Psychologist suicide: Incidence, impact, and suggestions for prevention, intervention, and postvention

Abstract

Psychologist practitioners are not immune to some mental health problems, including suicidality, for which they provide services. In the aftermath of two recent psychologist suicides, the American Psychological Association's Advisory Committee on Colleague Assistance (ACCA) initiated the formation of a conjoint ad hoc committee consisting of members from ACCA, the American Psychological Association (APA) Practice Directorate, and the Section on Clinical Emergencies and Crises (Section VII of APA's Division 12) to investigate the incidence of psychologist suicide and its impact on colleagues, students or interns, patients or clients, and the profession. The committee reviewed the extant empirical literature on suicide rates for psychologists, evaluated unpublished data on psychologist suicide provided by the National Institute of Occupational Safety and Health (NIOSH), interviewed colleague survivors, reviewed published case reports of the impact of therapist suicides, and linked their findings to the literature on professional distress, impairment, and self-care. The committee concluded that there is evidence suggestive of an elevated risk of suicide for psychologists in past decades. It further concluded that there is a need for further research to confirm if there is a heightened risk of suicide for psychologists in the present day, and to determine factors that might contribute to such risk. Accounts from colleague-survivors suggest that the impact of a psychologist's suicide can affect many people including family, colleagues, students, and patients or clients. This article offers suggestions for possible preventive approaches, for intervention with potentially at-risk colleagues, and for postvention efforts in the wake of a colleague suicide.

Large surveys have indicated that psychologists are at risk for mental health problems such as depression, anxiety, substance abuse, and suicidality. In a national sample of 800 psychologists, Pope and Tabachnick (1994) found that most participants had been in therapy, and, of those, 61% reported that they had suffered at least one episode of clinical depression. Over one in four (29%) disclosed that they had felt suicidal, and nearly 4% reported having made a suicide attempt. Likewise, in a sample of over 1000 randomly sampled counseling psychologists, Gilroy, Carroll, and Murra (2002) found that 62% of respondents self-identified as depressed. Of those with depressive symptoms, 42% reported experiencing some form of suicidal ideation or behavior. Finally, in a 2009 APA Colleague Assistance and Wellness Survey, it was found that 40-60% of the responding practitioners reported at least a little disruption in professional functioning due to burnout, anxiety, or depression. Eighteen percent acknowledged that they had had suicidal ideation while dealing with personal and professional stressors or challenges (American Psychological Association, 2010).

Were we to extend this line of thought our next question might be “What is the risk of dying by suicide among psychologists?” Certainly, there have been published reports of high profile cases such as those of psychologists Lawrence Kohlberg (Walsh, 2000) and Michael Mahoney (Warren, 2007). In addition, the American Psychological Association's Advisory Committee on Colleague Assistance (ACCA) was informed of two other recent psychologist suicides including the story of a forensic psychologist who died by suicide while under criminal investigation for voyeurism (Rowe, 2007). Such reports raise concerns for psychologists and for the profession of psychology. They, in fact, led to the formation of a conjoint ad hoc committee consisting of members from ACCA, the APA Practice Directorate, and the Section on Clinical Emergencies and Crises (Section VII of APA's Division 12) to investigate whether there is evidence of an increased rate of suicide among psychologists and what the impact of the suicide of a practicing psychologist might have on colleagues, patients or clients, and the profession. This article is a report of the committee's findings which are linked to the literature on professional distress and impairment. It includes a review of the extant literature on suicide rates for psychologists, an evaluation of unpublished data on psychologist suicide in the database of the National Institute of Occupational Safety and Health (NIOSH), a report on anecdotal observations from interviews with psychologists who experienced a colleague suicide, a summary of published accounts of the impact of a therapist's suicide on his or her patients, and sections on factors that may contribute to suicide risk for psychologists, barriers to recognizing and intervening with impaired colleagues, suggestions for intervention, the importance of self-care as a preventive measure, and suggestions in responding to the suicide of a colleague (i.e., so-called ‘postvention’ efforts; Shneidman, 1993).

Literature Review on Psychologist Suicide

A literature review of suicide rates by occupation revealed that many of the available studies are marked by inconsistent methodologies and disparate findings, with demographics accounting for many observed associations between occupational groups and suicide rates (Agerbo, Gunnell, Bonde, Mortensen, & Nordentoft, 2007; Stack, 2001; Wasserman, 1992). One consistent finding, however, has been elevated suicide rates for health care professionals —physicians, dentists, and nurses— with qualitative reviews (Boxer, Burnett, & Swanson, 1995; Horton, 2006; Nock, Borges, Bromet, Cha, Kessler, & Lee, 2008) and empirical investigations (Agerbo, et al., 2007; Kelly, Charlton, & Jenkins, 1995; Lindeman, Laara, Hakko, & Lonnqvist, 1996; Meltzer, Griffiths, Brock, Rooney, & Jenkins, 2008; Schernhammer, 2005; Stack, 2001; Stark, et al., 2006) suggesting that these occupations are at elevated risk for suicide.

Given that psychologist practitioners are also health care professionals, the aforementioned literature would suggest that psychologists might also be at elevated risk for suicide. Psychologists, however, have not been included as an occupational category in the epidemiological studies referenced above. The majority of the published literature on suicide among psychologists consists of uncontrolled case reports describing patient reactions to the loss of a therapist (Ables, 1974; Ballenger, 1978; Chiles, 1974; Graves, 1978; Reynolds, Jennings, & Branson, 1997; Soreff, 1975) and commentaries (Dunne, 1987; Guy & Liaboe, 1985; Lester, 1989) rather than empirical research.

There are three notable exceptions to the statement above. Two studies examined rates of suicide for members of the American Psychological Association compared to the general population (Mausner & Steppacher, 1973; Phillips, 1999). Mausner and Steppacher (1973) used a list of psychologist deaths obtained from the APA for 1960-1969. Rates of suicide for psychologists for this time period were compared to the general population in the form of standardized mortality ratios (i.e., adjusting for age) and were reported separately for males and females. The rate of suicide among male psychologists was comparable to the rate of males in the general population; however, the rate of suicide among female psychologists was significantly elevated compared to the rate of females in the general population – nearly three times greater (i.e., SMR = 2.8), suggesting that female psychologists, at that time, might have been at elevated risk for suicide compared to women in the general population.

The second study (Phillips, 1999) was a follow-up to the APA study described above. It examined suicide rates for psychologists in a more recent decade (i.e., 1981-1990). In contrast to Mausner and Steppacher (1973), elevated suicide rates (compared to the general population, adjusted for age) were not found for male or female psychologists. In fact, the suicide rate for males was significantly lower than the rate for males in the general population at that time (i.e., 7.8 per 100,000 for psychologists and 24.9 per 100,000 for males in the general population). How the cause of death was determined in either study is unclear. Moreover, psychologists who are not APA members might have different suicide rates, raising concerns about the generalizability of the findings to the profession as a whole.

The third study noted above was conducted at the Division of Surveillance, Hazard Evaluations, and Field Studies (DSHEFS) of NIOSH, but was not published. It was mentioned, however, in a commentary on occupation and suicide (Ukens, 1995). One of the authors of the current manuscript (P. K.) contacted the principal investigator for the study, Dr. Peter Boxer, as well as NIOSH-DSHEFS. NIOSH-DSHEFS provided a copy of slides from a presentation of the study¹. Dr. Boxer and his colleagues (Boxer, Burnett, and Swanson, 1992) completed a case control study in which occupation-coded death certificate data was collected on white males (ages 20-64) who had suicide as an ICD-9 coded cause of death in 25 states during the period of 1979-88². The control group consisted of white males (ages 20-64) who had died of any natural cause during this same time period and were matched for state of residence. Using an unconditional logistic regression analysis, while controlling for age, marital status, socioeconomic status, and contributory mental disorders, the authors found a significantly elevated odds ratio of 3.47 for male psychologists, while male pharmacists, male physicians, and male dentists had significantly elevated odds ratios of 3.35, 2.88, and 1.7 respectively.

Although the study by Boxer, et al., (1992) seemed to indicate that the risk of suicide was elevated for white male psychologists, at least during that particular time period, detailed information about the control group was no longer available to determine if it was fully representative of the occupations in the study group. Therefore, the findings should be viewed with great caution.

As a result of the authors' inquiry to NIOSH, however, NIOSH-DSHEFS accessed its National Occupational Mortality Surveillance (NOMS) database and found that 28 states had participated in a NIOSH project on cause of death by occupation/industry for two or more years from 1984 through 1998 (National Occupational Mortality Surveillance System, 2010)³. Cause of death was coded according to ICD-9 codes. Proportionate Mortality Ratios (PMRs) were calculated for all occupations, psychologists included. The PMR for suicide for psychologists, aggregated across all ages, both genders, and all races, was 166 (95% CI [139, 197], p<.01) indicating an increased risk of suicide⁴. PMRs were also calculated for psychologists by gender and race. White male psychologists (ages 15-120) were found to have an elevated risk for suicide (PMR=161; 95% CI [128, 199], p<.01) as were white female psychologists (ages 15-120) (PMR=212; 95% CI [154, 285], p<.01). In contrast, black male psychologists and black female psychologists were not found to be at elevated risk.

In sum, the literature on the question of whether psychologists have had an elevated rate of suicide provides some suggestive, albeit conflicting and flawed, findings, with the data and analyses provided by NIOSH offering the strongest evidence to date. There is a clear need for systematic research to determine if this risk continues for present day psychologists. In addition, research is needed to examine whether suicide rates may differ if patient or client care is a professional responsibility (as, e.g., with clinical, counseling, school, and industrial/organizational psychologists) or not (as, e.g., with academic or research psychologists).

Finally, it should be noted that under-reporting and under-detection of depression and suicidal behavior affects all research on suicide (Nock et al., 2008), and, thus, also applies to estimates of psychologist suicide. Furthermore, due to the stigma associated with suicide (Sudak, Maxim, & Carpenter, 2008), when there is doubt, medical examiners may be motivated to report a cause of death other than suicide.

Interviews on the Impact of Suicide on Colleague Survivors

Although there are reports of the impact of a therapist's suicide in the literature (as noted above), the committee, in order to better inform itself, also sought interviews with a small sample of psychologists who had experienced a colleague suicide. An inquiry was sent to several listserv (i.e., listserv to which one or more of the authors had access), specifically the State Psychological Associations' Colleague Assistance Program representatives, the California Psychological Association, APA Division 12, the Division 12 Section on Clinical Emergencies and Crises, and the Association of Behavioral and Cognitive Therapy, requesting a telephone interview with psychologists who knew of a colleague's suicide and who were willing to discuss it and its impact on self and others. Two of the committee members (M.H. and P.K.) developed a semi-structured interview and contacted the psychologists who volunteered. Through this method, it was possible to find 14 cases of psychologists who reportedly had killed themselves. Although we offer some reflections on these interviews here, it is clear that reflections on anecdotal data cannot be generalized to other cases of psychologist suicide. They may, however, generate interest in more systematic future investigations.

While psychologist suicides were identified, this fact per se does not imply a heightened rate of suicide among psychologists. What was impressive in these interviews, however, was the extent of the reported emotional impact of a psychologist suicide on family, on colleagues, on patients or clients, on students or interns, and on staff; and there is reason to believe that there could be an even wider impact. For instance, highly publicized cases are likely to create a negative impression of a profession that attempts to promote mental health.

More than half of the individuals who were said to have died by suicide were thought to have had problems with depression or substance abuse. An interpersonal, professional, or functional loss preceded the suicide for many. Only one was known to have problems of an ethical and legal nature. To borrow a phrase from O'Connor (2001), these psychologists who died by suicide seemed more likely to be troubled (i.e., distressed or impaired) than in trouble.

Consistent with data reported by Hendin, Lipschitz, Maltsberger, Haas, & Wynecoop (2000), participants frequently stated that they were “stunned” or “shocked” at the news of a colleague's suicide. None said that they had foreseen suicide as a possibility. Reasons for this are not known, but several studies have found that psychologists frequently withhold important clinical information from their therapist (e.g., Pope &Tabachnick, 1994) and perhaps also withhold such information from colleagues. For instance, in the APA survey referenced earlier, 14% of psychologists who reported suicidal ideation further indicated that they did not tell anyone about it, including their therapist.

Most of our interviewees reported that their colleague's suicide had a strong personal impact on them. For a few, the impact was reported as lasting for 1-2 years as they struggled to make sense of, and cope with, the loss. A majority said that the impact had also affected a wider circle of colleagues. One interviewee was a graduate student of a psychologist who took his own life. She reported that her mentor's death led her and a number of fellow graduate students to question their career choice.

Case Reports of the Impact of Therapist Suicide on Patients

A number of case reports about the impact of a therapist's suicide on his or her patients were published in the 1970s and were previously reviewed by Dunne (1987). There has been, for example, a report of the immediate impact of a therapist's suicide on a 10 year old boy (Ables, 1974) as well as a report of the impact of a therapist's suicide on two adolescent patients who were in a hospital at the time (Graves, 1978).

There have been at least three articles in which the reactions of a number of adult patients who lost a therapist to suicide were reported. In the first, Chiles (1974) contacted five of the seven former patients of a deceased female psychiatrist one year after her death. In this particular case, all five of these patients reported being aware of the fact that their therapist had psychological problems prior to her suicide. In fact, one patient stated that she frequently sought support from him and, at times, he felt as though he was treating her. A few focal concerns emerged for these patients. One was a concern about being responsible for her death, a second had to do with concern about the fallibility of the therapist (who could not help herself), and a third had to do with ambivalence toward, or rejection of, re-engaging in therapy with another therapist.

In a second article, Ballenger (1978) interviewed the five patients who were in psychotherapy with a psychiatrist who killed herself. These patients also reacted with feelings of shock, loss, abandonment, and distrust. At least two of these patients reported that they had sensed their therapist's depression prior to her death. Three of them wondered if they had somehow been to blame. They also questioned the value of their therapy, expressed a sense of demoralization, and pondered whether there was any hope for them if this could happen to a psychiatrist. Four of the five either did not engage with a new therapist or had difficulty in re-engaging in a relationship where they felt that they could trust the therapist.

In a more recent study, Reynolds, Jennings, and Branson (1997) developed a 44-item questionnaire based on the literature on grief and loss, and, one year after the event, contacted 34 former patients of a male psychologist who had died from a self-inflicted gunshot wound. Twelve (or 35.3%) of the former patients agreed to participate and completed the questionnaire. During the year after the suicide, many of the respondents reported feeling depressed, numb, angry, hopeless, and/or abandoned. They expressed a need to understand a cause or reason for the suicide. Half of them (or 17.6% of the psychologist's total patient case load) felt that they could have prevented the death had they been there at the time. Several reported behaviors that suggested dwelling on the loss. These behaviors included continuing to visit the therapist's gravesite, continuing to drive by his former office, and continuing to have episodes of crying. Denial of the cause of death also persisted with half of the sample believing that the therapist might have been murdered even though they were aware of the autopsy findings and the coroner's conclusion that the death was self-inflicted. Most rated suicide as an acceptable solution to a problem at least under certain conditions. All of the participants refused referrals to other therapists, but most said that they would be willing to seek out a therapist in the future if the need arose.

A common finding among these published case reports and the reports from our colleague interviews was that patients frequently had difficulty or refused to re-engage in therapy in the immediate aftermath of their therapist's suicide. The implications of such a tendency are not entirely clear, but perhaps there are clues in the frequent concerns about guilt for the therapist's suicide and the questions about the reliability of therapists and the value of therapy.

Factors That May Contribute to Suicide Risk for Psychologists

Knowledge and expertise in the assessment and treatment of mental disorders does not, unfortunately, foster immunity to mental disorders, nor does it ensure optimal functioning (Good, Khairallah, & Mintz, 2009). Psychologists, psychiatrists, and other mental health professionals are presumably vulnerable to the same risk factors for suicide that affect non-mental health professionals including mental disorders, social isolation, physical illness, unemployment, family conflict, hopelessness, impulsivity, and so forth (for a recent review of risk factors, see Van Orden, et al., 2010).

It is possible, however, that the professional responsibilities, the intense nature of the work, the work environment, or the characteristics of psychologists themselves are associated with a unique set of risk factors, moderators of risk factors, or barriers to care, as well as protective factors that either increase or decrease risk. Conceptual models for understanding the relations between occupation and suicide (Agerbo, et al., 2007; Horton, 2006; Stack, 2001; Wasserman, 1992) concur on the following factors as potential explanations: 1) demographic characteristics; 2) increased access to lethal means (e.g., medications for overdose); 3) job related social stressors (e.g., social isolation); and 4) self-selection processes whereby individuals with higher rates of psychiatric morbidity preferentially select certain occupations. Physician suicide in particular has been hypothesized to result, in part, from easy access to lethal means, with studies documenting that anesthesiologists disproportionately die from overdose with anesthetic agents (Hawton, Clements, Simkin, & Malmberg, 2000) and that the most common method for suicide among physicians is overdose, often with medications taken from work (Hawton, Malmberg, & Simkin, 2004). A consensus statement on depression and suicide among physicians that appeared in the Journal of the American Medical Association (Center, et al., 2003) proposed that the medical field's culture that de-emphasizes physician mental health combined with barriers to help-seeking are also likely contributors to elevated suicide risk in this occupational group.

Smith and Burton Moss (2009) have reviewed the risk factors for impairment in the functioning of psychologists. They include such characteristics of the occupation as the challenge of managing the intimate, confidential, and non-reciprocal nature of the client/therapist relationship, isolation in the work context (particularly in individual private practice), dealing with negative client or patient behaviors such as aggression and suicide potential, pressures associated with managed care, rapidly shifting role demands in institutional and clinic settings, decreased individual control over work, and increased time needed for paperwork and/or administrative duties. It is possible that such job-related stressors could create or exacerbate risk factors for suicide. Studies, however, have not examined these possibilities.

Barriers to Recognition of Impairment and Intervention

Were psychology graduate and training programs to include education about the risks of professional impairment and the benefits of wellness practices in their curriculum, it seems plausible that future psychologists would be more likely to see the signs of impending personal difficulty and be more aware of the option of seeking assistance. As Schoener (1999) has pointed out, however, psychology, as a field, has not emphasized this aspect of professional education. By way of example, Schwebel and Coster (1998) surveyed the heads of 107 APA-approved programs in professional psychology to learn their views on well-functioning in psychologists. They found a great lack of programmatic effort to institute means to prevent impairment. While the program directors had many good suggestions for ways to improve education about well-functioning and impairment prevention, they saw lack of time and space in the curriculum and lack of funding as great obstacles. In effect, they seemed to say that if change were to occur, it would need to be mandated and funded.

Floyd, Myszka, and Orr (1998) surveyed a large sample of psychologists in their state psychological association about their perceptions of colleague impairment and about whether or not they expressed concern about their colleague's perceived impairment. Of those who did not express concern, 43% felt that the perceived problems did not affect the individual's clinical work while 26% expected a negative outcome from an attempt at intervention and 22% felt uncertain about their professional responsibility. Still others reported feeling that an expression of concern involved risk to themselves or to the impaired colleague. Smith and Burton Moss (2009) have suggested that many of the barriers to intervening with an impaired colleague have to do with a lack of knowledge of how to go about it appropriately. Many psychologists state that they are uncertain about how to approach a potentially impaired colleague or about what information might be needed if patient care is being affected and it may be necessary to report him or her.

Points in Intervening with an Impaired Colleague

As O'Connor (2001) has pointed out, psychologists may naturally wish to distance themselves from a colleague who, due to distress or depression, appears to be practicing in a less than competent manner. In such circumstances, they may gravitate prematurely toward emphasizing strict adherence to a professional code but neglect the possible constructive effects of attempting an early intervention. Thus, at times, it may be possible to assist an individual by simply bringing a concern to his or her attention and noting such resources as those available through the state psychological association's colleague assistance program. It should be noted that, in some states, such programs provide specific information and support on how best to approach a colleague as well as where and how to refer them.

If such an informal approach proves inadequate, however, and the psychologist's patient care is being compromised, it may be necessary to consider making a plan (perhaps in consultation with another colleague or colleagues) about how to empathically start a conversation with the individual with clear criteria in mind and with well articulated examples of the perceived problem. A review of the proposals by an APA-affiliated work group on competence problems (Kaslow, et al., 2007) may prove useful in assessing and attempting to intervene.

If the psychologist is resistant to making efforts at remediation, but the evidence of harm to patients seems convincing, a consultation with a state psychological association's ethics committee could provide useful guidance. APA's code of ethics in Standard 2.06 (b) (American Psychological Association, 2002) states that if psychologists suspect that their personal problems or conflicts may interfere with the performance of their professional duties, they should make efforts to determine if the most appropriate course might be to “limit, suspend, or terminate” their work. Smith and Burton Moss (2009) have also noted that the code supports intervention by a colleague as an ethical responsibility when a provider's impairment interferes with patient care. Such intervention should, of course, be done with awareness of how stress inducing such an action may be. If possible, efforts at mobilizing emotional support should be encouraged.

Throughout efforts at intervention such as those noted above, it is important to remain mindful of the fact that depression, the co-occurrence of depression and anxiety, and substance abuse are conditions that have been associated with higher risk for suicidal behavior (Sullivan and Bongar, 2009). Since mental health practitioners may be reluctant to reveal that they are having suicidal thoughts, it is particularly important to be sensitive to subtle expressions of despair or frustration, and to inquire further about them and about possible feelings of hopelessness and wishes to escape or die. The major stigma associated with suicide among health care providers (e.g., Sudak et al., 2009) may also be an important barrier to identifying and intervening with psychologists at risk.

Self-Care as a Preventive Measure

More recently, most State Psychological Associations, and specifically those that have Colleague Assistance Programs, have focused on normalizing challenges that psychologists might encounter ranging from stress reactions to serious mood disorders. The intent is to mitigate the stigma involved in seeking assistance. Thus, even though psychologists are trained to support others in overcoming their challenges, on occasion they may need support themselves. In the 2009 APA Colleague Assistance survey, some of the most frequently reported barriers to the use of colleague assistance were: (1) lack of time (61% of respondents); (2) minimization or denial of issues (43%); (3) privacy or confidentiality concerns (43%); (4) shame, guilt, or embarrassment (40%); (5) lack of knowledge of available resources (31%); (6) fear of loss of professional status (29%); and (7) inadequate social support (27%) (American Psychological Association, 2010).

Perhaps a key to surmounting such barriers may be acknowledging the need for, and the regular use of, self-care strategies not unlike those recommended to clients or patients. The implementation of individualized self-care practices across the developmental spectrum of our professional and personal lives, from graduate school through retirement, needs to be accepted as a wise approach rather than one that springs from weakness or neediness.

Participation in regular self-assessments, some of which are available on colleague assistance program web sites (e.g., on the North Carolina Psychological Association website or the California Psychological Association web site), has been proposed by the APA-affiliated work group on competence problems mentioned earlier (Kaslow, et al., 2007). Such check-ins can be easy and confidential, and, if confidential, they can mitigate the barrier posed by fear of being reported to a licensing board or other governing bodies. Through a check-in process, the psychologist can identify early warning signs and navigate to sites for coping strategies or a phone line for confidential assistance. It should be noted, however, that the effectiveness of this type of self-assessment approach in mitigating stress, shame, embarrassment, or impairment has not yet been tested.

Suggestions for Postvention Following a Colleague Suicide

Edwin Shneidman (1993), the father of the modern study of suicide and suicidal behavior, coined the term postvention to refer to those appropriate and helpful acts that follow a suicide. With a psychologist's suicide, there can be many people who are affected; e.g., his or her family and friends, clients or patients, graduate students or interns, professional colleagues, the professional guild or association, and the local community. Given the recent findings on resilience in the face of loss and traumatic stress (Bonano, 2004), it is probably important to keep in mind that many of the affected individuals may not develop a protracted or pathological response. They may, in fact, utilize the coping resources they have developed and go through a relatively normal recovery. Postvention efforts might best be focused, at least initially, on providing information, support, and reassurance; i.e., on what has come to be called psychological first aid. Of course, family members, students or interns, and colleague survivors can be made aware that more intensive services are available if the need arises, but it is probably best not to presume that they will require them. In this way, those who, in the ensuing months, find that they are having difficulty with the loss will know that options exist for obtaining assistance. Thus, for example, professional colleagues and interns can be informed about whether their state psychological association has a colleague assistance program and whether it provides confidential access to care. In addition, the American Association of Suicidology (AAS) has a Clinician Survivor Task Force that provides consultation, education, support, and resources for mental health clinicians who have experienced a loss by suicide.

When a practicing psychologist commits suicide, his or her patients need to be contacted. It is often not clear whose responsibility it may be to do so and what information should be provided. Moreover, colleagues who are dealing with their own grief may find it emotionally difficult to make these calls. In our interviews of colleague survivors, we encountered several instances in which there was confusion about who could access the deceased psychologist's patient records in order to find contact information. There were also instances in which the deceased psychologist's family objected to having his or her patients informed that the death was a suicide (although patients subsequently learned of that fact through the media or other sources). In one case, difficulties related to patient care and the handling of confidential patient records following a psychologist's suicide were reported as factors in prompting the state licensing board to require that all psychologists prepare a professional will.

A professional will is a document that gives instructions and authority to a designated professional colleague (or executor) to act on the psychologist's behalf in the management of his or her practice in the event of unexpected death or incapacitation (Pope & Vasquez, 2005). Such an advance directive can be a benefit in the aftermath (or postvention phase) of a clinician's death from any cause. In addition to naming an executor, a professional will can, amongst other things, indicate the psychologist's wishes for the transfer and maintenance of confidential patient records, for the storage or destruction of computer files, and for how patients and other relevant parties should be notified.

Conclusion and Future Directions

Data to date present conflicting information in regard to risk of suicide among psychologists. The studies that have reported positive findings (i.e., risk of suicide) have either been flawed or have not been subjected to a peer review process. They raise sufficient concern, however, to lead the authors to conclude that there is an urgent need for further research to evaluate suicide risk for psychologists and determine whether risk varies by gender, ethnicity, or other demographic factors. Should it prove true that suicide is a significant risk for those in the field, it will be important to determine the degree to which stresses in the profession of psychology or obstacles to seeking mental health care might contribute to increasing the likelihood of suicidal behavior. In a similar vein, it would also be important to try to determine if there might be a self-selection process through which vulnerable individuals choose psychology as a profession. The answers to such questions would be invaluable in guiding preventive efforts.

Whether there is current risk or not, the shocked reactions reported by colleague-survivors suggest that psychologist suicides can cause significant distress for fellow professionals and possibly for graduate students or interns who may be led to question the meaningfulness of the career on which they have embarked. Case reports of the impact of a therapist's suicide on patients or clients lead us to question whether many of them may not be burdened with feelings of loss and guilt as well as with questions about the value of therapy and the ability of therapists to engage in a fiduciary relationship in which they put the interests of the patient or client first. These are serious issues and more systematic investigation is needed on each of them.

Suicide is almost always a multi-determined event. The 2009 APA Colleague Assistance Survey has suggested that 18% of survey respondents reported having suicidal ideation while dealing with personal and professional stressors or challenges. As noted above, only further research can inform us about the degree to which some of these stressors might contribute to the development of suicide risk in some psychologists. In the meantime, the preventive emphasis in colleague assistance programs on normalizing the stresses and challenges on the clinician, encouraging self-care, and improving the confidentiality of help-seeking efforts for professionals who worry about damage to their careers seems to make intuitive sense. We must also keep in mind, however, that we do not know how effective such an approach may actually be in reducing suicide risk for psychologists. Again, only future research on the effect of such systemic efforts will answer this question.

Finally, postvention efforts to address the needs of all survivors are needed. The existence of a professional will can aid the efforts of colleagues by clarifying the wishes of the deceased for bringing about closure to his or her practice. The extent to which colleagues may experience a complicated bereavement and need support or therapy is as yet undetermined. State Psychological Association colleague assistance programs may be one resource for services if needed. The AAS Clinician Survivor Task Force may be another (N. Gutin and V. McGann, personal communication, May 17, 2010). Even more concerning, however, is the question of how to repair the apparent rupture in the patient-clinician relationship so that patient or client survivors might wish to re-engage in therapy to deal with the issues for which they originally sought treatment. Offers by concerned colleagues to discuss the loss with the suicide victim's patients would seem to be a step toward assisting patients with bereavement and with re-building trust in the therapeutic process. Providing referrals to new therapists for ongoing treatment when the patient or client expresses a readiness to do so would be another. The information gathered from colleague interviews and case reports in the literature, however, suggests that, although it may be good to make such efforts, there is no guarantee that they will be effective in overcoming such a disruption of the therapeutic process.

Suicide is almost always profoundly disturbing to survivors. Suicide by psychologists, individuals with special expertise in human behavior, seems to be particularly fraught with challenges and raises concerns specific to psychology such as doubt in the value of therapy. Identifying risks, reducing the stigma associated with acknowledging hopelessness or despair, and overcoming other barriers to intervention are critical to reducing the incidence of suicide.

Biographies

Phillip M. Kleespies received his PhD in clinical psychology from Clark University. He is a consultant in mental health in the VA Boston Healthcare System and an Assistant Clinical Professor of Psychiatry at Boston University School of Medicine. He is a Diplomat in Clinical Psychology (ABPP) and a Fellow of the American Psychological Association. His clinical, scholarly, and research interests include behavioral emergencies (i. e., suicide risk, risk of violence, and risk of interpersonal victimization) and the ethics of end-of-life decision-making.

Kimberly A. Van Orden received her PhD from Florida State University. She is instructor and postdoctoral fellow at the Department of Psychiatry, University of Rochester Medical Center. Her research interests are in the etiology and prevention of late-life suicide, particularly the role of social connectedness as a protective factor and mechanism of intervention.

Bruce Bongar received his PhD from the University of Southern California in 1977. He is the Calvin Professor of Psychology at the Pacific Graduate School of Psychology at Palo Alto University, and Consulting Professor in the Department of Psychiatry and the Behavioral Sciences at Stanford University School of Medicine. Dr. Bongar's main research focus for many years has been on suicidal behavior and other clinical emergencies – with a particular interest on standards of care and risk management.

Diane Bridgeman maintains an independent practice in Santa Cruz, CA and previously taught at the University of California, Santa Cruz. She co-chairs the California Psychological Association's Colleague Assistance and Support Program (CLASP), and in 2008-2010 was a member of APA's Advisory Committee for Colleague Assistance (ACCA) chairing the committee the past two years. She gives workshops and presentations, and develops materials and policy on trust/vulnerability/reciprocity, resilience, disaster mental health, colleague assistance/professional development, and self-care/wellness. She chairs committees on disaster response and public education/outreach, Monterey Bay Psychological Association, is a member of APA's disaster response network (DRN), volunteer manager, responder and instructor for the Red Cross, and is chair of the California Disaster Mental Health Coalition.

Lynn F. Bufka received her PhD in psychology from Boston University. She is the Assistant Executive Director of Practice Research and Policy at the American Psychological Association. Her areas of professional interest include health care policy, practice patterns and trends, professional development, adult anxiety disorders and cross cultural issues in psychology.

Daniel I. Galper received his PhD in clinical health psychology from Virginia Tech. He is Director of Research and Special Projects at the American Psychological Association. His areas of research and clinical practice include evidence-based behavioral health, translational research, health care policy, practice patterns and trends, and models and interventions for health behavior change.

Marc Hillbrand received a PhD in clinical psychology from Kent State University in 1986. He is Director of Psychology at Connecticut Valley Hospital in Middletown, CT, and an Assistant Clinical Professor of Psychiatry at Yale University School of Medicine, New Haven, CT. He is a forensic psychologist with specialties in risk assessment and in the treatment of individuals found Not Guilty by Reason of Insanity. His primary research interests include suicidality, interpersonal violence, and innovations in patient-centered care.

Robert I. Yufit received his PhD from the University of Chicago in Psychology & Human Development. He is an Associate Professor at Northwestern University Medical School in Chicago & a Senior Consultant at St Joseph Hospital. He is Board Certified in Clinical Psychology by ABPP. His research and practice is in the area of suicide assessment and suicide prevention. He is a Consulting Editor in the Journal, Suicide & Life Threatening Behavior.