The Scottish Government

« Previous | Contents | Next »

Listen

SUICIDE AND SUICIDAL BEHAVIOUR: ESTABLISHING THE TERRITORY FOR A SERIES OF RESEARCH REVIEWS

CHAPTER THREE: IDENTIFYING AREAS OF INTEREST AND EXPERTISE: INTERNATIONAL SURVEY

RESPONSE

3.1 In total 229 researchers specialising in suicide and suicide behaviour were identified and invited to participate in the research. Of these, 20 were failed contacts, that is, the individual had left their post or the contact details available to the research team were incorrect. Therefore, the number of valid contacts and potential survey participants was 209.

3.2 Sixty two completed questionnaires from across the world were received, of which 13 were submitted 'manually', the rest using the electronic questionnaire system. Participants included 35 medical doctors or PhDs, 13 professors and 6 associate professors; the remaining participants did not have an academic title. The response rate of 30 percent was reasonable, allowing for the fact that the survey was 'cold calling'.

3.3 Twenty responses were received from experts based in the UK; they formed the largest group of participants in the study (32% of all responses). Of all UK research specialists invited to participate, 26% did so, and of all Scottish researchers invited to participate 36% did so. A higher response rate amongst UK and Scottish research specialists might have been expected, given that the survey was explicitly intended to inform Scottish suicide prevention policy, although it is encouraging that they formed the largest group of respondents.

3.4 Of the 20 UK research specialists participating in the survey, 5 were based in Scotland. Details of the institutions where the UK participants worked can be found in the table 3.1.

3.5 In addition to UK responses, 42 responses were received from international experts, from 13 different countries. Table 3.2 gives details about the number of responses by country and the institutions in which the experts were based.

3.6 Table 3.2 shows that, although the majority of participants were located in (post-) industrialised countries, the international research specialists who took part the study spanned a number of diverse cultures. North American research specialists formed the largest group of international participants with 17 responses, followed by Europe (16), Australasia (4), South East Asia (3), and Asia (one).

3.7 Table 3.3 presents information about the countries where the 147 non-respondents were located.

SUMMARY OF RESULTS

3.8 The 62 research specialists who participated in the survey were asked questions about their research interests and views on collating and co-ordinating the evidence base relating to suicide and suicidal behaviour. The following section gives details of:

• How they thought a series of research reviews on suicide and suicidal behaviour should be divided up

• Their specialist interest areas

• The issues which should be focused upon in the reviews

• What they perceived to be the gaps in the evidence base

How to divide up the reviews

3.9 Participants were asked what they would consider to be the most constructive and useful approach to dividing up the proposed reviews of previous research on suicide and suicidal behaviour to ensure comprehensive coverage of all relevant issues while avoiding duplication. Table 3.4 below provides a summary of the responses to this question.

3.10 Participants suggested splitting the reviews in a number of different ways. Deliberate self-harm and suicidal ideation were included, as well as suicide. Examining suicide and suicidal behaviour at different stages of the life span was considered to be of key importance. Several participants also noted the importance of dividing the reviews between interventions and determinants, taking into account the importance of preventative interventions and high risk groups, such as those with mental health problems. There was interest in taking a multi-disciplinary approach to the reviews and working towards a biopsychosocial model that would combine often conflicting perspectives (e.g. whether suicide is biologically or socially determined).

Interests, issues and gaps

3.11 Survey participants were asked:

• What are your specialist research interests relating to suicide and suicidal behaviour?

• What do you consider to be the main issue upon which the reviews should focus?

• What do you perceive to be the main gaps in the evidence base relevant to the implementation of the Scottish Executive's suicide prevention strategy?

For each of these questions, participants were invited to record up to 5 responses, although on many occasions the complexity of answers meant that on coding more than 5 interests, issues or gaps were actually mentioned. Responses were coded using the list of themes in annex 2 ( for full details of the coding method, see section 2.2.2).

3.12 The coded data were categorised under the following key themes:

• Epidemiology

• Determinants

• Risk factors and conditions

• Protective factors

• Interventions

• Cross-cutting (e.g. availability of means, media, stigma, contagion, attitudes)

Table 3.5 below presents a broad overview of the categories of responses to the questions on interests, gaps and issues according to the key coding themes.

3.13 Cells in bold in table 3.5 mark the largest column percentage. Survey participants viewed interventions as the priority area for the reviews and as an area where there is the largest gap in the existing evidence base.

3.14 Tables 3.6, 3.7 and 3.8 give a more detailed picture of the responses to the interests, issues and gaps questions indicating:

• The most common interests noted by participants

• The number of participants with each of these interests, the number of UK participants with each of these interests and the number of international participants with each of these interests

• The most common interests as a proportion of all interests recorded, the most common interests as a proportion of all UK interests and the most common interests as a proportion of all international interests

Specialist interest areas

3.15 Participants were asked to record up to 5 of their specialist interest areas relating to suicide and suicidal behaviour. A total of 282 interests was given, 94 of which were from UK participants and 188 from international participants. The interests of the participants covered 61 different themes; a full list can be found in annex 4. Table 3.6 below gives a breakdown of the most common specialist interest areas for both UK and international participants.

3.16 In respect of table 3.6, a common interest was defined as an interest held by 7 or more participants in the whole sample. Cell entries in bold represent the most common response in the column.

3.17 No area of interest represented more than 6% of the total set of interest themes (table 3.6), indicating a broad range of specialist interests amongst the participants to the survey and hence the perceived complexity of this research field. Prevention and young people were the most common specialist interests among participants, each representing 6% of interests overall.

3.18 Deliberate self-harm related interests (deliberate self-harm, interventions for deliberate self-harm, and epidemiology of deliberate self-harm) are all more common interests among UK participants than non-UK participants. For UK participants, deliberate self-harm was the most common interest (11%), with epidemiology and interventions for deliberate self-harm also highlighted (7% and 6%, respectively).

3.19 For international participants, prevention and interventions (miscellaneous) were the most common specialist interests (6%).

3.20 Scottish based participants shared the interests represented in table 3.6 apart from socio-economic determinants, media, stigma, contagion and attitudes and cross cultural epidemiology. The Scottish participants differed from other participants in that they had specialist interests in national and local suicide prevention strategies, rural and urban differences, community based interventions and integrated models for suicide interventions. This perhaps reflects some of the more immediate suicide prevention agenda in Scotland.

Main issues upon which the reviews should focus

3.21 Participants were asked to record up to 5 main issues on which they thought the reviews should focus. Many respondents actually mentioned more than 5 issues. A total of 346 issues was identified, with UK participants contributing 115 issues and international participants, 231. These issues could be categorised into 79 different themes; a full list can be found in annex 5. Table 3.7 below gives a summary of the issues that were most commonly mentioned.

3.22 In respect of table 3.7, a common issue was defined as an issue mentioned 7 times or more by participants in the whole sample. The numbers in cells in bold represent the issues mentioned most often.

3.23 Table 3.7 shows that prevention and interventions were the most common issues overall (7% and 11%, respectively). Epidemiology, risk factors, particularly societal risk factors, and mental illness were also key issues. The key issues for UK participants were: interventions, deliberate self-harm (both 6%), prevention, epidemiology and young people (all 5%) and alcohol and drugs (4%).

3.24 International participants' identification of issues was more aligned to the sample as a whole. Neurobiological determinants were identified as key issues by international participants but not at all by UK participants. Scottish based participants mentioned all of the issues in table 3.7. The only issue raised solely by a Scottish based participant was recovery.

3.25 A comparison of specialist interests (table 3.6) and identified issues (table 3.7) shows some overlap but also some difference at the individual level. Prevention was the most common interest amongst the participants and the top issue for the reviews. For international participants prevention was the most common interest and issue. However, out of the 17 participants who listed it as an interest, only 11 listed it as an issue. Fifteen participants suggested that epidemiology (miscellaneous) was a main issue for the reviews, but only 5 had cited this category as an interest. Ten out of 14 participants who listed interventions (miscellaneous) as an interest also listed it as an issue. However, a further 12 participants who did not list it as an interest also saw it as an issue. Mental illness was an interest for 14 participants, 7 of whom also saw it as an issue.

3.26 Overall, a substantial proportion of participants suggested issues for focus that did not concur with their specialist interest areas. It can be concluded that participants did not display bias towards their own specialist interest areas when considering what issues a series of reviews of the evidence around suicide and suicidal behaviour should focus on .

Participants' perceptions of the main gaps in the evidence base

3.27 Participants were asked to record up to 5 main gaps they perceived to exist in the current evidence base relevant to the implementation of the Scottish Executive's suicide prevention strategy. The participants identified a total of 294 gaps, 102 of which were identified by UK participants and 192 by international participants. The gaps identified could be categorised into 66 different themes; a full list can be found in annex 6. Table 3.8 below gives a summary of the gaps that were most commonly mentioned.

3.28 In respect of table 3.8, a common gap was defined as a gap mentioned 7 times or more by participants in the whole sample . The numbers in cells in bold represent the most commonly mentioned gaps.

3.29 Table 3.8 shows that, in the opinion of the participants, the main 3 gaps in evidence were all related to interventions. Effectiveness of interventions and preventative interventions were the most prominent gaps in the current evidence base (13% of all gaps mentioned). Participants also perceived gaps in the research evidence available on epidemiology, current data collection methods and the impact of medication on suicide.

3.30 UK participants perceived gaps in the evidence base around deliberate self-harm (7%) more commonly than international participants (1%). Effectiveness of interventions, deliberate self-harm and epidemiology were key gaps perceived by UK participants (8%, 14% and 5 %, respectively, of all UK gaps mentioned). Interventions spanning prevention and integrated models of intervention were the most common gaps for international participants.

3.31 A comparison of specialist interests (table 3.6) and identified gaps (table 3.8) shows some overlap but also some difference at the individual level. Seven out of the 19 participants who listed prevention as a gap had also listed prevention as an interest. Five out of 9 participants who listed epidemiology (miscellaneous) as a gap had also listed it as an interest. Three out of 13 participants who listed interventions (miscellaneous) as a gap had also listed it as an interest and 6 out of 9 participants who listed deliberate self-harm as a gap had also listed it as an interest.

3.32 Overall, for the key perceived gaps, fewer than half the participants identified gaps in the evidence base that concurred with their specialist interest areas, although many identified evidence gaps in understanding in their own specialist interest area, suggesting that they might be working to address this gap in some way.

Comments offered on additional components of the reviews

3.33 Participants suggested that the reviews should consider the role of various research methodologies and approaches within research on suicide and suicidal behaviour. Several participants emphasised the importance of an appreciation of the value of qualitative and mixed methods research and the use of multi-factorial, e.g. biopsychosocial, models of suicide and suicidal behaviour as a basis for the reviews. It was suggested that the reviews consider effectiveness of interventions based on different types of evidence, from randomised controlled trials to ethnographic qualitative studies of those affected by suicidal behaviour. Some participants felt that it would be important to include service user perspectives in the design and conduct of the reviews.

3.34 It was also suggested that the reviews should take into consideration whether the research evidence has a theoretical basis, whether evidence is generalisable and its suitability for replication and potential for transferability to the Scottish context. There was interest in reviews that would make international comparisons particularly on the impact of national suicide prevention strategies in different countries.

3.35 Participants also identified some key questions that they felt the reviews should address:

• What is known about the causes and determinants of suicide, what is unknown or inconclusive?

• What is the robustness of findings by risk group and issue?

• Where are there holes in the literature?

• What works?

• How can we identify effective preventative interventions?

• How can the impact or outcome of interventions be evaluated, monitored or measured at all levels?

• How can best use be made of local resources on all levels?

Developing a framework for dividing up the reviews

3.36 Using the findings from the survey data, the research team developed a framework model ( see figure 3.1) to illustrate how the territories for the reviews might be divided up.

Elements of the framework for dividing up the reviews

3.37 The framework firstly takes cognisance of the survey participants' ideas for the most useful ways to divide up the reviews ( see table 3.4). Therefore the framework divides understanding suicide and suicidal behaviour into the following main elements:

• Epidemiology (incidence)

• Suicide and deliberate self-harm

• Determinants, including societal, psychosocial, individual and service level risk and protective factors

• Interventions, including prevention, treatment and postvention at societal, psychosocial, individual and service levels

3.38 Other key ways of dividing up the reviews suggested by the survey participants in table 3.4, such as considering evidence relevant to suicide and suicidal behaviour over different stages in the life span or across high risk groups, could be incorporated into the framework as cross-cutting themes rather than as key elements.

Assessing the validity of the framework

3.39 The validity of the framework for dividing up the reviews can be assessed on the basis of its ability to incorporate all the issues that participants identified as important for focus in the reviews and all the perceived gaps in research. All the themes identified for issues (table 3.7 and annex 5) were matched to the framework elements. The following patterns emerged:

• There were more matches to the epidemiology of suicide (19) than to deliberate self-harm (4)

• Across the framework, there were more matches to interventions (134) than to determinants (76)

• Within determinants more matches were to general risk conditions (12) than to general protective conditions (9)

• Within risk conditions most matches were to risk conditions at the individual level (32), followed by societal level (18) and then psychosocial environments (4)

• Quality of services had only one match; this is because this element crosses over with quality of service interventions

• Within protective factors most matches were at the individual level (5), followed by societal (2), the psychosocial environment (1) and quality of services (1)

• Within interventions most matches were to promotion and prevention (40), then treatment (28) and postvention (1)

• At the next level most matches were to quality of services (52), then the individual level (11), societal (7) and psychosocial environment (3), however, it is likely that there is some overlap between the individual level and the quality of services which are targeted at individual risk groups.

3.40 When matches were made between the framework elements and key gaps identified (table 3.8 and annex 6), the following patterns emerged:

• There were more matches to the epidemiology of suicide (13) than to deliberate self-harm (3)

• Across the framework there were more matches to interventions (186) than to determinants (64)

• In broad terms, there were more matches to protective factors (8) than to risk factors (3), however, within risk conditions most matches were to individual (21) and societal levels (15), then psychosocial environment levels (3) and quality of services (1)

• Within protective factors there was little difference in the number of matches at individual level (4), psychosocial environments (3), societal (2) and quality of services (1)

• Within interventions most matches were to prevention (38), then treatment (29) and postvention (2)

• At the next level most matches were to quality of services (72), psychosocial environment (12), societal (11) and then individual (4) levels.

3.41 These patterns broadly reflect the priority issues for focus and perceived gaps in the evidence base generated through the survey. Because all the issues and gaps themes fit into the framework, it provides a robust tool for considering the links between survey responses.

KEY ISSUES ARISING FROM SURVEY

3.42 The following points give an overview of the key issues arising from the survey:

• The survey was useful in terms of producing a set of priority issues to guide the focus and structure of the research reviews on suicide and suicidal behaviour

• There was disparity between the research specialist's interest areas and the priority issues for focus of the research reviews they suggested.

• The survey has also served to raise some awareness across the world about Scotland's suicide prevention strategy and build an international network that Scottish practitioners, policy makers and researchers can access

• Survey participants were clear that the reviews should be split between suicide and deliberate self-harm, and across determinants, interventions and epidemiology

• The survey participants were most commonly interested in interventions for the prevention of suicide

• An extensive range of issues on which the reviews should focus and perceived gaps in the evidence on suicide and suicidal behaviour were suggested

• Participants felt that the reviews should focus on what is known about what causes suicide and how it can be prevented, with relevance to how local resources can be targeted effectively

• The key priorities for focus were understanding more about preventative interventions for risk groups, particularly those with mental illness and those who misuse alcohol and drugs (at service level), and for the general population (at societal level), another key suggestion for focus was epidemiological studies

• Participants perceived gaps in the evidence to be concentrated around the effectiveness of preventative interventions, interventions for deliberate self-harm and good quality epidemiological data

• The framework is a useful tool for considering the links between the issues and gaps identified in the survey within the key elements considered by the participants to be of most importance in understanding suicide and suicidal behaviour.

« Previous | Contents | Next »