Most anesthesiologists will have at least 1 catastrophic perioperative event, such as patient death or serious injury, during their working lives.1,2 The emotional impact for anesthesiologists involved in a perioperative catastrophe includes guilt, self-doubt, fear of judgment by colleagues, insomnia, anxiety, and depression.1 More than 70% of surveyed anesthesiologists in the United States reported “reliving” a catastrophic perioperative event,1 and 10% of surveyed Canadian anesthesiologists thought about an unanticipated perioperative death every day for more than a year after the event.3 Moreover, 19% of surveyed anesthesiologists in the United States said that they had never recovered fully after a catastrophic perioperative event,1 and 7% of surveyed anesthesiologists in the United Kingdom said that it either took them years to recover, or that they had never fully recovered, from an intraoperative death or serious patient injury.4
When error is associated with an unexpected catastrophic perioperative event, physician distress is likely to be intensified. The psychological effects of medical errors on doctors include shame, guilt, humiliation, self-doubt, anxiety, depression, and detachment.5–11 The potential moderators of emotional responses after errors include the severity of outcome of the error, support from peers and colleagues, workplace culture, and the personality characteristics of the doctor, although research on these factors is limited.8
After a major perioperative event, debriefing may be provided to anesthesiologists and other operating room team members. Debriefing is a generic term covering a variety of approaches. Critical Incident Stress Debriefing (CISD) is a structured, facilitated approach for the discussion of thoughts and feelings related to an incident.12–14 Importantly, CISD should be situated within a broader Critical Incident Stress Management strategy that includes pre-crisis education, posttrauma defusing, CISD, and specialist follow-up for further psychological support if required.15–17
There is very little research examining the attitudes of anesthesiologists to seeking professional help after a catastrophic perioperative event. In general, however, a person’s attitude toward seeking psychological assistance is one of the best predictors of whether they actually do seek psychological treatment.18–22 Doctors are generally poor at seeking help, especially for psychological symptoms and difficulties.23 Doctors do not accept psychological illness or distress in their colleagues or themselves as readily as they do in their patients.24 A pervasive belief held by doctors is that they are, or should be, invulnerable.25 Although 82% of a surveyed mixed specialty group (surgeons, internists, and pediatricians) stated they were somewhat or very interested in counseling after a serious error occurred, they perceived significant barriers to obtaining counseling.9
Stigma is one of the most frequently cited reasons in the general population for not seeking professional counseling.26 “Public stigma” or “social stigma” is associated with the negative, stereotypical labeling of others, and it is influenced by prejudice and discrimination.26 “Self-stigma” can be thought of as internalized social stigma, whereby individuals view themselves as having lower self-worth because of their problems and the need to seek professional help.26–30 Perceived stigma has been shown to reduce help-seeking by depressed doctors in a “dose-response” way, such that the more stigma that is associated with help-seeking, the less likely it is that a doctor will seek help either from their general practitioner or from colleagues.23
In our study, we examined the attitudes of anesthesiologists to supportive strategies and professional counseling in response to an unexpected perioperative catastrophe that they were asked to envisage: the death of one of their patients after anaphylaxis. By using a between-groups study design, we explored whether the presence or absence of an error on the part of the anesthesiologist as the cause of the anaphylaxis influences the anesthesiologist’s attitude to supportive strategies and professional counseling. In particular, we explored whether the presence or absence of the anesthesiologist’s error in the scenario influences the perceived social stigma and self-stigma associated with help-seeking by the anesthesiologist. We hypothesized that the presence of an error as the cause of the patient death would increase the perceived social stigma and self-stigma of help-seeking. Finally, we examined what strategies anesthesiologists believe would assist them in help-seeking.
Ethics approval was obtained from The University of Queensland. A waiver was granted for the requirement for individual signed consent by participants, in accordance with The University of Queensland’s Ethics Guidelines, given that the study involved an anonymous mailed survey with minimal risk to participants. The Participant Information Sheet explained that participation was voluntary. The Australian and New Zealand College of Anaesthetists (ANZCA) Trials Group gave gatekeeper permission to send the survey to a sample of anesthesiologists on its mailing list.
We conducted an anonymous, self-administered survey of 1600 anesthesiologists in Australia randomly selected from the ANZCA mailing list. A power calculation was performed to identify the sample size required to discriminate between a composite score difference for the stigma scales of 1 point, with a SD of 4 points, using 2 separate composite scale measurements (α = 0.025 and β = 0.80). Two groups of 306, or a total of 612 respondents, were required. Assuming an average questionnaire response rate of 40%, questionnaires were sent to 1600 participants. The number of responses was more than the number required for adequate power. The participant sample group is 38.5% of all anesthesiologists on the ANZCA mailing list in Australia.
Participants received a survey package consisting of a cover letter, Participant Information Sheet, questionnaire, and a stamped, addressed reply envelope. The questionnaire had 5 sections: 3 sections had statements to be rated using a 5-point Likert scale ranging from “strongly disagree” to “strongly agree,” a fourth section had questions on demographic characteristics, and a fifth section was for free text comments. The survey packages were identical except for the wording of the case scenario provided.
Section 1 of the Questionnaire
Section 1 contained the case scenario:
“You are anesthetizing a 32-year-old woman for an elective laparoscopic cholecystectomy. You give the patient an IV antibiotic after induction and the patient develops very severe anaphylaxis. You give appropriate treatment, including calling for extra anesthesia personnel to assist you in managing the anaphylaxis. Despite aggressive treatment, the patient cannot be successfully resuscitated and dies in the operating theatre.”
There were 2 different endings to the scenario:
- Error group: “You realize afterward that you should not have given the patient that particular antibiotic, because the patient had told you that she was allergic to it and you had written that on the anesthetic chart preoperatively.”
- No-error group: “The patient had no history of drug allergy.”
There were no features on the surveys to distinguish surveys from the error and no-error groups, apart from the difference in the ending of the case scenario.
There were 5 statements to rate, using a 5-point Likert scale, about generalized supportive strategies for anesthesiologists after this unanticipated perioperative death. The 5 statements have been previously used in a survey3 of Canadian anesthesiologists’ attitudes to unanticipated perioperative death and were used with permission. We made a minor change to 1 statement, replacing “anesthesiologists” with “anaesthetists,” the latter being the more commonly used term in Australia.
Section 2 of the Questionnaire
Section 2 included the following continuation of the case scenario:
“Several months after this patient’s death, you find yourself frequently reliving the event. You are having difficulties sleeping. You feel very upset about what happened. Although talking to family and colleagues has helped somewhat, you wonder whether you should seek some professional counseling assistance (from a psychologist, psychiatrist, or other qualified counselor).”
Section 2 contained 2 survey instruments, used and adapted with permission, to assess stigma related to seeking professional counseling.
- The 10-item Self-Stigma of Seeking Help (SSOSH)31 scale measures, on a 5-point Likert scale, people’s perceptions that seeking help from a counselor, psychologist, or other mental health professional would negatively affect one’s own self-regard and worth, self-confidence, and satisfaction with oneself. The SSOSH31 scale has good construct and criterion validity; for example, it correlates with attitudes toward seeking help (r = −0.53 to −0.63), intention to seek counseling (r = −0.32 to −0.38), and measures of public stigma (r = 0.46 to 0.48).31 Further, the SSOSH31 scale has strong internal consistency reliability (from 0.88 to 0.91) and good 2-month test–retest correlation of 0.72.31 The SSOSH31 scale has also been shown to differentiate people who sought psychological services from those who did not.31
- The 5-item Social Stigma for Receiving Psychological Help (SSRPH)32 scale uses a 4-point Likert scale to measure people’s perceptions of how socially stigmatizing it is to receive professional psychological help. The SSRPH32 scale correlates (r = −0.40)32 with the Attitudes Toward Seeking Professional Psychological Help Scale: Short Form33 and with a measure of public stigma (r =0.41),34 the Devaluation-Discrimination Scale.35 The internal consistency was reported at 0.73.32
For consistency within our survey, the SSRPH32 scale is adapted, with permission, to a 5-point Likert scale. Further, the SSOSH31 scale and the SSRPH32 scale are adapted, with permission, in our study to use the word “counselor,” instead of “therapist” or “psychologist,” as in the SSOSH31 and SSRPH32 scales, respectively, to keep terminology consistent. The SSOSH31 scale has been previously used with substitutions of the word “therapist,” including use of “group therapy” and “individual therapy” (internal consistency, 0.78 and 0.80),36 “group counseling” (internal consistency, 0.85),37 and “career counseling” (internal consistency, 0.89).34 The SSRPH32 scale has also had word substitutions for “psychologist,” including “career counseling” (internal consistency, 0.80)34 and “executive coaching” (internal consistency, 0.8).38 The SSRPH32 scale has been previously adapted to a 5-point Likert scale, with internal consistency of 0.75 and 0.72.39,40 In our study, the internal consistency of the modified SSOSH scale was 0.82 and for the modified SSRPH scale was 0.80.
Section 3 of the Questionnaire
Section 3 included 5 items covering anesthesiologists’ views of factors or resources that may assist them in seeking professional counseling after a major adverse event.
Section 4 of the Questionnaire
Section 4 gathered demographic characteristics, with 4 questions on type of practice (mostly private, mostly public, equal public/private, and other), age group (<30, 30–45, 46–60, 61–75, and >75 years), gender, and location of work (capital city or regional/rural).
Section 5 of the Questionnaire
This section was for participants’ free text comments about the survey or about seeking professional counseling.
The survey was piloted on 15 anesthesiologists to examine the usability of the survey, using contextual inquiry and modified cognitive walkthrough techniques. Minor changes were subsequently made to the wording of the survey instructions. The survey packages were prepared by the researchers in 2 separate groups, error group and no-error group, and delivered to ANZCA. Participants were randomly assigned by ANZCA staff using a randomization table41 to receive either the error or no-error version of the survey. The survey packages were addressed and mailed by ANZCA. The researchers did not have access to participants’ names and addresses. There were 3 follow-up reminders emailed at 2-week intervals to participants by ANZCA.
Each returned survey was given a unique identification number and an “E” or “NE” depending on whether it was in the error group or the no-error group. Missing data were treated as null. Data were entered into an Excel spreadsheet (Microsoft, Redmond, WA).
The data obtained from the Likert scales in sections 1 and 3 were treated as ordinal data, and nonparametric inferential tests were used. A 2 (groups) × c (categories) X2 test was used to compare the demographic characteristics of participants in the error and no-error groups. A series of 2 (groups) × 5 (levels of agreement) Wilcoxon-Mann-Whitney (WMW) tests was used to compare the responses of participants in the error and no-error groups for sections 1 and 3. For section 2, the scores on the SSOSH31 and SSRPH32 scales were compared between the error and no-error groups using Student t tests.
For sections 1 and 3, a series of r (demographic categories) × c (levels of agreement) Kruskal-Wallis tests was used to examine the effects of demographic characteristics on participants’ responses. For sections 1 and 3, post hoc analyses of statistically significant demographic influences were performed using 2 × c WMW tests. For section 2, a Student t test was used to compare the effect of gender on responses, and multivariate regression analysis was performed to examine the effects of demographic characteristics on participants’ responses.
To maintain the family-wise type I error rate at 0.05, all raw P values from the WMW tests and Kruskal-Wallis tests for sections 1 and 3 were adjusted for the effects of multiple comparisons using the Ryan-Holm step-down procedure,42–45 with raw P values arranged in an ascending order and sequentially adjusted until all hypotheses were tested or until the adjusted P′ value was >0.05. The corrected P values are designated by the use of a prime symbol adjacent to the letter P, P′, in distinction from the raw P values. For the Student t test comparison of the SSOSH31 and SSRPH32 scales for section 2, and the multivariate regression analyses, the P values were not adjusted. Data were analyzed using Stata version 12.1 (StataCorp, College Station, TX). An adjusted P′ value of <0.05 for sections 1 and 3, and an unadjusted p value of <0.05 for section 2, was considered statistically significant.
Thematic analysis of participants’ comments was performed by 2 of the researchers identifying the themes and subthemes that were present within the entire comments data set, using inductive analysis with a semantic approach in a phased iterative process.46 The entire comments data set was entered into an Access (Microsoft) database. The 2 researchers independently read and reread the data set, manually identifying the response patterns and meanings and generating the initial codes, with weighting given to prevalence of data items as well as relationship of data items to the overall research topics. The development of the themes was data driven, and using reflexive dialogue and visual representation of the codes, the initial codes were manually sorted into themes. A visual thematic map was created to identify the overarching themes and the subthemes. The data set was then manually recoded using the identified themes and subthemes into an Excel spreadsheet (Microsoft). The number of comments for each theme and subtheme was counted.
Eight hundred one surveys were returned. Of these, 16 surveys were returned to sender unopened. A further 19 surveys had been returned blank: 3 surveys were returned blank without explanation, 7 people explained that they no longer practiced clinical anesthesia due to retirement or other factors, 5 explained they were practicing in pain medicine only, 2 explained they were practicing exclusively in an intensive care unit, and 2 were nonanesthesiologists who were honorary fellows of ANZCA.
There were, thus, 766 usable returned surveys, with a usable response rate of 48.9%, calculated by dividing 766 by 1565, having subtracted the 16 returned to sender and 19 blank surveys from the 1600 that were mailed. Three hundred seventy-six surveys from the error group and 390 surveys from the no-error group were returned, with no statistically significant difference in the response rates. Seven hundred sixty-six participants had completed section 1, 756 completed section 2, 745 completed section 3, and 742 completed section 4. The percentage of missing data was 0% for section 1, 1.3% for section 2, 2.7% for section 3, and 3.1% for section 4. Participants worked mainly in public hospital practice in a capital city (Table 1). As would be expected from the randomization process,47 there were no statistically significant demographic differences between participants in the error group versus the no-error group with respect to whether they were in private, public, or mixed practice; capital city or regional/rural practice; age group; or gender.
Section 1: Generalized Supportive Strategies
For 4 of the 5 statements, there was a statistically significant difference between the error group and the no-error group (Tables 2 and 3), with the error group being more likely to agree/strongly agree that:
- Time off after the event is advisable.
- Counseling after the event is advisable.
- A formal strategy to help anesthesiologists deal with the aftermath of perioperative deaths is advisable.
- The anesthesiologist should not perform any more cases on the day of a perioperative death.
Eighty-three percent (confidence interval [CI], 80%–86%) of participants agreed/strongly agreed that “Debriefing the operating room team immediately after a perioperative death is advisable.”
Female anesthesiologists were more likely than males to agree/strongly agree with 3 of the statements in section 1 (Table 4):
- Counseling after the event is advisable.
- A formal strategy to help anesthesiologists deal with the aftermath of perioperative deaths is advisable.
- The anesthesiologist should not perform any more cases on the day of a perioperative death.
There were no other significant effects of demographic variables.
Section 2: Attitudes to Perceived Stigma of Counseling
There were no statistically significant differences between error group and no-error group scores for the SSOSH31 scale or for the SSRPH32 scale (Table 5). A t test showed no difference between genders for the SSOSH31 scale (P = 0.47) or SSRPH32 scale (P = 0.112). A multivariate regression analysis showed no correlations of interest for the SSOSH31 scale scores (Table 6). There was a correlation with the SSRPH32 scale scores for age (P = 0.003) and gender (P = 0.019; Table 7), with younger anesthesiologists and male anesthesiologists more likely to score higher, equating to more social stigma, than older anesthesiologists and female anesthesiologists.
Section 3: Strategies to Assist Help-Seeking
For 2 of the 5 statements, there was a statistically significant difference between the error group and the no-error group (Tables 8 and 9), with the error group being more likely to agree/strongly agree that the following would assist them in finding assistance:
- A formal hospital-based process that provides information to anesthesiologists on where to obtain professional counseling after a major adverse event.
- Availability of after-hours counseling services.
For the statements for which there were no differences between the error and no-error groups, the majority of participants agreed/strongly agreed that the following strategies would assist them in seeking counseling after a major adverse event:
- Role models, such as anesthesiologists or other medical practitioners, who talk openly about their own experiences in getting professional counseling to deal with major adverse events (90% [CI, 88%–92%] agree/strongly agree).
- Education about how professional counseling may assist doctors after a major adverse event (82% [CI, 80%–85%] agree/strongly agree).
Only 36% (CI, 32%–39%) of participants agreed/strongly agreed that “Availability of on-line or telephone, rather than face-to-face, counseling” would be an assistive strategy.
Female anesthesiologists were more likely to agree/strongly agree that the following would be assistive strategies (Table 10):
- Availability of after-hours counseling services.
- Availability of online or telephone, rather than face-to-face, counseling.
There were no other significant effects of demographic variables.
Two hundred eighty-two participants (37% of the 766 usable returned surveys) provided comments. Of these 282 participants, 92 (33%) provided comments that were supportive of counseling. Five (2%) participants were anticounseling. Only 4 (1%) participants wrote comments supporting mandatory counseling, whereas 30 (11%) participants emphasized that counseling should be an individual choice. Eighteen (6%) participants recounted good experiences with counseling, whereas 10 (4%) participants described unhelpful counseling experiences. There was concern over the qualifications of the counselor (20 comments, 7%), and that the counselor may not understand the unique circumstances of being an anesthesiologist and having a patient death or other adverse event (9 comments, 3%). “It is not good to engage junior, nonmedical counselors. They do not get it and can do great harm.” “I don’t think I would be much helped by a junior social worker…with little understanding of the work environment of anesthesiologists.”
A prominent theme was the importance of peer support by colleagues, with 56 (20%) participants commenting on the need for peer support, and 20 (7%) detailing their experience of good peer support. Some comments on peer support included “the next day (after the unexpected death) several colleagues phoned me to offer their support and listen to what occurred in the OT (Operating Theater) that night, this helped a great deal for me to handle a very traumatic situation,” and “the best ‘counseling’ is by talking to a skilled trusted senior anesthesiologist to put the case into perspective as they can much more understand the context and situation than a counselor.” The need for preemptive strategies was a further theme, with 22 (8%) participants emphasizing the need for education about the psychosocial effects of adverse events and 20 (7%) participants saying that greater awareness of supportive resources was required. “Since we are good risk managers it follows that early on (during the training period) is preferable to provide education than after an adverse event.” “We are doing far too little to prepare ourselves for these scenarios.”
The need for cultural change in medicine was mentioned by 17 participants (6%). “We had to ‘suck it up’ and carry on with the emergency list (after a patient death on the table).” “There is still a ‘macho’ ‘get over it’ mentality…increased visibility of support services may help.” Accounts were given of past, and ongoing, personal distress after a major adverse event, as well as distress and suicides of colleagues after major adverse events. Twenty-nine (10%) participants commented that they had experienced a patient death, and 15 (5%) wrote of other adverse events. Twenty-six (9%) recounted experiencing trauma after the event. “Suffered PTSD (posttraumatic stress disorder) for 18 months until I realized what was going on.” “Involved in unexpected death ... Very traumatized. Asked ‘Are you ok?’ by senior staff members. What am I to say?? No?? So you say I’m okay. Having nightmares etc.” “I felt terrible for many months after but eventually reconciled the fact that (the patient) was dead.” “NO ONE…sought to ask how I felt about it (patient death) and how it was affecting me.” “It took me 2 years to get over it (the patient death), and I did seek counseling, which helped.” Some participants provided detailed accounts of their adverse events or distress that they asked not to be published.
A further theme was time off from work, with 20 (7%) comments supportive of time off and 22 (8%) comments about other aspects of time off after the patient death. “I think a formal need to take time off…takes away any pressure people may feel to keep working.” Others recognized practical difficulties of arranging time off and also that individuals vary. “Personally, I found working on the same day after an unexpected death helped the healing process.” Some participants were concerned about confidentiality of counseling and also the consequences of seeking counseling on their job prospects and medical licensing, as well as medicolegal and personal insurance status.
Our participants appeared to recognize the need for supportive strategies after the hypothetical unexpected patient death, together with a greater need for support when an error had caused the patient’s death. Interestingly, the presence of an error did not influence the perceived self-stigma or social stigma of seeking counseling help, as measured by the SSOSH31 and SSRPH32 scales. As discussed below, however, there remain many questions for the anesthesiology profession in how to best provide the required support for anesthesiologists after a patient death.
Generalized Support and Coping Strategies
The 83% (CI, 80%–86%) of our participants who agreed/strongly agreed that “Debriefing the operating room team immediately after a perioperative death is advisable,” is a similar result to the 79% of anesthesiologists who agreed/strongly agreed with this statement in a 2010 Canadian study that examined attitudes to participants’ recalled significant unexpected perioperative patient death.3 Further, in a 2012 US survey, 68% of anesthesiologists believed that mandatory debriefing sessions would be a good strategy after an unanticipated death or serious injury of a patient.1 However, it is generally recommended that debriefing should not be forced or compulsory for any staff member.48,49 Mandatory debriefing may be traumatizing to those who recover best with solitude and isolation after an incident.50 More research is required to determine what types, and timing, of debriefing that anesthesiologists would find most helpful.
More of the participants in the error group than the no-error group agreed/strongly agreed that time off and counseling after the event were advisable, that the anesthesiologist should not perform any more cases on the day of a perioperative death, and that a formal strategy to help the anesthesiologist deal with the aftermath of the event was advisable. This may reflect participants’ awareness that a perioperative death associated with an error is likely to have a greater psychological effect on the anesthesiologist, in keeping with the known effects of medical errors on doctors including considerable guilt, shame, self-doubt, anxiety, and depression.5–10
Of note, 21% (CI, 16%–25%) of the error group and 28% (CI, 24%–33%) of the no-error group disagreed/strongly disagreed, or were neutral, that the anesthesiologist should not perform further cases on the day of a perioperative death. It is unclear whether these participants’ responses were motivated by probable production pressure to continue with the day’s work or by a view that subsequent work performance would be unaffected by the event. In the US survey of anesthesiologists,1 67% of respondents reported that in the first 4 hours after a perioperative catastrophe (unanticipated patient death or serious injury), their ability to provide anesthesia was compromised, with 32% reporting that they were greatly compromised during this period. Moreover, 51% of respondents stated that their ability to provide care was compromised for 24 hours after a catastrophic patient event.1 However, 46% of anesthesiologists in the Canadian survey3 performed further elective cases after an unanticipated perioperative death, with 32% doing so because there was no colleague to relieve them and 6% doing so because of production pressure from surgeons. Although 74% of these anesthesiologists thought time off should be offered after a perioperative catastrophe of patient death or serious injury, only 7% were given time off after a major event.3 In many other safety-critical industries, such as aviation, rail transport, and chemical and nuclear plants, it would be unacceptable for staff to continue with “business as usual” after involvement in a major adverse event; yet, this remains an unresolved question in our profession with respect to when, and for how long, time off should occur.
Our female participants were more likely to agree/strongly agree that counseling was advisable after the unexpected patient death. In general, women have more positive attitudes toward seeking professional psychological help than men.51 Traditional masculine role norms such as self-reliance, toughness, and restricted emotionality act as barriers to psychological help-seeking.52
For our participants, there was no significant difference between the error group and the no-error group for the scores on the SSOSH31 or the SSRPH32 scales, suggesting that for the given scenario, the presence of an error was not associated with increased perceived self-stigma or social stigma in help-seeking. These results did not support our hypothesis that the presence of an error causing the patient death would increase the perceived social stigma and self-stigma of help-seeking. One possible explanation is that the presence of an error contributing to the patient death may be viewed as a legitimate reason for psychological distress and the need for help-seeking. In contrast, if an anesthetist feels psychological distress after a death where they are not to blame, perhaps they may think they should be able to get over it and not need professional help. Other possible explanations for the lack of difference between the groups is that the stigma perceived around seeking counseling assistance may be more strongly mediated by other factors than by the cause of a patient adverse event. It may be that issues such as previous exposure to counseling in various settings, different organizational cultures toward help-seeking behaviors, perceived quality of supportive relationships with peers, and previous education about effects of adverse events on doctors may have a greater effect than committing an error on perceived social stigma and self-stigma for help-seeking after an adverse event.
There is little published research on stigma and help-seeking by doctors. Higher levels of perceived stigma are correlated with reduced help-seeking in depressed doctors.23 Further, doctors with moderate to severe depression were more likely to avoid seeking treatment because of stigma than doctors with minimal to mild depression scores,53 suggesting a relationship between stigma and severity of symptoms for which one is help-seeking. Whether this same effect would hold for doctors seeking help for symptoms related to a discrete event, such as the patient death in the scenario used in our research, needs further clarification. Conversely, even exceptional events such as a major adverse patient outcome will always be contextualized in terms of other present and past stresses experienced by an anesthesiologist, in both their professional and personal life. Exposure to previous traumatic life events affects responses to current traumatic events.54,55
Forms of Assistance
The error group was more likely than the no-error group to agree/strongly agree that it would assist them to have a formal hospital-based process to provide information about where to obtain counseling and for there to be after-hours counseling services. This result is in keeping with our findings in section 1, where significantly more of our error group agreed/strongly agreed with a formal strategy to assist with the aftermath of perioperative deaths, and that counseling after the event was advisable. These results suggest that our participants are recognizing that the distress experienced when a patient death is caused by an error may be more severe and more likely to require professional assistance.
A hospital-based formal process has the dual function of providing information and giving organizational recognition of, and support for, employee distress. The Scott et al.’s Three-Tiered Interventional Model of Support56 escalates support as required from Tier 1, which is departmental support, to Tier 2, which involves trained peer supporters, patient safety officers, and risk managers, to Tier 3, which involves expedited referral to professional counseling support. Further research is required to determine what aspects of a formal hospital-based process are the most valuable to anesthesiologists after a major adverse event.
Although the majority of both the error group (69%; CI, 65%–74%) and the no-error group (59%; CI, 54%–64%) agreed/strongly agreed with the assistive strategy of after-hours counseling services, only 38% (CI, 34%–44%) of the error group and 33% (CI, 28%–38%) of the no-error group agreed/strongly agreed with the strategy of online or telephone counseling rather than face-to-face counseling. Our participants appear to place a value on face-to-face interaction with a counselor, although such services may be more difficult to access after hours. Female participants were significantly more likely than males to agree with both the need for after-hours counseling and the availability of online or telephone counseling. It may be that family care responsibilities make it harder for women to attend face-to-face counseling after hours, or perhaps women are more comfortable with telephone and online counseling.
There has been little research comparing counseling modalities for doctors. A randomized trial of face-to-face counseling versus counseling by telephone versus the use of bibliotherapy (use of structured written materials) for occupational stress in health workers (nurses and managers) found all 3 interventions were equally effective,57 although participants with posttraumatic stress disorder were excluded in this trial. For stress-related symptoms, fewer doctors (23%) reported that they would use a telephone helpline than face-to-face counseling (35%).23 Our results suggest that the resources available for anesthesiologists after a perioperative death should include face-to-face, after-hours counseling.
Our participants were highly supportive (90% [CI, 88%–92%] agreed/strongly agreed) of the need for peers who act as role models to talk openly about their own experiences in getting professional counseling after adverse events. However, doctors are often reluctant to disclose to colleagues that they are suffering from psychological symptoms.58 On one hand, our results suggest that anesthesiologists want others to talk about their experiences, but on the other hand, the prevailing medical culture may preclude this. How to address this gap is a challenge that needs further research and support.
Reflections on Our Participants’ Comments
The accounts from our participants of their own or colleagues’ suffering after involvement in an adverse event highlight the need for improved support structures. One way to address the themes raised by our participants is to preemptively provide information and education to anesthesiologists about a comprehensive suite of resources, including peer support and how to access appropriate professional counseling if required. Readily available trusted peer supporters are required, together with simple referral pathways to experienced counselors who understand the work context in operating theaters and the inherent high-performance demands. Anesthesiologists pride themselves on being proactive risk managers of their patients’ safety, yet we may not be adequately equipped to protect our own psychological safety, should we be involved in a major adverse event. This issue deserves our profession’s attention and action.
LIMITATIONS OF THE STUDY
Our participants were anesthesiologists in Australia, and our results may not necessarily translate to other countries. Responder bias may have affected our results, in that those who responded may have been more interested in the topic than nonresponders, or more likely to have previously experienced a major perioperative adverse event. Previous experience with counseling may have biased participants’ responses. Similarly, previous exposure to patient death or major adverse events may have influenced responses. Social desirability bias59 may have been a confounding factor. Furthermore, a different case scenario and different patient outcome may elicit different results. Moreover, we examined attitudes rather than actual behaviors after a patient death. Although both the SSOSH31 and SSRPH32 scales have been used in adapted forms, we did not perform psychometric testing on our adapted version of the scales, other than a measure of internal consistency. It is possible that the word change from “therapist” and “psychologist” to “counselor,” and the change from a 4-point to a 5-point Likert scale for the SSRPH32 scale, may have affected our results.
Our participants were largely in agreement with approaches such as debriefing, time off, and formal hospital strategies to assist anesthesiologists after an unexpected intraoperative death. The presence of an error as the cause of the given adverse event did not appear to increase social stigma or self-stigma in relation to help-seeking, disproving our hypothesis. However, our participants appeared more likely to recognize the potential value of formal supportive processes, time off, and counseling when an error caused the patient death in our scenario, suggesting that they recognize the potential greater adverse effects of an error on the anesthesiologist. Our participants were highly supportive of educative, hospital-based, and role modeling strategies to assist help-seeking, together with the availability of after-hours, face-to-face counseling services. Given the potential for physician distress after an unanticipated patient death, this is an area deserving further research and discussion within our profession. E
Name: Gaylene C. Heard, MBBS, MHumanFact, FANZCA.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Attestation: Gaylene C. Heard has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
Name: Rowan D. Thomas, MBBS, MPH, FANZCA.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Attestation: Rowan D. Thomas has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Penelope M. Sanderson, BA(Hons), MA, PhD, FASSA, FHFES, FIEA.
Contribution: This author helped design the study, analyze the data, and write the manuscript.
Attestation: Penelope M. Sanderson reviewed the analysis of the data and approved the final manuscript.
This manuscript was handled by: Franklin Dexter, MD, PhD.
The authors thank Professor David Story, Chair of Anaesthesia, University of Melbourne, and Professor John Ludbrook, University of Melbourne, for advice. The authors also thank the ANZCA Clinical Trials Group for permission to mail the survey to ANZCA fellows. In addition, Ms. Stephanie Poustie, Ms. Anna Parker, and Ms. Karen Goulding of ANZCA, and Mr. Daniel Myles, provided valued assistance.
1. Gazoni FM, Amato PE, Malik ZM, Durieux ME. The impact of perioperative catastrophes on anesthesiologists: results of a national survey. Anesth Analg. 2012;114:596–603.
2. White SM, Akerele O. Anaesthetists’ attitudes to intraoperative death. Eur J Anaesthesiol. 2005;22:938–41.
3. Todesco J, Rasic NF, Capstick J. The effect of unanticipated perioperative death on anesthesiologists. Can J Anaesth. 2010;57:361–7.
4. Wee M. Catastrophes in anaesthetic practice: survey of AAGBI members in 2005. Anaesthesia News. 2005;224:2–3.
5. Fisseni G, Pentzek M, Abholz HH. Responding to serious medical error in general practice—consequences for the GPs involved: analysis of 75 cases from Germany. Fam Pract. 2008;25:9–13.
6. Hobgood C, Hevia A, Tamayo-Sarver JH, Weiner B, Riviello R. The influence of the causes and contexts of medical errors on emergency medicine residents’ responses to their errors: an exploration. Acad Med. 2005;80:758–64.
7. Schwappach DL, Boluarte TA. The emotional impact of medical error involvement on physicians: a call for leadership and organisational accountability. Swiss Med Wkly. 2009;139:9–15.
8. Sirriyeh R, Lawton R, Gardner P, Armitage G. Coping with medical error: a systematic review of papers to assess the effects of involvement in medical errors on healthcare professionals’ psychological well-being. Qual Saf Health Care. 2010;19:e43.
9. Waterman AD, Garbutt J, Hazel E, Dunagan WC, Levinson W, Fraser VJ, Gallagher TH. The emotional impact of medical errors on practicing physicians in the United States and Canada. Jt Comm J Qual Patient Saf. 2007;33:467–76.
10. West CP, Huschka MM, Novotny PJ, Sloan JA, Kolars JC, Habermann TM, Shanafelt TD. Association of perceived medical errors with resident distress and empathy: a prospective longitudinal study. JAMA. 2006;296:1071–8.
11. White AA, Gallagher TH, Krauss MJ, Garbutt J, Waterman AD, Dunagan WC, Fraser VJ, Levinson W, Larson EB. The attitudes and experiences of trainees regarding disclosing medical errors to patients. Acad Med. 2008;83:250–6.
12. Dyregrov A. Caring for helpers in disaster situations: psychological debriefing. Disast Manag. 1989;2:25–30.
13. Mitchell JT. When disaster strikes the critical incident stress debriefing process. JEMS. 1983;8:36–9.
14. Mitchell JT, Everly GS Jr. The scientific evidence for critical incident stress management. JEMS. 1997;22:86–93.
15. Mitchell JT. Stress. Development and functions of a critical incident stress debriefing team. JEMS. 1988;13:42–6.
16. Mitchell JT. The history, status and future of critical incident stress debriefing. JEMS. 1988;13:47–51.
17. Everly GS Jr, Mitchell JT. The debriefing “controversy” and crisis intervention: a review of lexical and substantive issues. Int J Emerg Ment Health. 2000;2:211–25.
18. Carlton PA, Deane FP. Impact of attitudes and suicidal ideation on adolescents’ intentions to seek professional psychological help. J Adolesc. 2000;23:35–45.
19. Cepeda-Benito A, Short P. Self-concealment, avoidance of psychological services, and perceived likelihood of seeking professional help. J Couns Psychol. 1998;45:58–64.
20. Codd RT, Cohen BN. Predicting college student intention to seek help for alcohol abuse. J Soc Clin Psychol. 2003;22:168–91.
21. Deane FP, Todd DM. Attitudes and intentions to seek professional psychological help for personal problems or suicidal thinking. J Coll Stud Psychother. 1996;10:45–59.
22. Vogel DL, Wester SR. To seek help or not to seek help: the risks of self-disclosure. J Couns Psychol. 2003;50:351–61.
23. Adams EF, Lee AJ, Pritchard CW, White RJ. What stops us from healing the healers: a survey of help-seeking behaviour, stigmatisation and depression within the medical profession. Int J Soc Psychiatry. 2010;56:359–70.
24. Armstrong E. Rehabilitating troubled doctors. BMJ. 1997;7095:314.
25. Adshead G. Healing ourselves: ethical issues in the care of sick doctors. Adv Psychiatric Treat. 2005;11:330–7.
26. Corrigan P. How stigma interferes with mental health care. Am Psychol. 2004;59:614–25.
27. Link B, Phelan J. Conceptualizing stigma. Annu Rev Sociol. 2001;27:363–85.
28. Link BG, Struening EL, Neese-Todd S, Asmussen S, Phelan JC. Stigma as a barrier to recovery: the consequences of stigma for the self-esteem of people with mental illnesses. Psychiatr Serv. 2001;52:1621–6.
29. Nadler A, Fisher JD. Berkowitz L. The role of threat to perceived control in recipient reaction to help: theory development and empirical validation. In: Advances in Experimental Psychology. 1986. New York: Academic Press. 81–121.
30. Vogel DL, Bitman RL, Hammer JH, Wade NG. Is stigma internalized? The longitudinal impact of public stigma on self-stigma. J Couns Psychol. 2013;60:311–6.
31. Vogel DL, Wade NG, Haake S. Measuring the self-stigma associated with seeking psychological help. J Couns Psychol. 2006;53:325–37.
32. Komiya N, Good GE, Sherrod NB. Emotional openness as a predictor of college students’ attitudes towards seeking psychological help. J Couns Psychol. 2000;47:138–43.
33. Fischer EH, Farina A. Attitudes toward seeking professional psychological help: a shortened form and considerations for research. J Coll Stud Develop. 1995;36:368–73.
34. Ludwikowski WMA, Vogel D, Armstrong PI. Attitudes toward career counseling: the role of public and self-stigma. J Couns Psychol. 2009;56:408–16.
35. Link B, Cullen F, Frank J, Wozniak J. The social rejection of former mental patients: understanding why labels matter. Am J Sociol. 1987;92:1461–500.
36. Shechtman Z, Vogel D, Maman N. Seeking psychological help: a comparison of individual and group treatment. Psychother Res. 2010;20:30–6.
37. Vogel DL, Shechtman Z, Wade NG. The role of public and self-stigma in predicting attitudes toward group counseling. The Couns Psychol. 2010;38:904–22.
38. McKelley RA. The effect of practitioner title and gender socialization on men’s attitudes, stigma and preferences for seeking help. PhD Thesis, University of Texas. 2008. Available at: http://repositories.lib.utexas.edu/handle/2152/17880
. Accessed August 4, 2014.
39. Lannin DG, Vogel DL, Brenner RE, Tucker JR. Predicting self-esteem and intentions to seek counseling: the internalized stigma model. Couns Psychol. 2015;43:64–93.
40. Vogel DL, Bitman RL, Hammer JH, Wade NG. Is stigma internalized? The longitudinal impact of public stigma on self-stigma. J Couns Psychol. 2013;60:311–6.
42. Brown BW, Russell K. Methods of correcting for multiple testing: operating characteristics. Stat Med. 1997;16:2511–28.
43. Holm SA. A simple sequentially rejective multiple test procedure. Scand J Stat. 1979;6:65–70.
44. Ludbrook J. Multiple comparison procedures updated. Clin Exp Pharmacol Physiol. 1998;25:1032–7.
45. Morikawa T, Terao A, Iwasaki M. Power evaluation of various modified Bonferroni procedures by a Monte Carlo study. J Biopharm Stat. 1996;6:343–59.
46. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006;3:77–101.
47. Altman DG. Practical Statistics for Medical Research. 1991. London: Chapman & Hall.
48. Hawker DM, Durkin J, Hawker DS. To debrief or not to debrief our heroes: that is the question. Clin Psychol Psychother. 2011;18:453–63.
49. Rose S, Bisson J, Churchill R, Wessely S. Psychological debriefing for preventing post-traumatic stress disorder (PTSD). Cochrane Database Syst Rev. 2002;CD000560.
50. Mitchell JT, Everly GS. Critical Incident Stress Debriefing. 19962nd ed. revised. Ellicott City: Chevron.
51. Nam SK, Chu HJ, Lee MK, Lee JH, Kim N, Lee SM. A meta-analysis of gender differences in attitudes toward seeking professional psychological help. J Am Coll Health. 2010;59:110–6.
52. Levant RF, Stefanov DG, Rankin TJ, Halter MJ, Mellinger C, Williams CM. Moderated path analysis of the relationships between masculinity and men’s attitudes toward seeking psychological help. J Couns Psychol. 2013;60:392–406.
53. Schwenk TL, Gorenflo DW, Leja LM. A survey on the impact of being depressed on the professional status and mental health care of physicians. J Clin Psychiatry. 2008;69:617–20.
54. Martin CG, Cromer LD, Deprince AP, Freyd JJ. The role of cumulative trauma, betrayal, and appraisals in understanding trauma symptomatology. Psychol Trauma. 2013;52:110–8.
55. Ogle CM, Rubin DC, Siegler IC. The impact of the developmental timing of trauma exposure on PTSD symptoms and psychosocial functioning among older adults. Dev Psychol. 2013;49:2191–200.
56. Scott SD, Hirschinger LE, Cox KR, McCoig M, Hahn-Cover K, Epperly KM, Phillips EC, Hall LW. Caring for our own: deploying a systemwide second victim rapid response team. Jt Comm J Qual Patient Saf. 2010;36:233–40.
57. Kilfedder C, Power K, Karatzias T, McCafferty A, Niven K, Chouliara Z, Galloway L, Sharp S. A randomized trial of face-to-face counselling versus telephone counselling versus bibliotherapy for occupational stress. Psychol Psychother. 2010;83:223–42.
58. White A, Shiralkar P, Hassan T, Galbraith N, Callaghan R. Barriers to mental healthcare for psychiatrists. Psychiatr Bull. 2006;30:382–4.
Copyright © 2016 International Anesthesia Research Society
59. Vogt WP. Dictionary of Statistics and Methodology. 2005. Newbury Park, CA: Sage.