As practising National Health Service (NHS) psychiatrists we can all expect at least one, and probably several, patients in our care to commit suicide over the course of our careers. The existing literature, which is mainly from North America and is generally anecdotal, suggests that patient suicide can be especially distressing (Reference Goldstein and BuongiornoGoldstein & Buongiorno, 1984; Reference Chemtob, Hamada and BauerChemtob et al, 1988; Reference Kaye and SoreffKaye & Soreff, 1991; Reference MenningerMenninger, 1991; Reference ValenteValente, 1994).
Chemtob et al (Reference Chemtob, Hamada and Bauer1988), in a survey of 643 psychiatrists in the USA achieved, after considerable effort, a 46% response rate. Of the respondents 51% reported experience of suicide of a patient in their care and many found such experiences to be particularly upsetting and to have an impact upon their professional and personal lives. Particularly distressing experiences of patient suicide were related to psychiatrists being younger, having lower levels of training and having spent fewer years in psychiatry. This would suggest that patient suicide might be especially traumatic for psychiatrists in training. Chemtob et al (Reference Chemtob, Hamada and Bauer1988) concluded that informal supports were of most help in enabling individuals to deal with the considerable “post-trauma symptoms” which followed patient suicide, that suicide of a patient was “a very real occupational hazard” for psychiatrists, and that training to prepare for patient suicide was lacking.
Our NHS differs in many respects from health services in North America and thus we do not know if patient suicide has a similar impact on NHS psychiatrists to that described by Chemtob et al (Reference Chemtob, Hamada and Bauer1988) and others. Although the UK literature in this area is minimal, anecdotal and tends to focus upon effects of suicide on nursing staff (e.g. Reference FarringtonFarrington, 1995), many NHS trusts now have critical incident review procedures apparently designed both to address staff's emotional needs and to ascertain deficits in care or disciplinary issues. Are these procedures just paying lip-service to staff's needs? Do unrealistic public expectations fuelled by political policy engender a ‘blame culture’? We decided to examine Scottish trainee psychiatrists' experiences of and reactions to suicide of their patients and to attempt to delineate issues of particular importance to be considered in constructing guidelines for the management of such incidents.
Following extensive consultation and a local pilot study. A 13-page questionnaire was developed. It comprised four sections: personal details; general views; the ‘most distressing’ suicide; and finally a free-text section for any individual thoughts and comments. Further details on the information sought in the questionnaire are summarised in Table 1.
|Personal details||Age, gender, time in psychiatry|
|General views||Training on and experience of patient suicide|
|The ‘most distressing’ suicide|
|Patient's details||Age, diagnosis, history of self-harm, respondent's involvement, legal status, where incident occurred|
|The suicide||When it occurred, method, if the body was seen|
|Consequences||Impact on personal and professional life, events (e.g. critical incident review) and individuals (e.g. family, general practitioner) involved in coming to terms with the suicide, time off work, effects on career plan;|
We targeted this questionnaire at trainees across all psychiatric specialities working as senior or specialist registrars in Scotland. Names and addresses were obtained from the Royal College of Psychiatrists, from postgraduate deans, from colleagues who had themselves recently surveyed higher trainees and through personal contacts generated a mailing list which we believe to have been comprehensive.
In view of the sensitive nature of some of the questions the support of the College was obtained before mailing the questionnaire. In order to protect anonymity, the researching clinicians (I.G.D., D.A.A. and J.M.E.) chose not to read individuals' responses. The non-clinical researchers (S.K. and N.M.G) were privy to this information, since it was necessary to identify respondents so that a repeat mailing could be sent to non-responders.
Personal details and general views
Of the 128 trainee psychiatrists identified for the survey, we received completed questionnaires from 103, representing a response rate of 81%. The duration of their experience in psychiatry ranged from 3-13 wholetime equivalent years.
While 96% of respondents had received training in suicide risk assessment, only 28% recalled any training in issues to be considered following patient suicide. All of the latter group rated this experience as moderately or extremely useful. Ninety per cent of trainees who had received such training reported that this included information on the potential effects of patient suicide on staff.
Although we had considered the possibility that relatively few trainees would report intimate experience of patient suicide, 47% reported the suicide of a patient under their care and a further 40% had been “closely involved in the consequences of the suicide” of a patient not directly under their care (e.g. out of hours duties). The number of suicides experienced ranged from 1-5.
The most distressing suicide
Trainees who had experienced more than one suicide were asked to focus upon their most distressing suicide in terms of its personal impact on them. The patients involved were predominantly young adults, 59% being in the 14-40-year-old age group. Fifty-nine per cent were male, and 62% had a known previous history of self-harm. Sixty-five per cent were current in-patients and 24% were detained. Ninety-three per cent of in-patients were on general observations only, that is, were not deemed to be at high risk of imminent suicide. The diagnoses encompassed all main areas of psychiatric practice.
The methods of suicide reported were as follows: hanging, 26%; jumping, 22%; poisoning/overdose, 17%; drowning, 9%; laceration, 5%; firearm, 3%; and other, 15%. None had died by asphyxiation from exhaust gases. The prominence of violent deaths may reflect the high numbers of in-patient suicides. Only 9% of respondents saw the body after death.
Why was this particular suicide experienced as being particularly distressing? For some, it was the only suicide that had been experienced, but those who reported more than one patient suicide frequently noted that their first experience of patient suicide had been the most distressing. The other particularly frequent theme was that the suicide was unexpected and had not been predicted, occurring when the patient seemed to be improving or making plans for the future. Several respondents felt to blame in some way. Others noted that the victim was young and on occasions had young children. Less frequently cited reasons for the degree of distress occasioned by the suicide were diverse but included knowing the patient well, disliking the patient, the method of suicide, the blame of relatives and having been the last to speak to the deceased.
Adverse effects on personal and professional life
Thirty-one per cent of trainee psychiatrists reported that the suicide had an adverse impact on some aspect of their personal lives. The most commonly reported effect was a continuing preoccupation about the suicide and how it could have been prevented. Also frequently mentioned were problems with anxiety, guilt, insomnia and loss of confidence.
Thirty-nine per cent recalled the suicide adversely affecting their work. Many reported increased anxiety and difficulty in making decisions, particularly when this involved patients with recognised increased risk of self-harm. Management became over-cautious, specifically when deciding on observation levels, passes and discharge for in-patients. Nine per cent of respondents reported giving consideration to a change of career. A few psychiatrists decided not to pursue careers in general adult psychiatry because of its perceived higher risk of patient suicide.
Very few doctors took any time off work as a consequence of the suicide.
Post-suicide management: what happened and what helped
When team meetings/reviews or critical incident reviews did occur most trainees reported them to have been personally helpful. Twenty-eight per cent indicated that the team meeting/review was ‘not applicable’ and 56% stated the same for the critical incident review. This suggests that in some instances these post-incident procedures do not occur or trainees do not attend them.
Many doctors suggested that it was not the formal supports that were of most value, but rather the informal. Most had discussed the death with involved team colleagues (95%), their own consultant (92%), family/partner (82%) and friends (69%). Other sources were seldom relevant, and notably, their general practitioner was consulted by only 2% of trainees. The perceived benefit from the individuals who may have been involved in assisting the trainee to come to terms with the suicide varied. Family and friends were usually helpful and never unhelpful. Team colleagues were often helpful but not always. The trainees' own consultants fare strikingly variably, being the most frequently cited ‘very helpful’ and ‘very unhelpful’ individuals: 39% thought the consultant's involvement had been ‘very helpful’ but 6% suggested the consultant had been ‘very unhelpful’.
Fatal accident inquiries were usually perceived as neither helpful nor unhelpful. Nineteen per cent of trainees were aware of press publicity about the suicide and only 33% of these trainees found this to be ‘moderately distressing’. Other events such as disciplinary and legal proceedings were exceedingly rare.
Open text comments
To maintain anonymity it is not appropriate to provide very detailed quotations, but the list below gives some of the themes which were characteristic of the points made by respondents in the open text sections.
(a) I felt terrible afterwards.
(b) No other doctors are made to feel so personally responsible or guilty at having a patient die as a result of their chronic illness.
(c) ‘If we don't see people who go on to commit suicide we are not seeing the right people’ is an oft quoted but extremely helpful comment.
(d) Trusts have protocols for these incidents. They don't get looked at until they're needed.
(e) There is a long held belief that we should all be able to deal with it ourselves and ‘get on with it’.
(f) I found personal approach by a consultant very helpful.
(g) It was helpful to hear senior colleagues discuss their own experiences - this made it easier to keep a sense of proportion about my own reactions.
(h) The lack of senior colleague interest in the junior doctor's experience of the process as a whole continually surprises me…I got better support from the Sheriff's office…
(i) The critical incident review should not be a scape-goating exercise.
This first UK study of the impact of patient suicide among psychiatrists in specialist training confirms some of the conclusions of previous work done overseas. Almost half of the respondents had close professional experience of suicide, and the findings are strengthened by the 81% response rate. It is clear that these experiences give rise to significant occupational stress, with potential adverse consequences upon the psychiatrists' work and personal lives. On occasions the suicide has been sufficiently distressing to encourage the doctor to consider a change of career. It is possible that some doctors did leave psychiatry or opted for a non-consultant post and thus have not been included in our study. Therefore, our findings may represent an underestimate of the impact of patient suicide on trainee psychiatrists.
Despite the frequency of experiences of patient suicide, only a small minority had been prepared by previous training for the potential consequences or the organisational procedures which follow suicide. Given that trainees who had received such training found it to be of value, and noting similar results from other studies in this area (e.g. Reference Chemtob, Hamada and BauerChemtob et al, 1988), we would contend that such training should be provided in postgraduate education programmes for psychiatrists. This should better prepare our junior psychiatrists to deal with patient suicide when it does occur.
The effects of stress and the risk of burn-out among the medical profession are factors which have been increasingly recognised over recent years (Reference CaplanCaplan, 1994; Reference Ramirez, Graham and RichardsRamirez et al, 1996; Reference Guthrie, Tattan and WilliamsGuthrie et al, 1999). As a buffer against work-related stress, the support which junior staff perceive from consultants may be crucial (Reference Firth-COZENSFirth-Cozens, 1987). In a large sample of psychiatrists working in Sweden and in Birmingham, Thomsen et al (Reference Thomsen, Dallender and Soares1998) concluded that support with work-related problems was crucial to their concept of a healthy workplace, and it seems that community-based mental health staff are more stressed than those based in hospital which may relate to the increased professional isolation among the former group (Reference Prosser, Johnson and KuipersProsser et al, 1996). Each of these findings underscores the need for interpersonal support among psychiatric professionals in general, and the provision of appropriate support and assistance at a time of considerable personal trauma (such as patient suicide) might be taken as a ‘marker’ of an appropriate working environment for trainees in psychiatry.
Only three-quarters of trainees appeared to have discussed the suicide and its aftermath at a team meeting, and less than half had attended a critical incident review. These events have other functions in addition to the provision of support and, given that trainees reported that they are helpful, this should reinforce the need for their routine occurrence following suicides. The trainees' consultant appears to be particularly influential, having an opportunity to provide potentially valuable advice. Most consultants appear to do well in supporting their trainees at these times, but some do poorly. Given the theme of self-blame which arose commonly, it would seem especially important that trainee psychiatrists are helped to disentangle issues of clinical management from less rational feelings arising from concerns about personal failure and responsibility. Recognition by all our consultant colleagues of the potential impact of patient suicide on training grade psychiatrists would, therefore, be of value, and trainee education could usefully include the issue of how, as consultants, support can be optimally provided.
We thank all colleagues who completed the lengthy questionnaire. The authors acknowledge the financial support of Grampian Healthcare NHS Trust. We are grateful to colleagues, including office bearers in the Royal College of Psychiatrists, who commented on our final draft of the questionnaire, and to those who provided trainees' names and addresses.