The death of patients by suicide has profound effects on the personal and professional life of many psychiatrists.Reference Séguin, Bordeleau, Drouin, Castelli-Dransart and Giasson1 Patient suicide can have an influence on staff recruitment and retention, quality of professional life, patient care and professional well-being. Currently there is limited and idiosyncratic support for clinicians who experience patient suicide. This important issue has not been well examined.
The majority of psychiatrists experience the death of a patient by suicide such at least once during their career, with many experiencing this more often.Reference Alexander, Klein, Gray N, Dewar and Eagles2,Reference Courtenay and Stephens3 In one study, 53% of psychiatrists reported stress levels in the weeks following a suicide comparable to those reported in studies of people seeking treatment after the death of a parent.Reference Chemtob, Hamada, Bauer, Kinney and Torigoe4 The intensity of this experience can result in a combination of symptoms of post-traumatic stress disorder, shame, guilt, anger, isolation, and fear of litigation and of retribution from the psychiatric community.Reference Gitlin5,Reference Ruskin, Sakinofsky, Bagby, Dickens and Sousa6 The stress following a patient suicide can be exacerbated by organisational responses, including serious incident enquiries, and the pressure of attending the coroner's court.Reference Campbell and Hale7 The result can be increased anxiety in clinical situations and changes in practice.Reference Alexander, Klein, Gray N, Dewar and Eagles2 In those suffering a severe stress reaction, retirementReference Alexander, Klein, Gray N, Dewar and Eagles2 and change in careerReference Dewar, Eagles, Klein, Gray and Alexander8 are often considered.
Aims
The aims of the present study were to investigate: (a) how patient suicide affects the emotional well-being and clinical practice of psychiatrists; and (b) the resources that psychiatrists would find helpful before and after the suicide of a patient. In addition, we wanted to find out about psychiatrists' attitudes to suicide and suicide prevention, and their perceptions of institutional and societal pressures.
Method
Design
A web-based survey was designed by a multidisciplinary team of clinicians experienced in this area and informed by past research.Reference Wolfart9 The survey consisted of 42 questions on SurveyMonkey and was sent out via medical directors to psychiatrists in two National Health Service (NHS) trusts where the researchers were based and where there were particular concerns about supporting clinicians affected by patient suicide, as well as two neighbouring trusts that had shown an interest in participation. Two trusts were based in London and two in south-east England. All had a similar local suicide rate, corresponding to the average rate in England of 9–10 per 100 000. Respondents were requested to focus on the patient suicide that had the most significant emotional effect on them (assuming they had experienced one). They were asked about the effects on them personally and on their clinical practice, what they had found helpful and unhelpful following the incident, and suggestions for possible helpful resources and actions. Respondents were also asked their opinion on: (a) the predictability of suicides in secondary care; (b) their experience of feeling under external pressure to prevent suicides; and (c) their views of their role in relation to suicide prevention. The layout of the questions varied (see Appendix 1). The survey was open for 1 month, and there was a follow-up email prompt after 2 weeks.
Participants
Psychiatrists at all levels of training and experience in four trusts were emailed about the survey.
Analysis
Data were examined numerically and via thematic analysis. Responses of men and women were compared with parametric t-tests.
Results
The survey was circulated to 520 psychiatrists, of whom 174 responded (34%). The characteristics of the respondents are shown in Table 1. Of the 140 respondents who had experienced death of a patient by suicide, 100 (72%) had experienced more than one, 18 (15%) had experienced more than six, and (3%) five had experienced more than ten. Looking at the number of suicides experienced by psychiatrists in the older age groups, the average number over a working lifespan was approximately four.
Table 1 Characteristics of psychiatrists responding (N = 174)
a. Included: unspecified CT1-3, liaison, perinatal, psychotherapy, rehabilitation, intellectual disability
Change in career path
Seventy-one psychiatrists reported experiencing their most impactful patient suicide as an ST 4–6 or a consultant in general adult services. Of these 71, 28 did not work in general adult services at the time of the survey. This indicates a possible 39% shift away from general adult psychiatry from the time of the death to the time of the survey. This shift might in part have reflected the usual changes during training. However, from the 16 consultants and one SAS doctor that reported experiencing their most impactful suicide when they were in a training grade in adult psychiatry, only one continued to work in general adult psychiatry. Furthermore, 33 professionals (27%) noted in the free text that they had considered a change in their career path as a result of the suicide of their patient.
‘The effect of this death heightened my determination to retire as soon as possible’
‘It was just before choosing higher training specialty, and I had applied for both adult and CAMHS, but felt quite disillusioned with adult services after the incident, which probably affected my decision to choose CAMHS.’
‘All the suicides directly influenced my career choice in not wanting to work in General Adult services.’
Effects on emotional well-being
Respondents were asked to rate the effect of the death on their emotional well-being on a Likert scale, where 0 = ‘no effect’, 50 = ‘some’ and 100 = ‘a very severe response’. Most (N = 105, 92%) rated their experience above 50, with an average rating of 66 out of 100. Only nine (8%) felt that their symptoms had met a clinical threshold for diagnosis of a psychiatric disorder at any time, although 15 (12%) were uncertain. Six (5%) had taken time off work. Leave, when taken, was of short duration (between 1 day and 1 week).
When asked to describe the primary emotional effect of the death, all 120 psychiatrists who answered this question reported experiencing painful emotions, such as sadness (N = 85, 71%); worry, anxiety and fear (N = 40, 33%); guilt and self-blame (N = 36, 31%); regret (N = 24, 20%); anger (N = 23, 19%); shame (N = 16, 13%); shock (N = 15, 12%); and blaming others (N = 4, 3%).
It was clear from the free text section that some psychiatrists had experienced serious emotional effects, for example:
‘initially shock and intrusive memories – things deteriorated after a month so tearfulness, unable to focus, intrusive thoughts and memories, nightmares resentment, shame, helplessness, finally PTSD’
‘like being deeply wounded’
‘Shame, guilt, it kept me awake at night. It affected subsequent career choices. It affected my capacity to work.’
‘I feel it's something I'll have to carry forever. I think of the boy often. And yet when it boils down to it I don't blame myself. It's the line of work, isn't it? One thought I had at the time (and still have) is that if I had another suicide then I'd resign and do something else with my life.’
Effect on clinical duties
One hundred and fifteen (98%) reported a detrimental effect on clinical practice in the period after the death, with women (mean Likert scale score = 48) affected more than men (mean = 35; P = 0.01). Forty-five (39%) respondents reported that the effect lasted between 1 week and 6 months, 18 (21%) reported that it affected them for between 6 months and 2 years, and 11 (13%) described an ongoing effect. This ongoing effect was described by some as helpful in that it increased their clinical sensitivity and awareness of their own vulnerabilities.
Ongoing thoughtfulness about patient contact – made me more vigilant and risk conscious.
Ongoing concern that although I believe that I could make valuable contribution, I feel ineffectual in changing systems that I recognize as being ineffective and fragmented… I regret that I am not more robust, but I am also now more realistic about my own limitations.
Factors that influenced psychiatrists' experiences following the deaths
The psychiatrists were asked to give a free text response about the people, interventions and experiences that either helped or did not help them in the aftermath of the death. The themes identified are shown in Table 2 and the free text responses that follow.
‘It helped knowing that some of my colleagues had been through it, that I was not the only one to feel the shame and guilt’
Table 2 What helped and what didn't help after the death
‘The patient's family were very thoughtful and supportive towards my team even in the midst of their grief. Discussing what had happened with them and attending the funeral and meeting with them again was helpful, hopefully on both sides.’
‘The suicide was upsetting, however the aftermath, the SIRI investigation and attending the Coroner's Court were very traumatic for me. It made me feel very upset, sad, angry and it felt like people were out to blame us/me/services.’
‘The sense of shame and responsibility that seems to be accepted in the whole psychiatric community about suicide didn't help.’
The psychiatrists were asked to select from a list those who supported them both inside and outside the work environment. At work, the most often selected were medical colleagues (N = 57; 46%), other disciplines (36; 30%) and managers (34; 28%). Twenty-six (21%) said they were not offered support by anyone at work. Outside work, family were reported to be helpful by 60 (49%) and friends by 35 (28%). Forty-three (35%) said they had no help from anyone outside work.
Support wanted after a patient suicide
This question included various answers, where those responding could select multiple alternatives. The seven most frequent responses are shown in Table 3.
Table 3 Support wanted after a patient suicide (N = 137)
Responsibility for the death
Respondents were asked to rate their feelings of responsibility for the death, first at the time, and then in hindsight, on a Likert scale from 0 = ‘not at all’ to 10 = ‘very responsible’. The sense of responsibility decreased with time, with the mean score falling from 42.7 (s.d. = 24.6) to 19.9 (s.d. = 17.0). There was a difference between genders, with women feeling somewhat more responsible at the time of the death (mean 46.9, s.d. = 23.2, N = 57) than men (mean 38.8, s.d. = 25.5, N = 60, P = 0.074).
Suicide prevention and the role of the psychiatrist
Psychiatrists rated the degree to which they felt that prevention of suicide was their role on a Likert scale where 0 = ‘not at all’, 50 = ‘to some degree’ and 100 = ‘very much’. The majority (N = 90, 66%) rated their responses between 50 and 100. There were 122 free text responses explaining this rating (excerpts below).
‘It is a very frightening world where one professional group is given an impossible task and then censured by society (and themselves) for failing to achieve it.’
‘The idea that suicide is always preventable (i.e. should be a ‘never event’ or we should have a ‘zero suicide’ policy) is invidious, leading to blaming of professionals. After all, if all suicides are preventable, it must be a professional's fault. On the other hand, clearly, some of the things we do reduce the risk of suicide, both long term (most obviously) and possibly short term.’
Respondents were asked how much pressure they felt from external sources to prevent patient suicide on a similar Likert scale ranging from 0 = ‘not at all’ to 100 = ‘very much’. The mean score from 134 responses was 77.6. They were then asked to what degree they thought that suicide was predictable. The midway point 50 was ‘to some degree’. Female psychiatrists thought suicide significantly more predictable than men (mean 50.0 (s.d. 16.4, N = 69) for women compared with 39.0 (s.d. 21.4, N = 64; P = 0.001) for men).
Discussion
Effects of patient suicide on psychiatrists
The results of this survey indicate a considerable emotional effect of patient suicide on psychiatrists, in line with findings from previous research.Reference Séguin, Bordeleau, Drouin, Castelli-Dransart and Giasson1,Reference Alexander, Klein, Gray N, Dewar and Eagles2,Reference Dewar, Eagles, Klein, Gray and Alexander8 General adult psychiatry was the specialty where most of the suicides occurred, and only one of the 17 psychiatrists who experienced a suicide while in training grade in adult psychiatry were still working in this specialty at the time of the survey. This suggests that the emotional effect of the death of a patient by suicide might influence specialty choice and contribute to the enduring recruitment and retention crisis in adult psychiatry. Despite reporting significant emotional effects, very few psychiatrists thought they had suffered a psychiatric disorder, such as depression or post-traumatic stress, as a result of the death. Also, very few took any time off work, which could indicate particular psychological resilience and/or a difficulty in acknowledging personal vulnerability.Reference Brooks, Gerada and Chalder10 This survey focuses specifically on psychiatrists; however, the findings are likely to apply to other mental health clinicians involved with a patient who dies by suicide. Further work is planned to examine the effects on non-medical staff.
Factors which affected the experience
Colleagues and families came out strongly as supportive factors after a patient suicide. Responses indicated that this distressing experience was made more or less painful depending on the response and attitude of the psychiatrist's trust, the family of the deceased and the coroner. If the response experienced was hostile or persecutory, it was harder for the psychiatrists to recover; if, however, it was understanding and compassionate, it was reparative for them.
Predictability and preventability of suicide
The responses indicated that most psychiatrists had high expectations about their own capacity to prevent suicide but found the weight of public expectation that they should always prevent suicide unreasonable. Our ability to predict patients at the highest level of risk is limited. Despite the ubiquity of advice to use suicide risk assessment in clinical practice, the positive predictive value is low and there is no evidence that these assessments can usefully guide decision-making. Meta-analyses indicate that no individual clinical risk factors, including suicidal thoughts and behaviours, are sufficiently accurate to be useful as the basis on which to allocate interventions.Reference Large, Ryan, Carter and Kapur11 The majority of patients who die by suicide have no contact with mental health services,12 and those that do often receive a low rating at their most recent risk assessment; this is the ‘low risk paradox’.Reference Berman13–Reference Windfuhr and Kapur15 A clinical review showed that only 2% of patients who went on to die by suicide were rated as immediate high risk in their most recent contact with services.Reference Windfuhr and Kapur15 Given this, attribution of personal responsibility for suicide prevention is clearly challenging.
Some responses directly commented on the government's focus on ‘zero suicide’,16 emphasising the different opinions that currently exist in the psychiatric community about this approach.Reference Karlsson, Helgesson, Titelman, Sjöstrand and Juth17 These respondents represented one side of this debate, expressing concern that this policy will increase the belief that suicide in secondary care is both predictable and preventable. On the other hand, there are psychiatrists who are supportive of the zero suicide movement, believing that it presents an aspirational challenge and practical framework for system-wide transformation which could help to alleviate feelings of individual blame (https://zerosuicide.sprc.org). This approach has clearly encouraged funding of suicide prevention; however, it is also crucial to manage expectations in order to avoid increasing the pressure on psychiatrists and other members of the clinical team following patient suicide.
Resources used and wanted
The results of this survey indicated a serious unmet need for support and information after these tragic and traumatic deaths. It is reasonable to think that intervention at this point may mitigate distress, effects on clinical practice and career change. The desire for information on resources for families would be relatively easily met, such as through the Help is at Hand booklet.18 Resources should also be developed to assist psychiatrists with the processes following patient suicide. There is also a clear need for support provided through both individual and group interaction.
Limitations
This survey was conducted in just four trusts, which were selected primarily through personal relationships and were restricted to the south of England. This may limit the representative nature of the experiences of the psychiatrists surveyed, compared with those working in other parts of the UK with higher suicide rates. The response rate was 34% and respondents were a self-selected group currently working in psychiatry. It was not clear whether those answering were more or less likely to have experienced a patient suicide. A number of responses detailed a desire to retire, change specialty or leave psychiatry altogether as a response to the suicide. It could be possible that some of the psychiatrists who were most severely affected would have chosen to leave psychiatry, and as a result would not be represented in this survey. The responses may be subject to some degree of bias due to social desirability, although the quotes from respondents suggested considerable frankness of expressed views. The choice of sources of support and help were based on a pre-determined list of options and may not have covered all possibilities.
Conclusions
Respondents to this survey confirmed that the death of a patient by suicide can have very strong emotional effects on psychiatrists. Their clinical practice was often negatively affected for a period of time, and over a quarter considered a change of career path as a result. There were differences in responses between genders in three areas. Women were more likely than men to believe that suicide is preventable, experience greater feelings of being responsible for the death and report a greater effect on clinical confidence for a period afterwards. The findings of the survey suggest that there is a need for a broad range of information and support to both prepare psychiatrists for the possible occurrence of patient suicide and provide helpful support following such incidents.
Acknowledgements
We thank Mark Hancock, Jonathan Bindman, Minoo Irani and Vincent Kirchner (Medical Directors) for approving the survey; Alexandra Pitman, Kate Saunders and Helen Killaspy for advice; Brad Mullins for assistance with the analyses; and Nora Turjanski, Siobhan Jeffreys, Nisha Shah and Rob Hale (consultants from the Suicide Group for Consultants) for piloting the survey.
About the authors
Rachel Gibbons is the Therapy Director of the Priory Group, and at the time of this work she was a Consultant Adult Psychiatrist and the Suicide Lead for Barnet Enfield and Haringey Mental Health Trust, UK. Fiona Brand is a psychiatric nurse with a clinical role in psychiatric liaison at Oxford Health NHS Foundation Trust and a research nurse role at the Oxford University Centre for Suicide Research, UK. Anne Carbonnier is a Consultant Child and Adolescent Psychiatrist in South Oxfordshire Child and Adult Mental Health Services (CAMHS) and Medical Lead for South Oxon CAMHS, UK. Alison Croft is a Consultant Clinical Psychologist with the Medic Support Service, Oxfordshire Psychological Services and Oxford Cognitive Therapy Centre, Oxford Health NHS Foundation Trust, UK. Karen Lascelles is a Nurse Consultant in Suicide Prevention, Oxford Health NHS Foundation Trust, UK. Gislene Wolfart is the Lead Psychologist at Oxford Mind & Body, and also holds a post in the Specialist Eating Disorder Service, Oxford Health NHS Foundation Trust, UK. Keith Hawton is Director of the Centre for Suicide Research at Oxford University Department of Psychiatry, and Consultant Psychiatrist, Oxford Health NHS Foundation Trust, Warneford Hospital, UK.
APPENDIX 1
Survey questions
1. Are you? Male, Female, Non binary
2. What is your age? 22–28, 29–34, 35–44, 45–54, 55–64, 65–74, 75 or older
3. How many years have you been working in psychiatry? 1–3, 4–6, 6–10, More than 10
4. What level are you currently working at? FY, CT, ST, Consultant, SAS, Other
5. Which age group of patients do you work with? Child and Adolescent, General Adult, Older Adult, Other (please specify)
6. Your area/s of speciality?
7. What setting do you work in?
8. Have you ever been working with a patient who died by suicide or suspected suicide?
9. How long ago was this? In the last three months, Between 3 and 6 months ago, Between 6– and 18 months ago, Between 18 months and 3 years ago Between 3 and 5 years ago Between 5 and 10 years ago, Over 10 years ago
10. When in your career did this suicide happen? Medical Student, Foundation, CT1-3, ST4-6, Consultant 1–3, Consultant 4–9, Consultant 10 or more, Staff Grade/Associate Specialist/Speciality Doctor
11. Please specify the service/client group in which you treated or encountered this patient.
12. In which setting did this suicide occur?
13. Approximately how many patients did you have some clinical responsibility for at that time?
14. Consider the effect that this patient's death had upon your own emotional well-being. Please rate the greatest extent to which it affected your emotional well-being at any point following the event.
15. Please use the box below to describe the effect it had on you. What type of emotions you experienced and if possible why you felt that way? (e.g. shame, sadness, anger, contempt, fury, regret, relief, paranoia etc)
16. At the time what extent did you feel responsible for the suicide/suspected suicide? Some psychiatrists feel responsible even where an event was judged to have been beyond their control.
17. With hindsight how responsible do you feel you were for this death now.
18. If this event affected your mental health do you think your symptoms met the clinical threshold for the diagnosis of a disorder at any point as a consequence of you patient's suicide?
19. At the most difficult point do you feel that this had a detrimental effect on your confidence in your clinical practice?
20. Approximately how long did this effect last?
21. Did the experience of losing a patient by suicide have an effect on your ability to carry out your clinical duties in the longer term?
22. Did you ever consider or act in any way to change your career path as a consequence of a patient suicide?
23. If you were working within a Trust or private hospital, how helpful do you rate the level of support that was provided by the Trust or private hospital at the time?
24. Did you take any annual leave or sick leave because of your experience?
25. Please indicate who you were offered support at work by:
26. Did you access support outside work?
27. How familiar/knowledgeable were you then with the process that takes place following a patient suicide?
28. At the time of the death which procedures were you not prepared for? (Select all that apply)
29. Were you offered advice/help/support with the processes that follows a patient suicide? (Select all that apply)
30. Did anything help make this event easier to cope with?
31. Did you find anything unhelpful?
32. If you were involved in the Trust or organisation's internal enquiry did you find it helpful or unhelpful?
33. How familiar/knowledgeable are you now with the process that takes place following a patient suicide?
34. Please select the procedures that you are not currently familiar/knowledgeable about now? (Select all that apply)
35. One of the aims of this survey is to identify doctors' specific needs and develop better support after the loss of a patient by suicide. If any of the following were available, which do you think would be helpful for doctors affected by the death of patient by suicide (select all that apply):
36. To what degree do you believe preventing suicide is the role of the psychiatrist?
37. Why do you think you gave the answer you did to question 36?
38. To what degree do you feel under pressure (by external sources) to prevent suicide in your role of the psychiatrist?
39. To what degree do you think suicide is predictable in secondary care?
40. Which Trust do you work for (optional)
41. If you do not give your permission to use anonymised extracts of your answers please select this box
42. Do you have any other ideas about what might be helpful for psychiatrists or other clinicians after a patient suicide (free text)?
Open access
eLetters
No eLetters have been published for this article.