Developing best practice guidelines for the psychosocial assessment of Aboriginal and Torres Strait Islander people presenting to hospital with self-harm and suicidal thoughts.

OBJECTIVE: To develop guidelines for the culturally responsive psychosocial assessment of Aboriginal and Torres Strait Islander people presenting to hospital with self-harm and suicidal thoughts.
METHOD: The Delphi method was used to establish expert consensus. A systematic search and review of relevant research literature, existing guidelines and grey literature was undertaken to develop a 286-item questionnaire. The questionnaire contained best practice statements to guide clinicians undertaking psychosocial assessment of Aboriginal and Torres Strait Islander people presenting to hospital with self-harm and suicidal thoughts. An expert panel comprising 28 individuals with clinical, community-based and lived experience in Aboriginal and Torres Strait Islander mental health and/or suicide prevention were recruited and independently rated the items over three rounds. Statements endorsed as essential or important by 90% or more of the expert panel were then synthesised into recommendations for the best practice guideline document.
RESULTS: A total of 226 statements across all relevant areas of clinical practice were endorsed. No statements covering the use of structured assessment tools were endorsed. The endorsed statements informed the development of a set of underlying principles of culturally competent practice and recommendations for processes of effective and appropriate engagement; risks, needs and strengths to be assessed; formulation of psychosocial assessment; and recommendations specific to children and young people.
CONCLUSION: The guidelines are based on recommendations endorsed across a range of expertise to address an important gap in the evidence-base for clinically effective and culturally responsive assessment of self-harm and suicidal thoughts by Aboriginal and Torres Strait Islander people in hospital settings. Further work is needed to develop an implementation strategy and evaluate the recommendations in practice.

Introduction

Previous self-harm, where a person intentionally injures or poisons themselves irrespective of suicidality, is one of the strongest predictors of death by suicide (Carroll et al., 2014; Ribeiro et al., 2016). It is estimated to account for a 100-fold increase in risk compared to the general population who have not experienced self-harm (Owens et al., 2002). This association has been confirmed for Aboriginal and Torres Strait Islander peoples in Australia in retrospective cross-sectional studies of suicides (De Leo et al., 2011; Kuipers et al., 2012; Parker and Ben-Tovim, 2002). Therefore, the high prevalence of non-fatal suicide attempts among Aboriginal and Torres Strait Islander people in the general population in Australia (Armstrong et al., 2017) and the persistently elevated and increasing rates of hospital presentations for self-harm among Aboriginal and Torres Strait Islander peoples in recent years (Harrison and Henley, 2014; Leckning et al., 2016) are cause for alarm.

Although most self-harm does not result in hospitalisation (Shand et al., 2018), the evidence suggests that presentation to hospital represents an important opportunity for engaging a high-risk population group in preventive intervention. One aspect of clinical care that is considered essential for hospital-treated self-harm is psychosocial assessment (National Collaborating Centre for Mental Health [NCCMH], 2011a, 2011b), which is used by clinicians to gather information about and evaluate the risks, needs and strengths of an individual to inform the most effective and appropriate care and treatment for underlying mental health and related issues. There is mixed evidence, however, about the role of these assessments in reducing suicidality and associated risks (Carroll et al., 2016). Accordingly, the focus of clinical assessments is increasingly expanding towards prevention rather than risk prediction and stratification alone (Pisani et al., 2016). At the same time, clinical guidance is important in order to minimise any negative experiences of assessment by patients treated for self-harm in hospitals, which have been shown to have adverse impacts on subsequent help-seeking behaviour (Hunter et al., 2013; Taylor et al., 2009).

While specific clinical guidance exists for assessing and managing self-harm in hospitals, it is unknown the extent to which these recommendations address the uniquely complex interplay of social, political and historical forces as well as the individual circumstances regarding Aboriginal and Torres Strait Islander suicidal behaviour (Dudgeon et al., 2017; Hunter and Harvey, 2002). The importance of such a gap cannot be underestimated given hospitals are the most common service encountered prior to suicide by Aboriginal and Torres Strait Islander people (Sveticic et al., 2012). Moreover, these health inequities are compounded by ongoing experiences of discrimination and racism across the healthcare system (Laverty et al., 2017). As a result, clinical mental health services are often experienced by Aboriginal and Torres Strait Islander people as unhelpful and inappropriate (Dudgeon et al., 2014). Addressing the gaps in clinical practice is urgently needed to improve the quality of care received by, and therefore, the outcomes achieved for Aboriginal and Torres Strait Islander people presenting to hospital with self-harm and suicidal thoughts.

Over the past two decades, cultural competence has informed numerous reforms and innovations to clinical health care designed to address these gaps and improve outcomes for Aboriginal and Torres Strait Islander people (Calma et al., 2017). That is, by helping health care practitioners to identify strategies for working safely and effectively across cultural differences, it is expected that Aboriginal and Torres Strait Islander people will receive better-quality care that should help to improve outcomes. Despite these efforts, interventions designed to improve cultural competence have had mixed results (Bhui et al., 2007; Clifford et al., 2015) with many clinicians feeling ill-equipped to put cultural competence into practice (McGough et al., 2018).

To address the gaps in existing clinical guidance and better support clinicians to be more culturally competent, a Delphi study was designed to develop best practice guidelines for the psychosocial assessment of Aboriginal and Torres Strait Islander people presenting to hospital with self-harm and suicidal thoughts. The Delphi method is an effective and reliable approach that has been used to systematically develop clinical guidelines in the absence of a comprehensive evidence-base and where the vulnerable state of patients warrants a less invasive approach (Jorm, 2015). Moreover, this approach has been endorsed by Aboriginal and Torres Strait Islander participants in previous mental health research (Chalmers et al., 2014).

Methods

The Delphi method used in this study involved presenting to a panel of experts a series of statements regarding best practice for the psychosocial assessment of Aboriginal and Torres Strait Islander people presenting to hospital with self-harm and suicidal thoughts. The statements were derived from a review of literature from a systematic search and presented to the expert panel in a questionnaire format. Statements were rated by the expert panel over three sequential rounds, with subsequent rounds being used to evaluate additional statements proposed by the expert panel in the first round as well as re-evaluate statements considered close to being endorsed by the expert panel in previous rounds. Feedback is given at each stage in order to help experts assess their opinions against those of the group. After the three rounds, endorsed statements were developed into recommendations for inclusion in the final guideline document. The research team (B.L., A.R., M.W.) were responsible for conducting the Delphi study with oversight and leadership from the working group (TH, TAC, GR, GA), comprising two clinical psychologist researchers (one identifying as Aboriginal), a social researcher and a public health researcher all with considerable experience in suicide prevention and/or Aboriginal and Torres Strait Islander health research. Experienced Aboriginal and Torres Strait Islander mental health practitioners and researchers were also recruited as advisers to provide feedback on the conduct of the research and its outputs at critical points outlined below. This study was approved by the NT Department of Health and Menzies School of Health Research Human Research Ethics Committee (Ref.: 2018-3200).

Literature search

Systematic literature search strategies (see Supplementary Table 1 for keyword groups and terms used) were designed to identify existing guidelines from Australia and other countries with similar postcolonial histories (Canada, New Zealand and the USA) (see Supplementary Figure 1), relevant Aboriginal and Torres Strait Islander gatekeeper interventions (Clifford et al., 2013; Ridani et al., 2015), evidence-based practice from the research literature (see Supplementary Figure 2) and any other relevant recommendations from grey literature discussing Aboriginal and Torres Strait Islander suicide prevention (see Supplementary Figure 3). Two evidence-based, five international and eight Australian guidelines were identified and reviewed for this study (see Supplementary Table 2). Of the three discrete gatekeeper programmes identified (Suicide Story, INSIST and Aboriginal Mental Health First Aid [MHFA]), only one programme had publicly available guidelines (Aboriginal MHFA) and another had an information DVD (Suicide Story) that were used in the study (see Supplementary Table 3). The search for identifying explicitly relevant academic journal articles yielded four results, with only one article from a New Zealand study (see Supplementary Table 4) containing a substantive enough discussion of assessment for inclusion. Given these limited results, the criteria for the search strategy were relaxed to include academic research discussing mental health assessment of Aboriginal and Torres Strait Islander peoples in any clinical setting (not just hospitals) and for any clinical presentation (not just self-harm and/or suicidal thoughts) (see Supplementary Figure 3). This amended search strategy identified 27 academic journal articles considered relevant to the study (see Supplementary Table 5). The grey literature search, which was conducted last to ensure the exclusion of search results already included from other searches, identified five additional resources that were included in the study (see Supplementary Table 6). In sum, 15 guidelines, 28 academic journal articles, three resources from two gatekeeper interventions, three research reports and one suicide prevention website were included in the literature review to develop the questionnaire for the expert panel.

Panel formation

A panel recruitment strategy was developed to engage a wide range of relevant perspectives, including lived experience, and priority was given to Aboriginal and Torres Strait Islander experts. Advertisements to nominate for the expert panel were broadcast publicly through relevant organisations (see Supplementary Table 7). Targeted recruitment was also undertaken based on recommendations provided by the organisations that were approached as well as Aboriginal and Torres Strait Islander advisers to the study. Nominations were received using an online form and nominees were excluded if they did not meet the selection criteria: five or more years of relevant clinical experience for non-Indigenous candidates and three or more years of relevant professional experience and/or lived experience for Aboriginal and Torres Strait Islander candidates. Nominees were informed during recruitment that a gratuity of AUD$200 would be offered after completing all three rounds of questionnaires as compensation for their time. All 32 nominees met the selection criteria and were invited to participate in the study as an expert panellist. Twenty-eight nominees accepted the invitation by consenting to participate in the study, with 11 identifying as Aboriginal, one as Torres Strait Islander and 16 as non-Indigenous.

Questionnaire development and administration

All literature identified in the search was then reviewed by two of the authors (B.L., A.R.) for statements containing recommendations for, or relevant to, the psychosocial assessment of Aboriginal and Torres Strait islander people presenting to hospital with self-harm and suicidal thoughts. These statements were reworded as recommendations, if needed, and presented for final review to the working group. Two hundred and eighty-six statements were included in the first round that were organised thematically into six sections reflecting principles and practices of different areas of clinical mental health assessment. The questionnaire asked participants to rate whether each statement should be included in the guidelines using a 5-point Likert-type scale: ‘essential’, ‘important’, ‘don’t know/depends’, ‘unimportant’ or ‘should not be included’. A free text response field was provided at the end of each section to allow for expert panellists to suggest new statements or revisions to existing ones for inclusion in the second-round questionnaire. Prior to entering the first-round questionnaire, expert panellists were presented with information about the study and asked to consent to participation. The questionnaire was constructed and administered online using Qualtrics® (www.qualtrics.com/).

Statistical analysis

It was determined a priori that a percent agreement calculation would be used to determine cut-offs for different levels of endorsement. Given the size and diversity of the expert panel, a higher than conventional level of endorsement was used for this study (Diamond et al., 2014). A statement was endorsed if it was rated as ‘essential’ or ‘important’ by at least 90% of the expert panel (80% agreement is more common). A statement was included for re-rating in subsequent rounds if it was rated as ‘essential’ or ‘important’ by 80–89% of the expert panel. This only applied to all statements in the first-round questionnaire and new statements presented in the second-round questionnaire. Any statements that were not endorsed after three rounds were excluded. At the end of each round, levels of endorsement were calculated for each statement and reported back to the expert panel for their review. At the end of the survey, the reliability of the consensus achieved was evaluated using Brennan-Prediger’s agreement coefficient (AC) (B-P’s ) and Gwet’s AC (Gwet’s ) (Klein, 2018). Agreement between Aboriginal and Torres Strait Islander and non-Indigenous panellists was determined by estimating the Pearson product–moment correlation (ρ) of average percentage agreement for each item by Indigenous status. All statistical analyses were conducted in Stata 15 (StataCorp, 2017).

Guideline development

After the statistical reliability of the results was established, a review of the final list of endorsed statements and how they were to be organised was undertaken by two of the authors (B.L., A.R.) to ensure the resulting guideline document would be clear and coherent. This began with a thematic review of the endorsed statements within the sections and sub-sections in which they were presented to the expert panel to suggest any structural changes that would improve the logic of the overall guideline document. A summary of these structural changes is presented in Supplementary Table 9. A detailed review of individual statements was undertaken to identify any superficial wording and grammatical changes required prior to inclusion in the final guideline document as recommendations. This review included identifying any overlapping statements that needed to be addressed. All changes recommended were reviewed and approved by the working group. The final guideline document was then drafted in accordance with recommendations outlined in the AGREE II statement (Brouwers et al., 2010), reviewed and approved by the working group, and sent to the expert panel and the study’s Aboriginal and Torres Strait Islander advisers with a detailed report of changes made to statements and how they had been organised. No objections were received from any panellists. Some amendments to wording in the guidelines, additional examples and revisions to the introductory sections were suggested by the Aboriginal and Torres Strait Islander advisers to the project. All expert panellists were given the opportunity to review and approve the final version of the guideline document prior to its release online (Leckning et al., 2019).

Results

Of the 28 participants who completed the Round 1 questionnaire, 25 (93%) returned to complete questionnaires in Rounds 2 and 3. Differences in the composition of the expert panel observed across these rounds were reviewed and considered to be very unlikely to bias results (see Supplementary Table 8). Participants reflected a range of expertise and experience from across Australia (see Table 1), ensuring the results are as representative as possible of a range of relevant perspectives.

Table 1.

Characteristics of expert panel participants by Indigenous status.

	Aboriginal and Torres Strait Islander		Non-indigenous		Total
	No.	%	No.	%	No.	%
Socio-demographic characteristics
Gender
Female	7	58.3	7	43.8	14	50.0
Male	5	41.7	9	56.2	14	50.0
Age group (years)
25–35	1	8.3	2	12.5	3	10.7
36–45	3	25.0	3	18.8	6	21.4
46–55	5	41.7	5	31.2	10	35.7
56–65	3	25.0	4	25.0	7	25.0
Over 65	0	0.0	2	12.5	2	7.1
Jurisdiction
NSW	0	0.0	2	12.5	2	7.1
NT	4	33.3	6	37.5	10	35.7
QLD	3	25.0	5	31.2	8	28.8
SA	2	16.7	0	0.0	2	7.1
VIC	1	8.3	1	6.3	2	7.1
WA	2	16.7	2	12.5	4	14.3
Characteristics of relevant experience
Roles[a]
Aboriginal mental health practitioner	3	25.0	0	0.0	3	10.7
Psychologist	3	25.0	2	12.5	5	17.9
Psychiatrist	0	0.0	2	12.5	2	7.1
Other medical doctor	0	0.0	4	25.0	4	14.3
Suicide prevention practitioner	4	33.3	1	6.2	5	17.9
Nurse	1	8.3	7	43.8	8	28.6
Researcher	3	25.0	4	25.0	7	25.0
Other	1	8.3	1	6.25	2	7.1
Settings[a]
Hospital	6	50.0	11	68.8	17	60.7
Community mental health service	10	83.3	7	43.8	17	60.7
Primary health care/Aboriginal medical service	4	33.3	8	50.0	12	42.9
Community-based suicide prevention	5	41.7	2	12.5	7	25.0
Other	1	8.3	2	12.5	3	10.7
Experience (years)
3–9	2	16.7	5	31.2	7	25.0
10–19	6	50.0	6	37.5	12	42.9
20+	4	33.3	5	31.3	9	32.1

Participants were permitted to specify more than one category as applicable to their experience. Therefore, the total of column percentages may exceed 100%.

Most of the statements from the first-round questionnaire were endorsed (n = 187; 65.4%). In addition to the 45 statements requiring re-rating, 22 new statements were developed for Round 2 using suggestions provided by the expert panel in Round 1. After completion of three rounds, a total of 226 statements were endorsed for inclusion in the guideline document. Figure 1 provides an overview of the results across all three rounds and Table 2 summarises the results according to the number of statements endorsed within each section of the questionnaire. As can be seen, most statements in each section were endorsed. The exception to this was a section on effective and appropriate engagement where only 42% (n = 29/67) of statements were endorsed.

Figure 1.

Overview of statements created, re-rated, included and excluded in each round of the survey.

Please note that an item from Round 1 that was meant to be re-rated in Round 2 was administered in Round 3 instead.

Table 2.

Number of statements endorsed and rejected in each section of the questionnaire.

Section title	No. endorsed	No. rejected
1. Principles of cultural competency	36	4
2. Effective and appropriate engagement	28	39
3. Elements of comprehensive assessment	68	11
4. Post-assessment process	23	2
5. Recommendations for young people	53	5
6. What to avoid	6	12
New statements	14	8
Total	228	81

It is also worth noting that within this section, no statements were endorsed from a sub-section relating to the use of structured assessment tools. Results for each round are presented in Supplementary Table 7 for all statements presented to the expert panel, organised by section and sub-section.

Analysis undertaken using the results from Round 1 shows strong agreement in the responses provided by the expert panel. The overall level of agreement was very high at 91.2%, with slightly higher levels of agreement among Aboriginal and Torres Strait Islander (93.0%) compared to non-Indigenous (91.1%) panellists. The statistical measures of interrater reliability used for this study indicate substantial agreement overall (B-P’s = 0.71, p < 0.001; Gwet’s = 0.84, p < 0.001). Moreover, the Pearson’s product–moment correlation coefficient (ρ = 0.73, p < 0.001) indicates a strong positive relationship between the responses provided by Aboriginal and Torres Strait Islander and non-Indigenous expert panellists. The results of this analysis confirm the very high and reliable level of consensus achieved across the diversity of perspectives represented by the expert panel.

Discussion

The aim of this study was to develop guidelines for implementing a best practice approach to the psychosocial assessment of Aboriginal and Torres Strait Islander people who present to hospital with self-harm and/or suicidal thoughts. This was motivated by the gaps in existing clinical guidelines and the ongoing adverse experiences of clinical mental health care by Aboriginal and Torres Strait Islander people. Despite being from diverse backgrounds and geographical locations across Australia, the expert panel was able to reach a high level of consensus on best practice statements, and 226 statements endorsed by ⩾90% of panellists were included in the guidelines document.

The recommendations cover all aspects of clinical practice relevant to psychosocial assessment in the context of evaluating self-harm and suicidal thoughts by Aboriginal and Torres Strait Islander people. The guidelines start with recommended principles of culturally competent practice relevant to making psychosocial assessment more responsive to the cultural needs and strengths of patients. Several strategies relating to interpersonal and communication skills are covered to ensure more effective and appropriate engagement of Aboriginal and Torres Strait Islander people during the assessment process. Informed by the concept of social and emotional wellbeing, the guidelines also recommend a comprehensive set of risks, strengths and needs to be considered and set out how this informs more culturally appropriate clinical formulation and further care for Aboriginal and Torres Strait Islander people. A section of the guidelines is dedicated to specific recommendations for assessing children and adolescents who are recognised as a high-risk group with specific developmental and socio-cultural needs. A few recommendations for what clinicians should avoid are also included to ensure that minimum requirements for both patient and cultural safety are satisfied in the assessment. The resulting guidelines, the first of their kind for Aboriginal and Torres Strait Islander people, ultimately aim to establish a standard of practice that could improve the quality of care and outcomes for patients.

Importantly, the approach used in this study has led to the development of guidelines that have maximised the existing evidence identifying best practice. The lack of current guidelines and absence of an explicitly relevant evidence-base mandated developing a strategy to identify recommendations from other sources that could be applicable to Aboriginal and Torres Strait Islander people presenting to hospital with self-harm and suicidal thoughts. Central to this was the emphasis placed on developing the cultural competence of clinicians and their sensitivity to the social and emotional wellbeing of Aboriginal and Torres Strait Islander people. The resulting guidelines provide comprehensive support for more culturally competent practice by not only informing clinicians of important historical and socio-cultural influences that may adversely influence their interactions with Aboriginal and Torres Strait Islander people, but by offering practical communicative and other interpersonal strategies that address known barriers to culturally competent practice (McGough et al., 2018) and better ensure culturally respectful and safe engagement during assessment.

However, there may be challenges for implementing the guidelines that are reflected in areas of clinical practice that did not receive relatively high levels of endorsement by the expert panel. Less than half of the statements pertaining to effective and appropriate engagement of Aboriginal and Torres Strait Islander people were endorsed. In some cases, this reflects responses from the expert panel consistent with the mutually exclusive nature of some statements, such as those recommending ways to start an assessment. However, in other sets of statements, these low levels of endorsement may be more meaningful. For example, only one statement on the role of ‘yarning’ in clinical assessment was endorsed. ‘Yarning’ has been described as a more conversational and indirect style of discussing mental health issues that is more familiar and appropriate to Aboriginal and Torres Strait Islander people (Nagel et al., 2012). At the same time, despite the increasing availability of culturally appropriate measures (Dingwall and Cairney, 2010; Le Grande et al., 2017), the expert panel did not reach a consensus on how these tools should be used with Aboriginal and Torres Strait Islander people. This is reflected in comments from the expert panel concerned about how ‘culturally useful’ or ‘culturally valid’ existing risk assessment tools are and whether they even stratify or predict risk accurately among Aboriginal and Torres Strait Islander people. In the end, the statements endorsed by the expert panel do not recommend a specific approach to assessment whether through yarning or use of structured tools, since both could form part of one process. Given these concerns and the importance of respectfully recognising the diversity of Aboriginal and Torres Strait Islander peoples, the recommendations in the guidelines emphasise the need to exercise discretion in determining the appropriateness of informal styles of interaction and/or formal assessment tools with Aboriginal and Torres Strait Islander people (Ralph et al., 2018). This also highlights the need for further local work to complement these general guidelines with contextually relevant additions or adjustments.

Other statements lacking strong endorsement were those relating to who should be involved in the assessment process. Despite growing evidence suggesting the involvement of Aboriginal Mental Health Workers/Liaison Officers improves the quality of care provided (McKenna et al., 2015; Nagel and Thompson, 2006), the expert panel did not arrive at a consensus on a prescribed or standard set of people to be involved in assessments. Rather, the consensus within the same section was for patients to be central to clinical decision-making, which includes asking them who should be involved. This reflects current trends in person-centred collaborative approaches to mental health care that are widely encouraged (Salvador-Carulla and Mezzich, 2012). It may also reflect, as one participant commented, on the differences in the need for, and availability of, Aboriginal and Torres Strait Islander clinical and cultural advisers across Australian hospitals. This points to the need to develop the workforce of Aboriginal and Torres Strait Islander mental health practitioners and address other barriers and inequities to accessing culturally appropriate care (Hinton et al., 2015). Ultimately, further research is needed to better understand what models of hospital mental health service are needed that can best facilitate access to appropriate and locally relevant Aboriginal and Torres Strait Islander–specific resources.

Although the Delphi method used in this study has achieved its goal of developing an expert consensus across areas of clinical practice most relevant to suicidal behaviours in hospital settings, there are limitations worth noting that may have implications for how the guidelines are to be implemented and continue to be developed. To some extent this study is both justified and constrained by the limited evidence base for practices of assessment of Aboriginal and Torres Strait Islander patients in existing guidelines and research. Most guidelines used were either developed through the Delphi method themselves or did not specify a methodology. All but two studies involving Aboriginal and Torres Strait Islander people reviewed in the research literature were observational in design and/or focused on process rather than patient outcomes. Future revisions to the guidelines would, therefore, benefit from greater input from lived experience or making use of other qualitative research to develop additional items to be rated by mental health professionals. Furthermore, given the diversity of experts in this study, the Delphi process may not have provided sufficient opportunities for participants to adequately identify biases and clarify assumptions that underpinned their opinions. Future revisions to the guidelines may benefit from a modified Delphi process that starts with a focus group and/or finishes with a workshop in which such issues can be addressed.

The guidelines were developed to reflect and respect the diversity of Aboriginal and Torres Strait Islander peoples and their experiences. However, it is questionable whether any guidelines can fully reflect the needs of all groups such as the elderly, disabled and gender- and sexuality-diverse Aboriginal and Torres Strait Islander people along with other demographic and socio-cultural differences. It is expected that such limitations can be addressed within the design of training interventions coupled with an appropriate implementation strategy intended to optimise responsiveness to specific areas of need. These should be evaluated both to determine the general effectiveness of the guidelines and, as far as possible, to test their sensitivity to important and often neglected Aboriginal and Torres Strait Islander population groups.

Conclusion

This study has informed the development of guidelines for the best practice psychosocial assessment of Aboriginal and Torres Strait Islander people presenting to hospital with self-harm and suicidal thoughts, which addresses gaps in the evidence-base for clinically effective and culturally appropriate care. It draws on the widest available evidence-base of clinically effective practice and is informed by the key concepts of cultural competence and social and emotional wellbeing that are central to increasing the cultural responsiveness of mental health care. The synthesis of these sources of recommendations was systematically developed through the Delphi method, which made use of a diverse and experienced panel of experts to arrive at a consensus on best practice that covers all areas of clinical practice relevant to psychosocial assessment. The next step requires developing an implementation strategy and evaluation framework to further advance the evidence-base and ensure these guidelines help to overcome inequities in health care and the consequent outcomes for Aboriginal and Torres Strait Islander people experiencing self-harm and suicidal thoughts.

Click here for additional data file.

Supplemental material, BestPrAxIS_study for Developing best practice guidelines for the psychosocial assessment of Aboriginal and Torres Strait Islander people presenting to hospital with self-harm and suicidal thoughts by Bernard Leckning, Tanja Hirvonen, Gregory Armstrong, Timothy A Carey, Mark Westby, Alix Ringbauer and Gary Robinson in Australian & New Zealand Journal of Psychiatry