Effectiveness of a French program to prevent suicide reattempt (VigilanS)

Aim: Brief contact interventions (such as letters, green cards, telephone calls or postcards) for reducing suicide reattempt (SR) and suicide have been evaluated since the 1980s, but results have been inconsistent. VigilanS is one of these programs that has benefited patients hospitalized for suicide attempt (SA) after discharge in 2 departments of northern France since 2015. The purpose of this study is to demonstrate its effectiveness in reducing SR.

Methods: Patients exposed to VigilanS in 2016 were recruited from the medical administrative database of the program, and the nonexposed patients from a database of the medico-surgical ward outside the scope of the program. First, a Cox model was used to compare the probability of SR during the 12-month follow-up period between the 2 groups. Second, a propensity score using the variables sex, age, source, SA history and SA method was used to match the VigilanS-exposed and the nonexposed patients. A Cox model propensity score adjusted analysis was reiterated on the matched data.

Results: The exposed and nonexposed groups included 3,068 and 3,694 individuals, respectively. In the bivariate analyses, the cumulative probability of SR at 12 months was significantly lower in the exposed group (6.0%, 95% confidence interval (CI): 5.5-6.5%) than in the nonexposed group (16.8%, 95% CI: 15.9-17.7%; p < 0.001). In the Cox model, the hazard ratio of SR was 0.38 in the exposed patients (95% CI: 0.36-0.40, p < 0.001). After matching, the cumulative probability of SR at 12 months was 5.2% in exposed versus 22.2% in nonexposed patients (p < 0.001). In the propensity score-adjusted Cox model, the hazard ratio of SR in the exposed patients was 0.19 (95% CI: 0.14-0.24, p < 0.001).

Conclusion: The results suggest the effectiveness of this real-life program for reducing SR. However, VigilanS only benefits a portion of the patients hospitalized for SA and therefore could be extended.