Suicide is recognized widely as a problem, with nearly 800,000 people worldwide completing suicide every year. A history of having attempted suicide is an important risk factor for suicide. Suicide aftercare interventions aim to reduce future suicidal behavior (i.e., suicide, attempt, or ideation) of people who have attempted suicide. Aftercare refers to interventions that aim to benefit people who have attempted suicide and interventions that address their family members. Interventions might intend to facilitate psychosocial adjustment, prevent and reduce suicidal behavior in the future, and promote psychological well-being.
The purpose of this systematic review and the meta-analyses of key outcomes was to synthesize the existing evidence on aftercare interventions, addressing the following key questions (KQs) and subquestions:
- KQ1: What is the effect of aftercare interventions on uptake, retention, effectiveness measures, and unintended consequences for individuals who have attempted suicide?
- KQ1a: Do the effects vary by the intensity of the intervention?
- KQ1b: Do the effects vary by the type of intervention?
- KQ1c: Do the effects vary by intervention target?
- KQ1d: Do the effects vary by population?
- KQ2: What is the effect of aftercare interventions on uptake, retention, effectiveness measures, and unintended consequences for the family members of people who have attempted suicide?
- KQ2a: Do the effects vary by the intensity of the intervention?
- KQ2b: Do the effects vary by the type of intervention?
- KQ2c: Do the effects vary by intervention target?
- KQ2d: Do the effects vary by population?
Methods
We searched research databases (e.g., PubMed, PsycINFO, CINAHL, Web of Science) and trial registries (e.g.,
We included studies that evaluated the effects of an aftercare intervention on individuals with a history of having attempted suicide or those individuals' family members (broadly defined as family members, caregivers, or friends). Eligible studies included clinical trials (randomized controlled trials [RCTs] or nonrandomized trials) and evaluations of large-scale interventions, such as screening or monitoring programs, that reported on a concurrent or historic comparator (e.g., pre-post studies, cohort studies comparing two cohorts).
Two reviewers screened publications for inclusion, abstracted study-level information, and assessed the risk of bias of included studies. The primary outcome of the review was repeated suicide attempts. Critical appraisal focused on selection bias, performance bias, detection bias, attrition bias, and study-specific sources. The quality of the body of evidence (QoE) for the effect estimate of each outcome was evaluated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach.
Results
In total, 73 studies met inclusion criteria, including 52 RCTs. Risk of bias was acceptable with most issues due to selection bias and confounding.
Interventions showed a statistically significant reduction in further suicide attempts for intervention participants (relative risk [RR] 0.78; confidence interval [CI] 0.67, 0.91; 33 studies; moderate QoE). There was evidence of publication bias, but the effect remained significant after taking publication bias into account. In addition, interventions reported an effect on suicide deaths (RR 0.71; CI 0.50, 0.99; 16 studies; low QoE). Interventions also reduced depression (SMD –0.32; CI –0.57, –0.06; 17 studies; low QoE) and hopelessness (SMD –0.42; CI –0.78, –0.05; 10 studies; low QoE). Uptake of interventions varied widely by intervention. Similarly, the treatment retention of participants varied considerably. Only three studies reported on unintended consequences; these were adverse events of medications.
Regarding KQ1a, we found no indication that the intensity of the intervention is systematically associated with the treatment success. Although we found 21 studies comparing two interventions directly, we were unable to determine which intervention types systematically produce better outcomes for patients (KQ1b). We could not explore the effects of the intervention target (e.g., on those participants who attempted suicide versus on family members or both) because of the paucity of studies addressing family members of people who attempted suicide (KQ1c). Populations in the studies that we included were varied, but we found only two studies reporting on military samples, which hindered analyses to identify population-specific effects (KQ1d).
We were unable to meaningfully address the effects of interventions on family members (KQ2) because these were very rarely included in existing research studies. Studies did not evaluate interventions that incorporated family members, and studies rarely reported on the effects of the suicide aftercare interventions on family members.
Conclusions
Across studies, we found that suicide aftercare can reduce the risk of further suicide attempt, but we found no evidence that the type of intervention or intensity of it affected outcomes. Research is needed to explore which interventions will produce the greatest clinical improvements and reduction in future suicide attempts and to identify effective interventions for service members and for family members of people who have attempted suicide.
