Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality

Abstract

Objectives: We examined whether a reduction in youth suicide mortality occurred between 2007 and 2010 that could reasonably be attributed to Garrett Lee Smith (GLS) program efforts.

Methods: We compared youth mortality rates across time between counties that implemented GLS-funded gatekeeper training sessions (the most frequently implemented suicide prevention strategy among grantees) and a set of matched counties in which no GLS-funded training occurred. A rich set of background characteristics, including preintervention mortality rates, was accounted for with a combination of propensity score-based techniques. We also analyzed closely related outcomes that we did not expect to be affected by GLS as control outcomes.

Results: Counties implementing GLS training had significantly lower suicide rates among the population aged 10 to 24 years the year after GLS training than similar counties that did not implement GLS training (1.33 fewer deaths per 100 000; P = .02). Simultaneously, we found no significant difference in terms of adult suicide mortality rates or nonsuicide youth mortality the year after the implementation.

Conclusions: These results support the existence of an important reduction in youth suicide rates resulting from the implementation of GLS suicide prevention programming.

FIGURE 1—

Model of the comprehensive, community-based Garrett Lee Smith Memorial Suicide Prevention Program.

FIGURE 2—

Main and control outcomes after Garrett Lee Smith Memorial Suicide Prevention Program training implementation for (a) suicide among those aged 10–24 years, (b) suicide among those aged ≥ 25 years, (c) mortality among those aged 10–18 years, and (d) mortality among those aged 19–24 years: Substance Abuse and Mental Health Services Administration cross-site evaluation, United States, 2007–2010. Note. Solid lines represent the estimated trajectory of the outcome after Garrett Lee Smith (GLS) program training implementation. Dashed lines represent the estimated trajectory of the outcome during the same period had GLS not been implemented. The 90% and 50% confidence intervals around the trajectory are in dark and light gray, respectively.