Background
Methods
Study aim and design
Study procedures and setting
Search strategy and selection criteria
Data extraction and management
Study quality
Description of materials
Data analysis
Results
Overview of the interventions
Quality assessment
Universal campus-wide interventions
Authors (year of publication) | Details of the intervention | Study design | Target of the intervention | Main findings | Effect Size |
---|---|---|---|---|---|
Aseltine et al., 2004 [42] | Signs of Suicide (SOS) prevention program | RCT (wait-list control group with follow up at 3 months postintervention) | Reduce suicidal ideation | No significant reduction in suicidal ideation. | Small (d = 0.15) |
Reduce suicide attempts | Significant reduction in suicide attempts. | Small (d = 0.26)a | |||
Change knowledge | Significant increase in knowledge. | Small (d = 0.38)a | |||
Change attitudes | Significant increase in adaptive attitudes toward suicide. | Small (d = 0.40)a | |||
Increase in help-seeking behaviour | No significant increase in help seeking in the form of treatment. No significant increase in help seeking from a friend. No significant increase in help seeking from an adult. | Small (d = 0.12) Small (d = 0.04) Small (d = 0.13) | |||
Aseltine et al., 2007 [43] | Signs of Suicide (SOS) prevention program | RCT (wait-list control group with follow up at 3 months postintervention) | Reduce suicidal ideation | No significant reduction in suicidal ideation. | Small (d = 0.10) |
Reduce suicide attempts | Significant reduction in suicide attempt. | Small (d = 0.26) | |||
Change knowledge | Significant increase in knowledge about suicide. | Small (d = 0.32) | |||
Change attitudes | Significant increase in adaptive attitudes toward suicide. | Small (d = 0.10) | |||
Increase in help-seeking behaviour | No significant increase in help seeking in the form of treatment. No significant increase in help seeking from a friend. No significant increase in help seeking from an adult. | Small (d = 0.02) Small (d = 0.02) Small (d = 0.08) | |||
Britton et al., 2014 [44] | Classroom-based, teacher implemented, mindfulness meditation intervention | RCT (active control condition with postintervention follow up only) | Reduce suicidal ideation Reduce suicidal behaviour | Significant reduction in suicidal ideation and self-harm. | Large (d = 1.13) |
Kalafat & Elias, 1994 [45] | Suicide awareness curriculum | Solomon four-groups design (physical education content control group with postintervention follow up only) | Change knowledge | Significant increase in suicide knowledge. | Large (d = 0.91) |
Change attitudes | Significant change in positive attitudes towards suicide and helping others. | Small (d = 0.17) | |||
Nasution et al., 2019 [46] | Combination of training for mental health nurses (TKN), CBT and a peer leadership (PL) program | Quasiexperimental pre-post test design (care as usual control group with postintervention only follow up) | Reduce suicidal ideation | Significant reduction in suicidal ideation. | Insufficient statistics |
Randell et al., 2001 [47] | C-CARE - Counselors CARE | Experimental three-group design (care as usual control group with follow up at 4-weeks and 10-weeks) | Reduce suicide risk: unspecified | Significant reduction in suicide risks behaviours (thoughts, threats, and attempts) in intervention and control group. | Large (d = 0.90) |
Randell et al., 2001 [47] | CAST - C-CARE plus a 12-session Coping and Support Training | Experimental three-group design (care as usual control with follow up at 4-weeks and 10-weeks) | Reduce suicide risk: unspecified | Significant reduction in suicide risks behaviours (thoughts, threats, and attempts) in intervention and control group. | Large (d = 0.90) |
Rogers et al., 2018 [48] | Psychoeducation and Interpersonal exposure | RCT (control group browsed the National Diabetes Education website with a postintervention and 1-month follow up) | Reduce stigma | Significant reduction in stigma of suicide. | Small (d = 0.46) |
Significant greater reduction in suicide-related stigma among those with prior exposure to suicide at post-intervention. | Small (d = 0.28) | ||||
Significant greater reduction in suicide-related stigma among those with prior exposure to suicide at one-month follow-up. | Small (d = 0.40) | ||||
Schilling et al., 2016 [49] | Signs of Suicide (SOS) prevention program | RCT (wait-list control group with a 12-weeks post-baseline follow up) | Reduce suicidal ideation | No significant reduction in suicidal ideation. | Small (d = 0.01) |
Reduce suicide plan | Significant reduction in suicide plan. | Large (d = 1.05) | |||
Reduce suicide attempts | Significant reduction in suicide attempt. | Large (d = 0.72) | |||
Change knowledge | Significant increase in knowledge. | Small (d = 0.28) | |||
Change attitudes | Significant change in adaptive attitudes about suicide. | Small (d = 0.05) | |||
Wasserman et al., 2015 [50] | Question, Persuade, Refer (QPR) | Cluster-RCT (control group exposed to six educational posters displayed in their classrooms with a 3-month and 12-month follow up) | Reduce suicidal behaviour | No significant reduction in suicidal behaviour at 3-month follow up. | Small (d = −0.26) |
No significant reduction in suicidal behaviour at 12-month follow-up. | Small (d = −0.20) | ||||
Wasserman et al., 2015 [50] | ProfScreen | Cluster-RCT (control group exposed to six educational posters displayed in their classrooms with a 3-month and 12-month follow up) | Reduce suicide attempts | No significant reduction in likelihood of suicide attempt at 3-month follow-up. | Small (d = − 0.14) |
Significant reduction in likelihood of suicide attempt at 12-month follow-up. | Large (d = −0.44) | ||||
Wasserman et al., 2015 [50] | Youth Aware of Mental Health Programme (YAM) | Cluster-RCT (control group exposed to six educational posters displayed in their classrooms with a 3-month and 12-month follow up) | Reduce suicide attempts | No significant reduction in the likelihood of suicide attempts at 3-month follow-up. | Small (d = − 0.14) |
No significant reduction in the likelihood of suicide attempts at 12-month follow-up. | Small (d = −0.44) | ||||
Wulandari et al., 2019 [51] | Peer leadership training (team formation and building, adolescent related projects, and team activities) | Quasi-experimental pretest–posttest design (no control group with postintervention follow up) | Reduce suicidal ideation | Significant reduction in suicidal ideation. | Insufficient statistics |
Wyman et al., 2010 [52] | Source of Strengths prevention program | RCT (wait-list control group with postintervention and 1 year follow up) | Increase help-seeking behaviour | Significant increase in positive expectation that adults at school would help suicidal students. | Medium (d = 0.75)a |
Significant increase in norms for help-seeking from adults at school. | Medium (d = 0.62)a | ||||
No significant increase in connecting distressed peers to adults. | Small (d = 0.21)a |
Reducing non-fatal suicidal behaviour
Changing knowledge, attitudes, and stigma
Increasing help-seeking behaviour
Selective interventions
Authors (year of publication) | Details of the intervention | Study design | Target of the intervention | Main findings | Effect Size |
---|---|---|---|---|---|
Gatekeeper interventions for students | |||||
Mitchell et al., 2013 [53] | Brief psychoeducation Question, Persuade, Refer (QPR) gatekeeper training | Open trial (no control group with a postintervention and 3 to 6 month follow up) | Change knowledge | Significant increase in knowledge of suicide prevention facts. | Large (d = 1.46) |
Increase in help-seeking behaviour | No significant increase in ability to referred anyone to on-campus mental health services. | Small (d = 0.12) | |||
Pasco et al., 2012 [54] | Campus connect (didactic training and experimental exercises) gatekeeper training | Open trial (control group received a 1.5-h adapted format of Campus Connect with postintervention only) | Increase in help-seeking behaviour | Significant increase in crisis intervention skills. | Large (d = 1.21)a |
Rallis et al., 2018 [55] | Brief Psychoeducation and experimental (modelled after the Campus Connect training) | Open trial (no control group and postintervention and 3 month follow up) | Change knowledge | Significant increase in declarative knowledge. | Large (d = 1.62)a |
Significant increase in perceived knowledge. | Large (d = 1.41)a | ||||
Significant reduction in declarative knowledge at 3-month follow-up. | Large (d = 0.94)a | ||||
Significant reduction in perceived knowledge at 3-month follow-up. | Large (d = 1.10)a | ||||
Increase in help-seeking behaviour | Significant increase in identifying any suicidal students. | Small (d = 0.12)a | |||
Significant increase in making at least one referral. | Small (d = 0.24)a | ||||
Taub et al., 2013 [56] | Knowledge and crisis communications skills | Open trial (no control group and postintervention follow up) | Change knowledge | Significant increase in knowledge of suicide among new resident assistants. | Small (ηp2 = 0.16)a |
Significant increase in knowledge of suicide warning signs among new resident assistants. | Small (ηp2 = 0.24)a | ||||
Significant increase in places to refer among new resident assistants. | Small (ηp2 = 0.30)a | ||||
No significant increase in knowledge of suicide among returning resident assistants. | Small (ηp2 = 0.00)a | ||||
No significant increase in suicide warning signs among returning resident assistants. | Small (ηp2 = 0.00)a | ||||
No significant increase in places to refer among returning resident assistants. | Small (ηp2 = 0.00)a | ||||
Increase in help-seeking behaviour | No significant prediction of crisis communication skills among new resident assistants. | Small (ηp2 = 0.00)a | |||
No significant prediction of crisis communication skills among returning resident assistants. | Small (ηp2 = 0.15)a | ||||
Tompkins and Witt, 2009 [57] | Brief psychoeducation Question, Persuade, Refer (QPR) gatekeeper training | Quasi-experimental non-equivalent control group design (control group option to be waitlisted or treatment as usual with postintervention and 6 month follow up) | Change knowledge | Significant increase among intervention group for self-evaluation of knowledge. | Medium (d = 0.51)a |
Increase in help-seeking behaviour | Significant increase among intervention group for perceived efficacy to refer. | Small (d = 0.49)a | |||
Wachter Morris et al., 2015 [58] | The ALIVE @ Purdue train-the trainers program | Open trial (no control group with postintervention follow up) | Change knowledge | No significant increase in knowledge about suicide. | Medium (d = 0.62)a |
No significant increase in knowledge about potential warning signs. | Small (d = 0.14)a | ||||
No significant increase in knowledge about places to refer. | Small (d = 0.00)a | ||||
Increase in help-seeking behaviour | Significant increase in crisis-related communication skills. | Large (d = 0.95)a | |||
Gatekeeper training programmes for staff | |||||
Cimini et al., 2014 [59] | Gatekeeper training (tailored to group specific needs) involving didactic and experiential learning components highlighting the opportunity for behavioural rehearsal | Open trial (no control group with postintervention and 3-month follow up) | Change knowledge | Significant increase in knowledge about suicidal behaviour at postintervention. | Large (d = 0.78)a |
Significant reduction in knowledge about suicidal behaviour at follow up assessment. | Small (d = 0.4)a | ||||
Increase in help-seeking behaviour | Significant increase in comfort level to intervene with suicidal behaviour at postintervention. | Medium (d = 0.74)a | |||
Significant reduction in comfort level to intervene at follow up assessment but remained significantly higher than baseline. | Medium (d = 0.58)a | ||||
Cross et al., 2010 [60] | Brief psychoeducation - QPR (Question, Persuade, Refer) gatekeeper training | Open trial (no control group with a postintervention follow up) | Change knowledge | Significant increase in knowledge about suicide at postintervention assessment. | Large (d = 2.28)a |
Increase in help-seeking behaviour | Significant increase in perceived efficacy to intervene in suicide at postintervention assessment. | Large (d = 2.94)a | |||
Hashimoto., 2016 [61] | Gatekeeper-training based on the mental health first aid program | Single-arm follow-up study (no control group with postintervention and 1-month follow up) | Increase in help-seeking behaviour | Significant improvement in the competence of managing suicidal students and behavioural intention at postintervention. | Small (d = 0.46) |
Significant improvement in the competence of managing suicidal students and behavioural intention at follow-up. | Small (d = 0.35) | ||||
Significant improvement in the confidence of managing suicidal students and behavioural intention at postintervention. | Medium (d = 0.59) | ||||
Significant improvement in the confidence of managing suicidal students and behavioural intention at follow-up. | Small (d = 0.35) | ||||
Mclean et al., 2017 [62] | Adapted version of brief psychoeducation - Question, Persuade, Refer (QPR) gatekeeper training | RCT (stress and time management skills training program with a 16 weeks postintervention follow up) | Increase in help-seeking behaviour | Non-significant increase in number of interventions performed. | Small (ηp2 = 0.002)a |
Non-significant increase in number of times approached by a resident. | Small (ηp2 = 0.001)a | ||||
Non-significant increase in number of suicidal residents reported. | Small (ηp2 = 0.005)a | ||||
Non-significant increase in suicidal thought severity. | Small (ηp2 = 0.012)a | ||||
Shannonhouse et al., 2017 [63] | Brief Psychoeducation -Applied Suicide Intervention skills training (ASIST) | Quasi-experimental pretest–posttest design (wait-list control group with postintervention follow up only) | Change knowledge | Significant increase in knowledge about suicide across time. | Small (ηp2 = 0.28)a |
Change attitudes | Significant increase in participants’ attitudes about suicide across time. | Small (ηp2 = 0.32)a | |||
Increase in help-seeking behaviour | Significant increase in comfort to respond to persons-at-risk. | Small (ηp2 = 0.25)a | |||
Significant increase in competence to respond to persons-at-risk. | Small (ηp2 = 0.38)a | ||||
Significant increase in confidence to respond to persons-at-risk. | Small (ηp2 = 0.14)a | ||||
Wyman et al., 2008 [64] | QPR (Question, Persuade, Refer) gatekeeper training versus wait-list control group | RCT (wait-list control group with postintervention and 1 year follow up) | Change knowledge | Significant increase in self-reported knowledge. No significant increase noted among staff who received a 30-min refresher training several months after initial training. | Small (d = 0.41)a |
Increase in help-seeking behaviour | Significant increase in appraisals of efficacy to perform a gatekeeper role. | Large (d = 1.22)a | |||
Significant increase in access to services for suicidal students. | Small (d = 1.07)a | ||||
No significant increase in comfort in asking about suicide. | Small (d = 0.18)a | ||||
No significant increase in referral behaviours. | Small (d = 0.07)a | ||||
No significant increase in asking about distress. | Small (d = 0.27)a | ||||
Gatekeeper training programmes for staff and students | |||||
Indelicato et al., 2011 [65] | Brief psychoeducation - QPR (Question, Persuade, Refer) gatekeeper training | Between-subjects design (no control group with 1 month and 3 month postintervention follow up) | Change knowledge | Significant increase in self-reported knowledge about suicide. | Insufficient statistics |
Significant increase in self-reported knowledge about facts on suicide prevention. | Insufficient statistics | ||||
Significant increase in self-reported knowledge about warning signs of suicide. | Insufficient statistics | ||||
Significant increase in self-reported knowledge about how to ask someone about suicide. | Insufficient statistics | ||||
Significant increase in self-reported knowledge about how to persuade someone to get help. | Insufficient statistics | ||||
Significant increase in self-reported knowledge about how to get help for someone. | Insufficient statistics | ||||
Significant increase in self-reported knowledge about information about local resources. | Insufficient statistics | ||||
Significant increase in self-reported knowledge about belief that asking about suicide is appropriate. | Insufficient statistics | ||||
Significant increase in self-reported knowledge about likelihood to ask someone about thoughts of suicide if concerned for them. | Insufficient statistics | ||||
Increase in help-seeking behaviour | Significant increase in confidence in how to respond to the situation. | Insufficient statistics | |||
Significant increase in comfort talking about suicide. | Insufficient statistics | ||||
Significant increase in effectiveness of the suicide prevention intervention. | Insufficient statistics | ||||
No significant were found regarding making a referral for help and taking the person to a mental health professional. | Insufficient statistics |
Gatekeeper training programmes for students
Gatekeeper training programmes for staff
Gatekeeper training programmes for staff and students
Indicated interventions for high-risk students
Authors (year of publication) | Details of the intervention | Study design | Target of the intervention | Main findings | Effect Size |
---|---|---|---|---|---|
Eggert et al., 1995 [67] | At risk high school students - assessment protocol plus 1-semester Personal Growth Class (PGC l) | Controlled before and after study (care-as-usual control group with a postintervention follow up) | Reduce suicidal behaviour | A total of 85% of the youth in Groups I reduced suicide-risk behaviours by 25%, with Group I showing a greater decline in suicide-risk behaviours than Group II. | Insufficient statistics |
Eggert et al., 1995 [67] | At risk high school students - assessment protocol plus 2- semesters Personal Growth Class (PGC ll) | Controlled before and after study (care-as-usual control group with a postintervention follow up) | Reduce suicidal behaviour | A total of 65% of Group II showed reduced suicide-risk behaviours by 25%. | Insufficient statistics |
Eggert et al., 2002 [68] | Counselors-CARE (C-CAST): assessment interview, counselling session, and social ‘connections’ intervention | Controlled before and after study (care-as-usual control group with a postintervention and 10 week follow up) | Reduce suicidal behaviour | Statistics were not reported for interventions and control separately. Group x Trend Interaction demonstrate that the pattern of change differed significantly between at least one of the three groups. Not clear from the stats how these differed. | Insufficient statistics |
Eggert et al., 2002 [68] | CAST: combination of the C-CARE intervention (i.e. assessment interview, counselling session, and social ‘connections’ intervention) followed by a small group prevention program | Controlled before and after study (care-as-usual control group with a postintervention and 10 week follow up) | Reduce suicidal behaviour | Statistics were not reported for interventions and control separately. Group x Trend Interaction demonstrate that the pattern of change differed significantly between at least one of the three groups. Not clear from the stats how these differed. | Insufficient statistics |
Fitzpatrick et al., 2005 [69] | Brief video intervention regarding problem solving and coping skills | RCT (time-matched intervention focusing on physical health issues for control group with a 1 week, 2 weeks, and 1-month postintervention follow up) | Reduce suicidal ideation | No significant difference between intervention and control group with regard to suicidal ideation at baseline. | Insufficient statistics |
Fukumori et al., 2017 [70] | Three-day individual intervention program of structured writing that incorporates the emotional regulation group program and the DBT workbook | RCT (wait-list control group with a postintervention, 2 week and 1-month follow up) | Reduce suicidal ideation | No significant reduction in suicidal ideation. | Small (d = 0.35)a |
Hetrick et al., 2017 [71] | Internet-based cognitive behavioural therapy (Reframe-IT) | RCT (treatment-as-usual control group with a 10 week and 22 week postintervention follow up) | Reduce suicidal ideation | No significant reduction in suicidal ideation at postintervention assessment. | Small (d = −0.35) |
King et al., 2015 [72] | Electronic bridge mental health services (eBridge) | RCT (treatment-as-usual control group with an 8 week postintervention follow up) | Increase in help-seeking behaviour | Significant increase in readiness to intervene with own suicidal behaviour by talking to family. | Large (d = 2.74)a |
Significant increase in readiness to intervene with own suicidal behaviour by talking to a friend. | Large (d = 2.48)a | ||||
Significant increase in readiness to intervene with own suicidal behaviour by seeing a mental health professional. | Large (d = 3.16)a | ||||
No significant increase in readiness to seek information. | Large (d = 1.60)a | ||||
No significant increase in readiness to seek out self-help or a support group. | Small (d = 0.50)a | ||||
No significant increase in readiness to seek academic support services. | Small (d = −0.44)a | ||||
Reduce stigma | Significant reduction in level of personal stigma scores at postintervention. | Large (d = −1.07)a | |||
Significant reduction in level of perceived public stigma at postintervention. | Medium (d = −0.59)a | ||||
Lin et al., 2019 [73] | Cognitive therapy group program | RCT (cognitive therapy control group with a 4-, 8-, 20-, and 32-week postintervention follow up) | Reduce suicidal ideation | Significant reduction in suicidal ideation at 4 weeks follow up. | Large (d = 5.24) |
Significant reduction in suicidal ideation at 8 weeks follow up. | Large (d = 4.39) | ||||
Significant reduction in suicidal ideation at 20 weeks follow up. | Large (d = 3.67) | ||||
Significant reduction in suicidal ideation at 32 weeks follow up. | Large (d = 3.30) | ||||
Reduce suicidal behaviour | Significant reduction in suicide attempt 4 week follow up. | Small (d = 0.32) | |||
Significant reduction in suicide attempt at 8 weeks follow up. | Small (d = 0.23) | ||||
Significant reduction in suicide attempt at 20 weeks follow up. | Small (d = 0.18) | ||||
Significant reduction in suicide attempt at 32 weeks follow up. | Small (d = 0.14) | ||||
Lin et al., 2019 [73] | Dialectical behaviour therapy group program | RCT (cognitive therapy control group with a 4-, 8-, 20-, and 32-week postintervention follow up) | Reduce suicidal ideation | Significant reduction in suicidal ideation found at 4 weeks follow up. | Large (d = 5.24) |
Significant reduction in suicidal ideation found at 8 weeks follow up. | Large (d = 4.39) | ||||
Significant reduction in suicidal ideation found at 20 weeks follow up. | Large (d = 3.67) | ||||
Significant reduction in suicidal ideation found at 32 weeks follow up. | Large (d = 3.30) | ||||
Reduce suicidal behaviour | Significant reduction in suicide reattempt at 4 weeks follow up. | Small (d = 0.32) | |||
Significant reduction in suicide reattempt at 8 weeks follow up. | Small (d = 0.23) | ||||
Significant reduction in suicide reattempt at 20 weeks follow up. | Small (d = 0.18) | ||||
Significant reduction in suicide reattempt at 32 weeks follow up. | Small (d = 0.14) | ||||
Pistorello et al., 2012 [74] | 12-month long term Dialectical Behaviour Treatment | RCT (optimised treatment-as-usual control group with a 3 month and 18 month follow up) | Reduce suicidal behaviour | Significant reduction in suicidality (i.e., suicidal thoughts and the person’s estimation of the likelihood they would consider, attempt, and die from suicide in the future). | Medium (d = 0.53)a |
Tang et al., 2009 [75] | Program of Intensive Interpersonal Psychotherapy for depressed adolescents with suicidal risk (IPT-A-IN) | RCT (treatment-as-usual control group with a postintervention follow up) | Reduce suicidal ideation | Significant reduction in suicidal ideation. | Medium (d = −0.78) |
Thompson et al., 2000 [76] | Personal Growth Semester 1 | Three-group, repeated measures design (Measure of Adolescent Potential for Suicide control group with 18 week postintervention follow up) | Reduce suicidal behaviour | Significant reduction in suicide risk behaviours. | Small (d = 0.12) |
Thompson et al., 2000 [76] | Personal Growth Semester 2 | Three-group, repeated measures design (Measure of Adolescent Potential for Suicide control group with 18 week postintervention follow up) | Reduce suicidal behaviour | Significant reduction in suicide risk behaviours. | Small (d = 0.21) |
Xavier et al., 2019 [77] | Problem solving intervention | RCT (care-as-usual control group with 1-, 3-, and 6 month follow up | Unspecified: suicidal orientation | Significant reduction in suicidal orientation at postintervention assessment. | Large (ηp = 0.91)a |
Significant reduction in suicidal orientation at 6-months follow up assessment. | Medium (ηp = 0.65)a |
Discussion
• Expand research in LMICs and diverse cultural settings. • Conduct translational research to guide the cultural adaptation and application of suicide prevention interventions that have been developed and tested in high-income settings. • Develop and test interventions not premised on an “identify-and-refer” model of suicide prevention for use in low-resource environments where there are not adequate referral networks. • Increase epidmiological research and population survailance of suicdal behaviour among adolescents and young-adults in LMICs, to advocate for making suicide prevention a priority in high-schools and univeties. • Draw on implementation science research to better understand how the implementation of interventions influences their effectiveness. • Increase the number of high quality studies that have suicide deaths as the primary outcome. • Increase the use of well-designed multi-site studies to explore contextual variables influencing implementation and outcomes. • Utilise multi-site studies, where the campus is the unit of analysis and/or a key variable for assessing outcomes. • Utilise cluster randomization trials and co-ordination of studies across a large number of sites in a range of diverse settings. • Utilise well designed randomized controlled trials and pragmatic trials to culturally adapt and test gatekeeper training in LMICs. |