Active contact and follow-up interventions to prevent repeat suicide attempts during high-risk periods among patients admitted to emergency departments for suicidal behavior: a systematic review and meta-analysis

Abstract

Background: There is evidence that several intervention types, including psychotherapy, reduce repeat suicide attempts. However, these interventions are less applicable to the heterogeneous patients admitted to emergency departments (EDs). The risk of a repeat suicide attempt is especially high in the first 6 months after the initial attempt. Therefore, it is particularly important to develop effective ED interventions to prevent repeat suicide attempts during this 6-month period.

Methods: We systematically reviewed randomized controlled trials of ED-initiated interventions for suicidal patients admitted to EDs using the databases MEDLINE, PsychoINFO, CINAHL, and EMBASE up to January 2015 in accordance with an a priori published protocol (PROSPERO: CRD42013005463). Interventions were categorized into four types, including active contact and follow-up interventions (intensive care plus outreach, brief interventions and contact, letter/postcard, telephone, and composite of letter/postcard and telephone), and a meta-analysis was conducted to determine pooled relative risks (RRs) and 95% confidence intervals (CIs) of a repeat suicide attempt within 6 months.

Results: Of the 28 selected trials, 14 were active contact and follow-up interventions. Two of these trials (n = 984) reported results at 6 months (pooled RR = 0.48; 95% CI: 0.31 to 0.76). There were not enough trials of other interventions to perform meta-analysis. Some trials included in the meta-analysis were judged as showing risk of bias.

Conclusion: Active contact and follow-up interventions are recommended for suicidal patients admitted to an ED to prevent repeat suicide attempts during the highest-risk period of 6 months.

Systematic review registration: PROSPERO CRD42013005463 (27 August 2013).

Keywords: Emergency department; Meta-analysis; Self-harm; Suicide; Systematic review.

Fig. 1

Study selection. Two and 11 trials, respectively, were included in a meta-analysis of the effect of active contact and follow-up interventions on repeat suicide attempts at 6 and 12 months. There have been no new publications on psychotherapy and pharmacotherapy interventions for suicide attempts since our previous meta-analysis. Therefore, we did not perform meta-analyses on the effect of these interventions in the present study

Fig. 2

Primary outcome: Suicide attempts within 6 months for active contact and follow-up interventions. Two trials of active contact and follow-up interventions reported suicide attempts within 6 months [41, 45]. Two trials were included in the meta-analysis [41, 45]. The number of included participants and the number of participants who made repeat suicide attempts in each trial are shown in Table 1. To assess heterogeneity, we used the Cochrane Q statistic to examine heterogeneity among the trials in each analysis. We regarded heterogeneity as substantial if the Cochrane Q test produced a low p-value (< 0.10)

Fig. 3

Secondary Outcome: Suicide attempts within 12 months for active contact and follow-up interventions. a The meta-analysis included 11 trials [, , , , –, , –52]. The number of included participants and the number of participants who made repeat suicide attempts in each trial are shown in Table 1. To assess heterogeneity, we used the I² and Cochrane Q statistics to examine heterogeneity among the trials in each analysis. We regarded heterogeneity as substantial if I² was greater than 30% or if the Cochrane Q test produced a low p-value (< 0.10). b We investigated publication bias by constructing a funnel plot and by using the Egger test