Caring Letters and Suicide Mortality Among US Veterans Who Contacted the Veterans Crisis Line

Importance: Caring letters (CL) is one of the only suicide prevention interventions to ever reduce suicide mortality in a randomized clinical trial, but the results have not been replicated. The US Department of Veterans Affairs Veterans Crisis Line (VCL) CL intervention was previously evaluated, but suicide mortality data were not available at that time. Objectives: To examine the association of the VCL CL intervention with suicide mortality among US veterans, and to determine whether there are differences in outcomes by signatory. Design, Setting, and Participants: This prespecified secondary analysis of a randomized clinical trial includes a randomized parallel trial of US veterans, combined with an observational comparison of no CL receipt with any CL receipt. Veterans Health Administration (VHA) patients (aged ≥18 years) who contacted the VCL between June 11, 2020, and June 10, 2021, were randomized to receive CL from either a VA clinician or a peer veteran signatory. Veterans who contacted the VCL between June 11, 2018, and June 10, 2021, and did not receive the intervention were included as an observational historical control group. Suicide mortality analyses were completed between March and September 2025. Intervention: Veterans received 9 CL messages over a period of 1 year from either a clinician or peer veteran signatory. The signatory was randomly assigned. Main Outcomes and Measures: The main outcome for these analyses was suicide mortality among veterans in the 12 months following the index VCL contact. Suicide mortality was compared using χ2 tests and time-to-event models. Results: This study included 227 502 VHA patients (mean [SD] age, 53.6 [16.9] years; 84.4% men) who contacted the VCL. A total of 87 926 veterans received the CL intervention; the remaining 139 576 comprised the historical control group. A total of 625 suicides (0.27%) occurred during the study period. The suicide mortality rate was similar between the CL and historical control groups (0.29% and 0.27%, respectively; P = .31). Within the CL group, outcomes did not differ between peer letter and clinician letter recipients (0.21% and 0.22%; P = .95). Conclusions and Relevance: In this analysis, among VHA patients who contacted the VCL, CL receipt was not associated with suicide mortality, and no differences in outcomes were identified by signatory. Future research should explore the potential benefits of increased health care utilization, a finding reported in the parent trial. Trial Registration: isrctn.org Identifier: ISRCTN27551361.