“Ms. C” was a woman in her 40s with major depressive disorder who was treated over several months by a psychiatry resident in a medication-management clinic. She had several previous hospitalizations for suicidal ideation and attempts but had recently stabilized. Her course of treatment was characterized by sporadic missed appointments but ongoing phone contact. Several months into treatment, the resident became concerned after several “no-shows.” Ms. C then called the clinic complaining of depressed mood and was urged by staff to go to the emergency department. When she did not present, the mobile crisis team was called to do a welfare check, and police arrived to find Ms. C dead by suicide.
After the patient's suicide, the resident felt like “the only one” to whom this had happened. He was unsure with whom he could discuss the case and feared that he might be perceived as a “bad psychiatrist.” He felt guilty for being angry and concerned about his liability in such a tragic situation. He wondered if those quiet conversations in the hall or the unexpected pat on the back by a supervisor were really a questioning of his capabilities and capacity to cope. As co-residents became aware of the situation, they were unsure how to support their colleague; some secretly questioned the primary clinician's capabilities, and others feared that the same could happen to them.
The department arranged a meeting to process the situation, but, of necessity, the meeting focused on practical medical-legal issues. Several residents came away feeling that they knew more about the specific logistics of what to do after a patient suicide, but that their emotional reactions had not been adequately processed. The residents and training program leadership recognized the need for organized core training on coping with a patient's suicide.
National surveys have found that more than half of practicing psychiatrists and psychiatric residents have had a patient die by suicide (1, 2). Although many mental health clinicians will need to cope with a patient suicide over the course of their careers, completed suicide is not often talked about, and psychiatry training programs do not routinely provide the necessary training to cope with the many complex issues and reactions that follow a patient's suicide (3–5). Sudak (6) asserted that, “Suicide is the trauma for the mental health professional.” Like the family and friends left behind, the treating clinician is also a survivor of suicide but is often disenfranchised in that he or she is rarely acknowledged as such (1, 7–9).
A cohort of residents in the authors' training program felt committed to respond after several had experienced patient suicides. A literature review revealed that, in 1986, Lomax (10) proposed a curriculum on suicide care that included building knowledge, skills, and functional attitudes about the phases of suicide prevention, intervention, and postvention. Balon recommended, in 2007, that residency programs have an organized, and perhaps mandatory, process similar to school-based suicide postvention (11). Campbell developed an active postvention model for reaching out to survivors of suicide, beginning at the scene of the suicide, and suggested that something similar could be done with respect to suicides occurring in the hospital by reaching out to affected patients and staff (7). A 1998 survey of residency training directors found that fewer than half of psychiatric programs instruct residents on what to do if a patient should commit suicide (12). Melton and Coverdale (5) found that only 19% of chief residents reported feeling prepared for the possibility of having to manage the aftermath of a patient's suicide. The editors of Academic Psychiatry have asserted that academic psychiatrists must help residents learn how best to respond to the tragedy of suicide (13).
On the basis of this literature, we developed a curriculum on coping with patient suicide that includes two distinct components: a biennial half-day workshop on medical-legal issues and coping skills attended by all residents, and an as-needed module for individual clinicians should they experience the death of a patient by suicide. The current study describes this curriculum and reports on an evaluation of whether participation in the biennial workshop was associated with increases in knowledge about and perceived competence in coping with patient suicide.
The biennial curriculum is a half-day workshop designed to teach trainees how to manage both the logistical and emotional aspects of patient suicide. The workshop's participants include all four classes of the residency, and it has three components: 1) a large group lecture; 2) small group discussions, each led by faculty members who have experienced a patient suicide, as recommended by Lomax (10), and 3) a reconvening of the large group with a guest speaker who had experienced the loss of a relative by suicide. The overall goal of the lecture is to help learners develop a basic understanding of issues that arise when a patient completes suicide.
Topics covered include discoverability and confidentiality, risk-management and malpractice insurance, whether or not to contact the family, what to do with unpaid bills, and documentation (14–18). It also reviews relevant institutional policies of the university and hospital and provides guidelines to follow when in independent practice. Last, the lecture covers common emotional reactions of clinicians and their colleagues after a patient suicide. The goal of the small-group session is to give residents an opportunity to reflect upon and discuss their emotional reactions to the topic. The guest speaker provides a valuable perspective for trainees regarding the impact of suicide on a family member. The workshop's total time is 4 hours, presented in a single afternoon.
The as-needed curriculum, including a comprehensive handout, is implemented after a suicide has occurred. The faculty member and chief resident most associated with the clinical site of the affected trainee provide necessary information, and the primary purpose of this postvention is to focus on emotional support, institutional processes, and learning. This intervention is in addition to the responsibility of hospital and clinic organizations to objectively review the care that was provided to patients who died by suicide (i.e., through morbidity-and-mortality conferences, root-cause analyses, etc.).
This project was conducted in a large, academic psychiatry department. A total of 42 individuals participated in the workshop, including 39 residents (9 PGY-1, 12 PGY-2, 9 PGY-3, 8 PGY-4, and 1 PGY-5) and 3 faculty psychiatrists. Of the participants, 12% (5/42) reported previously having had a patient commit suicide. The University's Committee on Human Research approved the protocol, and learners gave informed consent to participate in the anonymous evaluation of the training.
Before the workshop, participants completed a questionnaire about their professional background and perceived competence in coping with patient suicide. Also, participants read a brief fictional clinical vignette about a patient who completed suicide, and they responded to a series of open-ended questions assessing their knowledge of pertinent emotional and medical-legal issues.
After the workshop, participants again answered questions about their perceived competence in responding to patient suicide. Also, they read a different fictional clinical vignette about a patient who completed suicide, and they responded to a series of open-ended questions assessing knowledge of relevant emotional and medical-legal issues.
The pretest and posttest included the two different vignettes, which were counterbalanced so that about half of the participants (17/42) had one vignette as the pretest and the other vignette as the posttest; the other participants were given the vignettes in reverse order.
+
Self-Perceptions of Competence
Before and after the training, participants rated items on a 5-point scale concerning their knowledge about the documentation needs after the suicide of a patient, their comfort with supporting a colleague who has experienced a patient suicide, and how capable they felt about actively participating in a discussion with involved clinicians after a patient suicide.
+
Content Analysis of Responses to Case Vignettes
The responses to knowledge-questions based on the vignettes were independently rated by two psychiatrists who were blind to whether they were pretests or posttests. Scoring criteria for the ratings were based on the literature on professional responses to completed suicide, including risk-management principles, documentation practices, attending to the needs of the surviving family, legal issues such as confidentiality, institutional issues such as morbidity-and-mortality reviews, and emotional processing by the clinicians. A total of 17 dimensions were rated for responses to questions about one vignette and 18 dimensions were rated for responses to questions about the other vignette. For purposes of data analysis, we computed the sum of the component dimensions rated for responses to questions about each vignette.
Interrater reliability was measured with the average intraclass correlation coefficient (ICC2). Data analyses were based on the average ratings of the two coders. Across the component dimensions of participants' responses to questions about the vignettes, the mean ICC2 was 0.85.
To evaluate changes in participants' self-assessments of competence, paired t-tests were conducted in which each participant's pre-training self-assessment was compared to his or her post-training self-assessment. A repeated-measures General Linear Model (GLM) was used to test for changes in knowledge while controlling for baseline knowledge and characteristics of the vignettes used to measure knowledge. To characterize the magnitude of the effects, Cohen's d was used; conventionally, d values of 0.20 are considered small, 0.50 as medium, and ≥0.80 as large effect sizes, respectively (19).
+
Changes in Self-Perceptions of Competence
Paired t-tests showed that, after training, participants showed large and statistically significant increases in their ratings of how knowledgeable they felt about documentation needs after the suicide of a patient, (t[1, 41]=10.76; p<0.001; Cohen's d=1.73; mean [standard deviation {SD}]=3.8 [0.9] versus 2.1 [1.0]); how comfortable they felt supporting a colleague who has experienced a patient suicide (t[1, 41]=7.11; p<0.001; Cohen's d=1.04; mean=4.1 [0.7] versus 3.2 [1.0]); and how capable they felt about actively participating in a discussion with involved clinicians after a patient suicide (t[1, 41]=6.84; p<0.004; Cohen's d=1.04; mean=4.1 [0.6] versus 3.2 [1.1]).
The analysis of changes in knowledge used a repeated-measures GLM predicting the post-test knowledge score based on 1) the baseline or pretest score; 2) a dummy variable indicating which of the vignettes was used at pretest (i.e., the order of presentation of the vignettes); and 3) a change score, defined as the difference between the post-test and pre-test score. The overall model was significant: Wilk's λ=0.45, df: [3, 39]; p<0.001. Controlling for scores on the pre-test vignette (F[1, 39]=59.79; p<0.001) and vignette order (F[1, 39]=1.57; p=0.22), participation in the workshop was associated with a statistically significant increase in knowledge about coping with patient suicide (F[1, 39]=48.66; p<0.001). The size of this effect was large (Cohen's d=1.08).
Unfortunately, many psychiatrists, psychologists, and other mental health clinicians will experience the death of a patient by suicide at some point in their professional life. Multiple authors have called for greater openness in discussing patient suicide, more education before an event, and greater emotional support after a patient suicide (1–18, 20). This article describes the development, implementation, and evaluation of a structured curriculum on coping with suicide in a large academic psychiatry department. The curriculum augments the required processes of the hospital and clinics to conduct morbidity-and-mortality conferences and root-cause analyses to examine clinical and systems aspects of care. Review of the literature revealed the importance of explicitly preparing residents for the potential impact of a patient death from suicide and of attending to individual residents' emotional needs after the fact. A similar training program could also be helpful in medicine, pediatrics, and surgical residencies, as well, because those trainees also encounter patient suicide and other unexpected deaths (20).
The results of the current study show that participation in a 4-hour workshop on building skills in coping with suicide is associated with large and statistically significant increases in knowledge about the topic. Also, participants in the program showed substantial increases in self-perceptions of confidence in addressing the emotional, clinical, and medical-legal aspects of coping with suicide.
Limitations of this study include its small sample size, the fact that it was conducted at a single institution, lack of a comparison or control group, lack of assessment of validity of the outcome measures, and the possibility that self-ratings of confidence could be affected by social desirability. Although participants showed an increase in knowledge and confidence immediately after participation in the training, future research is needed on the extent to which these positive outcomes are maintained longitudinally. Also, future study is needed to assess the impact of the as-needed component of the curriculum. To assess the generalizability of this study's findings, evaluation of the implementation of the curriculum in other residency training programs is needed.
Psychiatric trainees are often not aware of suicide's complex ramifications, and many lack relevant knowledge for responding to this difficult situation should it arise. However, specific training about this topic can yield measureable improvement in residents' knowledge and perceived ability to cope with patient suicide.
This work was presented in part at the American Association of Directors of Psychiatry Residency Training (AADPRT) Annual Meeting on March 14, 2008, as a Senior Resident talk at the UCSF Department of Psychiatry on March 21, 2008, and at the Northern California Psychiatric Society (NCPS) Annual Meeting on March 29, 2008.
None of the authors have any competing interests.