The literature concerning psychiatrists' responses to patient suicide is sparse (Reference BrownBrown, 1987, Reference Chemtob, Hamada and BauerChemtob et al, 1988; Reference Alexander, Klein and GrayAlexander et al, 2000) but even less attention has been given to the psychiatrist's role in the aftermath of such an event. Psychiatrists infrequently discuss their own experience of patient suicide with their colleagues, either at an individual level (Reference Kaye and SoreffKaye & Soreff, 1991) or in group settings such as team meetings (Reference Ruben, Blumenthal and KupferRuben, 1990). This is all the more remarkable when one acknowledges that the suicide of a patient is arguably the event that causes most concern for clinicians, irrespective of their experience or seniority (Reference Kaye and SoreffKaye & Soreff, 1991). As Brown (Reference Brown1987) reminds us, working with the suicidal is intense and demanding. The stakes are high and psychiatrists work with the constant background pressure that arises from the real possibility that the patient may commit suicide during treatment. Patient suicide is a relatively frequent event (Reference Kozlowska, Nunn and CousensKozlowska et al, 1997) and may even be regarded as an occupational hazard (Reference Chemtob, Hamada and BauerChemtob et al, 1988). The reluctance of the profession as a whole to address systematically the issues that arise in the aftermath of patient suicide has far-reaching implications for mental health professionals, the families of patients who have committed suicide and, of course, the patients themselves.

The psychiatrist

Most psychiatrists will experience the suicide of a patient during their professional career. In their questionnaire survey of consultant psychiatrists in Scotland, Alexander et al (Reference Alexander, Klein and Gray2000) reported that 68% had a patient who committed suicide under their care. Looking at trainees in particular, for whom the impact of suicide may be greater, Dewar et al (Reference Dewar, Eagles and Klein2000) found that 47% had experienced the suicide of a patient and many trainees consider it one of the most adverse experiences of their training (Reference Kozlowska, Nunn and CousensKozlowska et al, 1997). Similarly, Courtenay and Stephens (Reference Courtenay and Stephens2001) report that 54% of their sample of trainees had a patient under their care who had committed suicide, a figure consistent with a rate of 0.23 suicides per training year. It is likely that even these figures are conservative estimations (Reference HennHenn, 1978).

As one might expect, the impact of a patient suicide on psychiatrists is enormously variable and dependent on several factors, not least the nature and extent of the doctor's relationship with the patient (Reference LitmanLitman, 1965) and the support provided by the context within which he/she works (Reference Jones, Dunne, Mcintosh and Dunne-MaximJones, 1987). Nevertheless, some universal features are apparent and, in particular, several authors have noted that psychiatrists appear to react on two different levels, namely personal and professional (Reference LitmanLitman, 1965; Reference Jones, Dunne, Mcintosh and Dunne-MaximJones, 1987; Reference Ruben, Blumenthal and KupferRuben, 1990).

On a personal level, the psychiatrist's reaction is related to the experience of loss (Reference Ruben, Blumenthal and KupferRuben, 1990) and, as such, there may be features common to any bereavement reaction. Initially there are often feelings of shock or disbelief (Reference Kozlowska, Nunn and CousensKozlowska et al, 1997), which may evolve into denial that the death has taken place or that the death was in fact suicide. Anger may become prominent and may be directed towards the deceased or his/her family (Reference Jones, Dunne, Mcintosh and Dunne-MaximJones, 1987). Feelings of sadness or loss also are common and may be associated with other features of depression (Reference Jones, Dunne, Mcintosh and Dunne-MaximJones, 1987).

Psychiatrists also must deal with the loss on a professional level and frequently react in a way that is related to their special role (Reference Jones, Dunne, Mcintosh and Dunne-MaximJones, 1987). Feelings of guilt and self-doubt (Reference Kozlowska, Nunn and CousensKozlowska et al, 1997), which may be both strong and persistent, are frequently associated with relentless self-questioning about whether something important had been missed (Reference Ness and PfefferNess & Pfeffer, 1990). Such guilt and self-doubt may be exacerbated by anger and blame from others, such as the deceased's family and legal representatives (Reference Ruben, Blumenthal and KupferRuben, 1990). In contrast, more narcissistic defences may be manifest, such as the denial of feelings of guilt or failure, adoption of an air of indifference and, perhaps, even blaming of others (Reference Ruben, Blumenthal and KupferRuben, 1990).

Fear concerning the family's reaction to the suicide and the prospect of litigation may cause anxiety (Reference Kozlowska, Nunn and CousensKozlowska et al, 1997). This fear, together with that of censure from colleagues or of adverse publicity, may lead to increasing withdrawl and isolation (Reference Jones, Dunne, Mcintosh and Dunne-MaximJones, 1987). As Wu (Reference Wu2000) points out, although there is a norm of not criticising, reassurance from colleagues often is grudging or qualified. Levels of distress may escalate to the extent that psychiatrists question their ability to cope and they may fear that they will break down psychologically (Kozlowska 2, 1987). Indeed, 57% of Chemtob et al's (Reference Chemtob, Hamada and Bauer1988) sample of psychiatrists who had experienced the suicide of a patient reported post-traumatic symptoms comparable with those found in clinical groups. A sense of impotence and powerlessness may cause psychiatrists to feel demoralised and disheartened. They may question the ability of their profession to make a positive difference and may even consider leaving the speciality.

The experience of patient suicide may be particularly traumatic for trainees, who in general are more inexperienced, may feel unsupported by senior staff (Reference Kozlowska, Nunn and CousensKozlowska et al, 1997; Reference Dewar, Eagles and KleinDewar et al, 2000) and are more likely to perceive the suicide as a personal rather than professional failure (Reference BrownBrown, 1987). They frequently report feeling ill-prepared and that they are left to their own resources to try to assimilate the experience (Reference Chemtob, Hamada and BauerChemtob et al, 1988). In their sample of trainees, Courtenay and Stephens (Reference Courtenay and Stephens2001) found that 46% of those who had experienced the suicide of a patient had expressed the need for help in coping with the associated emotional upset. In 40% of these cases this need was ignored, despite the fact that those who were offered it in the period following a suicide appreciated almost any form of support.

Irrespective of seniority, the impact of patient suicide on psychiatrists is compounded by their need to continue to minister to the needs of others at a time of great personal distress (Reference Jones, Dunne, Mcintosh and Dunne-MaximJones, 1987). Professional responsibilities may disrupt their personal grieving and it may be intensified by the need to maintain a professional balance (Reference Ruben, Blumenthal and KupferRuben, 1990). Almost invariably they must return to work immediately, yet Ruben (Reference Ruben, Blumenthal and Kupfer1990) argues that psychiatrists should accept that the suicide of a patient would have an unavoidable negative impact on professional functioning for a short period of time at least. The impact of patient suicide often results in an inability to concentrate, associated with intrusive ruminations and a pressing need to talk over what happened with colleagues. Doubts over professional competence and fear of repetition may result in psychiatrists becoming overcautious with subsequent patients who are suicidal or may impair their ability to deal with patients even if they are not suicidal (Reference Jones, Dunne, Mcintosh and Dunne-MaximJones, 1987).

Although the suicide of a patient can affect psychiatrists deeply on both a personal and professional level, Kaye and Soreff (Reference Kaye and Soreff1991) emphasise that, despite the lack of guidance from the literature, it is important to prioritise the various tasks that must be undertaken in the aftermath of patient suicide and compile these into a checklist. Perhaps the first task is to inform others of the event. Naturally this includes the family and the staff directly involved in the patient's care, but also the various other individuals and agencies involved, such as the general practitioner (GP), senior medical and nursing staff, hospital management, and, if relevant, the police (Reference Hodelet and HughsonHodelet & Hughson, 2001). In our experience it can be useful to write a detailed letter to the GP very shortly after the suicide. This letter should summarise the patient's clinical history prior to the suicide, describe the clinical team's work with the patient and their estimation of the patient's suicide risk and the treatment plan at the time of the suicide. Such a letter can be a useful source document in subsequent reviews or enquiries into the patient's suicide. The psychiatrist also must liaise with the Coroner's Office (Reference Hodelet and HughsonHodelet & Hughson, 2001). He/she may be required to provide a report and also give evidence in the Coroner's Court.

Although it is paramount that the psychiatrist be available for others, he/she must also secure support for him-/herself. This is critical, as Ness and Pfeffer (Reference Ness and Pfeffer1990) emphasise: “with striking unanimity [psychiatrists] have said that formal and informal consultation with colleagues is one of the most important and helpful actions to take in coping with a patient's suicide”. It is, therefore, crucial that they are aware of the sources of support available to them. Initially, it is helpful to review the case with a supervisor or mentor, particularly if he/she has had a similar experience (Reference Jones, Dunne, Mcintosh and Dunne-MaximJones, 1987), and to reflect on questions that may arise (Reference Kaye and SoreffKaye & Soreff, 1991). Reviewing the case in this way also may act as a check against the development of possible countertransferance phenomena (Reference Dunne, Dunne, Mcintosh and Dunne-MaximDunne, 1987). More informally, it may be helpful to talk with staff and family members and also take time to reflect alone (Reference Jones, Dunne, Mcintosh and Dunne-MaximJones, 1987). Binder (Reference Binder1978) describes the benefits of groups support and found that group self-examination in a supportive atmosphere reduced feelings of being ‘alone’ and provided a supportive environment in which to work through feelings regarding the death. Ultimately this resulted in diminished feelings of pain and loneliness and a greater willingness subsequently to treat other suicidal patients. Advice and support also may be sought from the British Medical Association, The Royal College of Psychiatrists and the National Counselling Service for Sick Doctors (Reference Hodelet and HughsonHodelet and Hughson, 2001). None the less, in certain cases it may be appropriate to seek more formal, professional help and, importantly, Ruben (Reference Ruben, Blumenthal and Kupfer1990) cautions against self-medication.

Psychiatrists also should give serious consideration to attending the patient's funeral. The funeral may provide an important opportunity to grieve in a group setting and, as Kaye and Soreff (Reference Kaye and Soreff1991) point out, attendance is by no means an admission of responsibility and is often appreciated by the family.

In the longer term, psychiatrists have an important duty to participate in local audit proceedings. This is an opportunity to reflect on how one might act differently if a similar situation arose again and serves also to identify deficiencies in practice or service provision with the ultimate aim of preventing further suicides (Reference MorganMorgan, 1993; Reference Lelliot, Jenkins, Griffiths and WylieLelliot, 1994). The psychiatrist most likely to have knowledge of the circumstances surrounding the deceased's illness and death also will be contacted by the National Confidential Enquiry into Suicide and Homicide by People with Mental Illness. The enquiry aims to determine, by means of a questionnaire, whether or not, in hindsight, there were any ways in which the likelihood of the death could be decreased (Reference Boyd, Jenkins, Griffiths and WylieBoyd, 1994). It uses this confidential information as the basis of recommendations regarding different approaches to clinical management or provision of care that may prevent patient suicide (Reference BoydBoyd, 1993).

The family

The nature of the way in which those who commit suicide die — often violent, usually sudden or unexpected (Reference Hauser, Dunne, Mcintosh and Dunne-MaximHauser, 1987) — and the stigma associated with the death means that the families of the deceased often experience great difficulty in coming to terms with their bereavement (Reference HawtonHawton, 1986). Normal coping mechanisms are especially likely to be thrown into disarray (Reference WertheimerWertheimer, 1991) and Hauser (Reference Hauser, Dunne, Mcintosh and Dunne-Maxim1987) argues that there is an “almost certain likelihood of disturbances in the grief process”. This may be particularly likely if there is preexisting family pathology or ongoing stressors (Reference RetterstolRetterstol, 1993). In addition to the normal grief reaction there is often a greater sense of shock followed by a protracted search for an explanation (Reference Kaye and SoreffKaye & Soreff, 1991). Alternatively, there may be an implicit agreement within the family not to discuss the death, or the subject may be avoided for fear of upsetting or burdening others (Reference HawtonHawton, 1986). This may lead to feelings of isolation, which are exacerbated by the sense of shame associated with the intense stigma of suicide. Prolonged feelings of guilt may be intensified by the inevitability of the death and the sensation of ‘knowing’ that the suicide would occur sooner or later, which Scott (Reference Scott1989) describes. Some families describe a feeling of relief, especially if the death is preceded by chronic stress, suicidality and aggravation (Reference HawtonHawton, 1986). This, too, can exacerbate feelings of guilt.

It is important to remember that the impact of a suicide within the family is extremely variable and can affect individuals in different ways. For example, the reaction of fathers of children who have committed suicide may be predominantly hostile with evidence of denial and a strong tendency to blame external factors for the death (Reference HawtonHawton, 1986). Parents may develop a ‘parental identity crisis’, with prominent self-doubt resulting in difficulties in their relationships with surviving children (Reference Mcintosh, Dunne, Mcintosh and Dunne-MaximMcIntosh, 1987). Others may appear unaffected initially because they are ‘buoyed up’ by necessary practical matters, only to experience difficulties later on when support may be less available (Reference ScottScott, 1989). In the aftermath of suicide, it is important to realise that some of the deceased's relationships may remain hidden and such people will find it even more difficult to obtain support (Reference Morgan, Jenkins, Griffiths and WylieMorgan, 1994).

The families of those who have committed suicide are more likely to need professional help during the grieving process and may be at an increased risk of suicide themselves (Reference Ness and PfefferNess & Pfeffer, 1990). However, this excess risk is difficult to assess because there are often confounding factors, such as a family history of depression. The reaction of others towards a family in which there has been a suicide is another factor that influences the grieving process. Society continues to have a judgmental attitude towards suicide (Reference RetterstolRetterstol, 1993). Friends often do not know what to say to the family (Reference ScottScott, 1989) and instead withdraw (Reference RetterstolRetterstol, 1993) or avoid them (Reference Ness and PfefferNess & Pfeffer, 1990). The loss of previous sources of social support inhibits mourning and perpetuates the sense of stigma associated with suicide. The modern notion that suicide is no longer primarily a personal failure but a family failure (Reference Colt, Dunne, Mcintosh and Dunne-MaximColt, 1987) is reflected in the observation that families of those who have committed suicide tend to be blamed more, are seen as more emotionally disturbed (Reference HawtonHawton, 1986) and are more often viewed as accountable for the death (Reference Rudestam, Dunne, Mcintosh and Dunne-MaximRudestam, 1987). Such views socially reinforce the feelings of guilt and self-blame that already preoccupy the family (Reference Ness and PfefferNess & Pfeffer, 1990). The perception that the public is unsympathetic or unresponsive to the family's emotional needs hampers their adjustment to the death and also may result in their inadequate use of mental health resources (Reference Rudestam, Dunne, Mcintosh and Dunne-MaximRudestam, 1987).

At a time of great distress the family of the decreased also must undertake various practical tasks. They may be required to identify the body of the decreased. Almost invariably a post-mortem examination is undertaken (Reference ScottScott, 1989), which is just one factor in the inevitable delay in making funeral arrangements after a suicide (Reference Barraclough and ShepherdBarraclough & Shepherd, 1977). The registration of the death and funeral or memorial service must fit in with the schedule of the Coroner's Office (Reference Dolman, Jenkins, Griffiths and WylieDolman, 1994) and this can cause additional strain on the family. The Coroner's hearing itself is often found to be traumatic for family members. In Barraclough and Shepherd's (Reference Barraclough and Shepherd1977) sample, a family member was present at 96% of such hearings. They found that the most distressing aspects of the hearing for the family were giving evidence in front of strangers, the judicial atmosphere and the reading of the post-mortem report. Barraclough and Shepherd (Reference Barraclough and Shepherd1976) argue that the distress caused by the hearing “could be eliminated almost entirely by changes of procedure”, and Chambers (Reference Chambers1989) goes as far as to suggest that the necessity for an inquest in all cases of suicide should be removed. A further aspect of the Coroner's hearing that relatives found distressing was the presence of the local media, whose reports often were sensationalist, intrusive and inaccurate, serving only to enhance the stigma associated with suicide (Reference Barraclough and ShepherdBarraclough & Shepherd, 1976; Reference Dunne-Maxim, Dunne, Mcintosh and Dunne-MaximDunne-Maxim, 1987).

With this in mind, psychiatrists have a valuable role in helping the family of a patient who has committed suicide to come to terms with their loss. There is no single formula (Reference Morgan, Jenkins, Griffiths and WylieMorgan, 1994) and, just as each family member reacts to the death of his/her relative in an individual way, his/her needs are equally varied and psychiatrists must be sensitive to this. Contact should be made with the family, in person, as soon as possible and preferably within 24 h (Kaye & Soreff, 1990). This initial contact is critical, because early help and support has been shown to reduce the displacement of anger onto the psychiatrist (Reference Ruben, Blumenthal and KupferRuben, 1990). Ruben (Reference Ruben, Blumenthal and Kupfer1990) emphasises that the greatest mistake that psychiatrists can make is to avoid the family and that contact should be made even if there had been only minimal contact previously. If, for whatever reason, the psychiatrist feels unable to offer the family support, contact still should be made with the family even if it is simply to refer them on to a colleague. Hawton (Reference Hawton1986) suggests that the first meeting primarily should be supportive, concentrating on practical issues and spending time listening in a constructive way. The family should be given ample opportunity to ask questions and it may be worthwhile to discuss the efforts made to treat their relative prior to his/her suicide and emphasise that realistically all that could be done for the patient was done (Kaye & Soreff, 1990). The atmosphere of the meeting must permit, and even encourage, the venting of anger and hostility (Reference HawtonHawton, 1986). It may be helpful at this time for the psychiatrist to acknowledge his or her own personal grief to the family (Reference Dunne, Dunne, Mcintosh and Dunne-MaximDunne, 1987) and acknowledge that the grieving process for the family may be more complex and painful that normal (Kaye & Soreff, 1990). Some assessment of the family's expectations of help in the future should be made and the psychiatrist should leave a contact number. As such, the family can dictate the nature and frequency of further meetings from, perhaps, weekly, to an ‘on required’ basis (Reference HawtonHawton, 1986). It may be appropriate also for contact to be made with the family at potentially stressful times, such as the anniversary of the death (Reference HawtonHawton, 1986). In the immediate aftermath of the suicide, the family should be supported in making positive decisions about practical matters but advised not to make any major decisions at a time of such emotional turmoil. It may be helpful to explain the natural course of grief and emphasise that it may take some time (Reference ScottScott, 1989).

Support for the family in the long term may take several forms and the families should choose the type of assistance that they feel able to accept (Reference Morgan, Jenkins, Griffiths and WylieMorgan, 1994). Discussing suicide itself and its effects on the family is often helpful (Reference WertheimerWertheimer, 1991) and can be expanded to include an exploration of how the family think the suicide might affect them in the long term (Reference HawtonHawton, 1986). More structured, cognitive approaches can be used, such as reality testing of the relatives' feelings of guilt or correcting cognitive distortions to develop a more realistic view of the deceased (Reference HawtonHawton, 1986). A group counselling approach where relatives can share and vent feelings with those who have experienced a similar bereavement may be useful. Such groups provide a source of support and reassurance and a safe environment in which to express hostility and anger (Reference HawtonHawton, 1986). This has been shown to reduce the incidence of depression in relatives and to increase their ability to plan for the future.

In considering the needs of the family, psychiatrists must be aware of the possible obstacles to finding support, not least the family's ambivalence towards mental health services (Reference WertheimerWertheimer, 1991). Anger may be projected onto the psychiatrist for providing inadequate treatment or for giving the patient too much freedom (Reference RetterstolRetterstol, 1993), and this can influence the family's likelihood of accepting help from mental health services. At such times it may be appropriate to refer the family to a colleague or to other sources of support. The GP, who has often known the family for many years, can be particularly helpful, as can the Samaritans at times of crisis. Several more specialised self-support groups also exist, such as Shadow of Suicide and Bereavement by Suicide, which can offer support and friendship at a local level. It is often useful to involve the clergy, who are able to provide comfort and support and perhaps provide answers that medicine cannot (Kaye & Soreff, 1990).

Colleagues

The suicide of a patient may have a profound effect not only on psychiatrists but on all members of nursing staff and other multi-disciplinary team members who came into contact with the patient. It is, therefore, an important responsibility of the psychiatrist to inform all staff of the event as soon as possible. In addition to the typical grief reaction, staff may fear being held responsible and may question their own skills and competence (Reference Ruben, Blumenthal and KupferRuben, 1990). Many will review their actions and interactions prior to the suicide and it is important to encourage group support and the venting of emotions, possibly through informal meetings (Reference Kaye and SoreffKaye & Soreff, 1991). The suicide of a patient may put extreme strain on working relationships between staff members and affect both their ability to work together and with other patients. Psychiatrists must be vigilant for evidence of blame or splitting among colleagues and seek to remedy this if necessary.

Staff should be made aware of the agenda for dealing with the death and, for example, should be given the opportunity to attend the funeral (Reference Kaye and SoreffKaye & Soreff, 1991). In the more formal structure of a psychological autopsy (Reference Ruben, Blumenthal and KupferRuben, 1990), staff may examine their feelings and attitudes surrounding the suicide and give consideration to possible contributory factors. This process has two stages, namely the gathering of all pertinent information, usually by a clinician unfamiliar with the case, followed by the presentation of his/her findings. Such a process has been shown to diminish feelings of grief and guilt while avoiding ascribing blame or responsibility (Reference Ruben, Blumenthal and KupferRuben, 1990). It also provides an opportunity to reflect on how one might act differently in the future and may provide the stimulus for ongoing in-service training. This may address issues such as personal beliefs regarding suicide, feelings surrounding responsibility and its limitations, guidelines for the management of the suicidal patient and procedures in the event of patient suicide. One should bear in mind that some members of staff may feel threatened by this process and also that it is not a privileged forum in terms of the confidentiality of what is discussed.

In order to remain effective staff members, individuals must work through their own grief (Reference Dunne, Dunne, Mcintosh and Dunne-MaximDunne, 1987) and this may take time. Kaye and Soreff (Reference Kaye and Soreff1991) noted that in one case healing and closure for the staff did not occur until more than a year after the patient's death, with the presentation of the case at a Grand Round.

Other patients

Particularly in the in-patient setting, other patients must not be neglected in the aftermath of patient suicide. Kaye and Koreff (1991) advocate a rapid, organised meeting of patients and staff and suggest that this reinforces the sense of community on the ward. Patients will react in different ways, often depending on their own psychopathology and the nature of their relationships with the deceased. They need clear, accurate and timely information from a single authoritative source in order to prevent the emergence and proliferation of rumours (Reference Dunne, Dunne, Mcintosh and Dunne-MaximDunne, 1987). To varying extents they may experience a typical grief reaction and need the provision of appropriate opportunities for the expression of emotional responses. Suicidal patients may be especially vulnerable and develop concerns regarding the ability of staff to help them. Practical measures such as limiting passes may be appropriate in the short term.

Conclusion

Patient suicide is a tragic and not infrequent event. The psychiatrist has an important role in the aftermath of a patient suicide, both in communicating with the patient's family, other clinical staff and the authorities, as well as encouraging and participating in discussion about the patient's care. He/she must also monitor his/her own reactions to the suicide because failure to do so may have a significant impact on his/her mental health and his/her ability to function professionally. Clinical services should have written guidelines giving advice to staff on good practice following the suicide of a patient. The psychiatrist has an important contribution to make to the process of discussion, reflection and enquiry that must follow a patient suicide, so that lessons are learned and the best standard of clinical practice is maintained.