Development of best practice guidelines for suicide-related crisis response and aftercare in the emergency department or other acute settings: a Delphi expert consensus study

Suicide causes significant public health burden worldwide. In Australia, suicide is a leading cause of death in Australians aged 15–49 years, accounting for approximately 2800 deaths [1] and over 26,000 hospitalisations annually [2]. However this number significantly underestimates the proportion of people who present to the acute settings (hospital emergency departments and Psychiatric Emergency Care Centres) for suicide-related behaviour and who are discharged into the community or leave prior to receiving treatment. A history of suicide attempt is a robust predictor of repeat attempt and suicide death in Australia [3] and internationally [4, 5]. Risk of subsequent suicide death is highest within the first year of an index suicide attempt [6, 7] and up to six times higher among those who present to acute settings compared to the general population [8]. The heightened risk of suicide during the period immediately following discharge means that the acute setting is a critical point of intervention for those who are most at risk of subsequent suicide.

People who present to the acute settings for suicidal behaviour (including suicide ideation and suicide attempt) often have multiple psychological, social and interpersonal needs [9]. Apart from cognitive behavioural therapies, there is limited evidence for the effectiveness of psychosocial interventions on rates of suicidal behaviour following an episode of self-harm [10]. Although previous studies have shown people who receive some form of intervention, such as a psychosocial assessment [11‐14], admission to a hospital bed [7, 15], or referral to outpatient follow-up care [6, 7] are less likely to engage in repeat suicidal behaviour or subsequently die by suicide, factors such as negative staff attitudes and stigma can impact patient engagement in follow-up services, as well as future help seeking behaviour [16‐19]. Thus, even when patients receive assessment or referral to follow-up care, patients’ interactions and experience with healthcare staff can have a significant impact on their perception of care, and has the potential to mediate subsequent suicide risk.

Clinical Practice guidelines for the management of people who present to the acute settings for suicidal behaviour are available in Australia and New Zealand [20], Canada [21], the UK [22] and the US [23]. The purpose of these guidelines is to provide recommendations for the organisation of clinical services and management of medical and mental health. Nevertheless, these guidelines are subject to several limitations. For example, with the exception of the National Institute for Health and Care Excellence (NICE) in the UK [24, 25], few guidelines have been developed in consultation with consumers or people with lived experience of suicidal behaviour. The inclusion of consumers with lived experience is a quality indicator using the AGREE II (Appraisal of Guidelines Research and Evaluation II) instrument [26], is a key component of service planning, delivery and evaluation in the recent Fifth National Mental Health and Suicide Prevention Plan in Australia [27], and has been a policy directive in the National Health Services research and development process in the UK for the last two decades [28]. This approach, provides service users with lived experience the opportunity to identify existing gaps and identify services which may best meet their needs [28]. The inclusion of consumers with lived experience may be particularly relevant to the development of guidelines in the acute settings, given recent evidence that suggests those who have experienced a suicide attempt are often discharged from the acute settings without being offered the help they need [19].

In addition to the limited involvement of external stakeholders in the development of best practice recommendations for the management of suicidal behaviour in the acute settings, to date, existing guidelines have overwhelmingly focused on only a few aspects of the care continuum, from the point of presentation, triage, assessment, and discharge into the community. Consequently, current guidelines typically lack enough detail for implementation when faced with common organisational barriers in the acute settings such as high patient volumes, time pressures, availability of hospital beds, access to trained mental health consultants and presentation during afterhours [29‐31]. For example, Clinical Practice Guidelines from both Australia and New Zealand [20] and the NICE guidelines in the UK [22] recommend that all people at risk of suicide who present to acute settings should receive a psychosocial assessment by a mental health consultant. But these guidelines do not describe what should be done in the absence of a mental health team and provide limited information on the content that should be included during the psychosocial assessment. It is precisely in these contexts, for example when there is limited access to trained mental health consultants, in which guidelines for the management of suicide risk may be most beneficial [30].

In an effort to further inform best practice, we conducted a Delphi expert consensus study to develop guidelines for suicide-related crisis response and aftercare in acute settings. The study comprised two panels. The first panel consisted of consumers with lived experience of suicidal behaviour, and the second consisted of professionals from acute settings, crisis and aftercare, healthcare, and suicide prevention sector. The Delphi method recruits panels of experts who make private ratings on statements until group consensus is met [32, 33]. In its original form, the Delphi method has been used in healthcare research to reach consensus when evidence is lacking, or when there is conflicting information [32]. Whilst there is some evidence for effectiveness of interventions for suicidal patients [34, 35], the present study aims to develop guidelines that take into account multiple aspects of the care continuum (i.e. from the point of presentation, triage, psychosocial assessment, discharge planning, and referral into the community), which as noted previously, are not routinely included in existing guidelines. For this reason, we chose the Delphi method in order to develop a comprehensive set of guidelines that take into account recommendations from the existing guideline literature and academic literature, while also assigning equal weight to the expert opinions of consumers and professionals. This method has been used previously to identify which approaches may be most beneficial, in the absence of empirical or conflicting evidence [33]; and has been broadly applied in the development of best practice guidelines [36‐39].

This study aimed to identify best practice strategies for crisis response and follow-up care in the acute settings for people who present for suicidal behaviour. The final guidelines consist of 420 items that outline the team structure; the provision of peer support; procedures and actions staff can take from initial contact, assessment, referral, discharge and follow-up; staff training; and linkage with community aftercare services. These items have been written into a set of guidelines that provide guidance to acute settings with respect to policy and practice [41]. In these guidelines, items were broadly grouped into areas of responsibility to ensure that the guidelines were targeted to the needs of different audiences. When responses between panels were compared, a higher proportion of consumers were more likely to endorse statements as essential or important when they involved non-clinical factors such as positive lifestyle skills, parenting skills, domestic support, harm minimisation plans, and the inclusion of peer support. In contrast, professionals were more likely to endorse items specific to assessing the lethality of the suicide attempt; the person’s involvement with the criminal or youth justice system; and items which included contacting the person’s carer, family, or friends as part of the psychosocial assessment process. This focus is more aligned with emergency medicine models that favour the identification and mitigation of short-term morbidity and mortality. Lower endorsement rates for some strategies, such as direct referral to aftercare services and the assessment of non-clinical factors may also reflect the difficulties health professionals experience in identifying appropriate aftercare services that meet the diverse needs of people who present to the acute settings for suicidal behaviour [9]. Nevertheless, inclusion of assessment of broader psychosocial factors in the current guidelines represents an important step in recognising and responding to the sometimes complex needs of individuals at risk of suicide who present to acute settings.

Participants’ feedback provided important insights into the rationale behind both consumers’ and professionals’ choices. When consumer feedback was incorporated into the second questionnaire, both panels were equally likely to endorse items as important or essential if they included the person’s consent. Similarly, the provision of brief discharge planning was included in the guidelines only when feedback involving contingent arrangements for more extensive aftercare planning in the community was incorporated into the second questionnaire. Feedback from the professionals showed that items that were judged as requiring significant allocation of financial resources were endorsed at a much lower rate, even if they were considered best practice. This may account for lower endorsement rates among the professional panel for statements involving the allocation of staff to accommodate suicide-related presentations during after-hours as well as higher endorsement rates among professionals for items that included the use of teleconferencing to fill provider gaps.

In contrast, feedback from the consumer panel emphasised the importance of empathy and compassion as well as concerns about stigma. The therapeutic value of empathy is a key theme in several qualitative studies [16, 17, 19, 42, 43] systematic reviews [18, 44] and clinical practice guidelines [20, 22]. For example, Taylor et al. [18] conducted a systematic review of qualitative studies investigating the healthcare experiences of people who have made a suicide attempt. In this review, the majority of studies that provided recommendations on service improvement identified a need for improved staff knowledge towards people who have self-harmed, as well as increased sympathy characterised by listening to patients and responding to them in a non-judgmental way. In a recent study of Australians who had experienced a suicide attempt, McKay et al. [42] found empathy was central to patients’ sense of support and care even among those who experienced ongoing suicidal ideation and vulnerability. In this study, empathy was characterised by the authors as ‘small kindnesses’ such as providing patients with privacy and a quiet place to rest. Whilst it is difficult to mandate empathy through clinical practice guidelines, results from the current Delphi study, and previous studies combined, suggest the attitudes of staff ought to be a priority for the response to suicidal behaviour in acute settings.

To the authors’ best knowledge, this is the first study that has used the Delphi method to develop best practice guidelines for the management of suicidal behaviour in acute settings. Particular strengths of the approach outlined here include the equal weight given to the views of people with lived experience. This is particularly relevant given consumer opinion is considered a key priority in healthcare policy directives in Australia and internationally, but is often overlooked in the delivery of health care responses. Further, the current guidelines address important gaps in the current guideline literature. For example they provide information on how to respond to people who present for suicidal behaviour in the absence of trained mental health staff and provide specific details on what constitutes a safe environment, as well as important considerations during assessment, referral and discharge.

These guidelines provide recommendations and strategies for responding to people who present for suicidal behaviour in acute settings based on the consensus of two expert panels comprising professional involved in the care of people who have experienced suicidal risk and people with lived experience. They provide recommendations that address some of the organisational barriers encountered by staff when working with suicidal persons, and give equal weight to the perspective of end users, whose experience with suicidal behaviour is a valuable source of expertise for informing evidence based practice. As with any guidelines, implementation will vary according to context and it is our hope that this research will be useful to the scientific and clinical community outside Australia. Future research should investigate ways to optimise implementation of these guidelines in order to improve equal access to quality care for people at risk of suicide who present to the acute settings.