, this may be because of small sample size. We did not find a relationship between a history of being sued and the decision to seek involuntary commitment. The very small number of subjects (N = 4) who reported a history of being sued again may have precluded the ability to identify any true statistical differences between groups.

The relationship between the level of training and the decision to seek involuntary commitment was more complex. The level of training was related to the probability of seeking involuntary commitment, although not in a simple way. In our sample, PGY-4 residents were the least likely group to seek involuntary hospitalization for patients in the five vignettes. However, for residents in their PGY-5 or beyond, the number of involuntary commitments recommended for the patients in the five vignettes actually increased. Our relatively small sample size makes generalizing these results difficult. However, this may suggest a trend in which psychiatry residents in their junior years are less prone to taking risks associated with discharging patients. This behavior may change with increasing experience, as senior residents show a willingness to assume more risk associated with not seeking involuntary commitment and discharging patients. The response of residents in their PGY-5 of training and beyond may reflect resurgence in concern for risks associated with discharging patients as residents complete their core psychiatry training and assume more of the responsibility and risk for these complex decisions. An association between decisions to seek commitment and a history of being sued was not seen in our study. This may be due to the small number of residents reporting a history of having been sued, likely in part related to the lack of experience of our subject population. Future studies examining these issues in more experienced psychiatrists could shed more light on this potential association.

From the psychiatry residency training director’s perspective, these findings may be very important. If certain nonclinical variables (whether these include previous exposure to patient violence and/or suicide or their personality traits, like risk-taking behavior or other variables that were not investigated in this study) influence psychiatry residents’ decisions to seek involuntary commitment, every effort needs to be made to minimize the inappropriate influence of these variables on the decision to prevent negative outcomes. This might be done through peer review, supervision, and, in some cases, individual counseling and psychotherapy.

This study has limitations that restrict the general application of the results and interpretation of the findings. The use of deliberately ambiguous clinical vignettes limits the data available to the clinician and thus does not reflect actual practice. In an emergent situation, the psychiatry residents should be able to ascertain further information from patients or collateral sources before making the decision to seek involuntary commitment. Further, it is possible that the respondents (42% of those contacted) were a self-selected group that was invested and interested in these difficult decisions and thus should not be considered reflective of the residents’ practice as a whole. Also, only a very limited number of variables were investigated in this study; certainly other variables, such as resident values of paternalism versus patient autonomy, might also influence the decision to seek involuntary commitment. Finally, since all subjects were psychiatric residents, it is possible that fully trained psychiatrists would have responded differently. Despite the study’s limitations, the findings reflect that decisions to seek involuntary commitment of some patients may not be based solely on clinical variables, but may be associated with a number of other variables, including those related to clinicians themselves.

More work is needed with larger samples of psychiatrists at all levels of training to investigate the impact of temperamental predispositions and other clinician-centered variables on psychiatrists’ decisions concerning involuntary commitment.