Suicide prevention is a critical priority for the Veterans Health Administration (VHA), as Veterans face a suicide rate 1.5 times higher than the general U.S. adult population. The Veterans Crisis Line (VCL), VHA’s 27/7 crisis support hotline, has managed over 5 million calls since 2007. Caring Letters interventions, involving periodic mailings to individuals expressing care and concern, are effective in reducing suicide rates among high-risk individuals post-hospitalization. However, the effectiveness of caring letters for crisis line callers has not been evaluated previously.
New Research
In April 2024, evaluators from the Puget Sound VA and Partnered Evidence-based Policy Resource Center (PEPReC) published a paper “Caring Letters Sent by a Clinician or Peer to At-Risk Veterans” in the Journal of the American Medical Association Network Open. The primary objective was to examine the outcomes of the VCL caring letters intervention and determine whether there were differences in outcomes by signatory (letters sent by a clinician versus a peer). Authors previously reported details implementing the intervention.
Methods
Authors conducted an effectiveness-implementation hybrid type 1 evaluation. They performed a randomized evaluation of the impact of peer versus clinician letters on outcomes, as well as an observational evaluation of outcomes associated with caring letters receipt. Data collection included linking VCL contact records with national VHA administrative. All Veterans who contacted the VCL with an identifiable address in the VHA’s Corporate Data Warehouse (CDW) and who contacted the crisis line between June 11, 2020, and June 10, 2021, were screened for inclusion in the caring letters cohort. Veterans were randomized to receive 9 caring letters for 1 year from either a clinician or peer Veteran signatory. Enrollment occurred between June 11, 2020, and June 10, 2021, with 1 year of follow-up. Analyses were completed between July 2022 and August 2023.
Main Findings
Peer signatory and clinician signatory recipients had similar rates of suicide attempts within 12 months of their call to VCL (7.4%, 7.6%, respectively). They also had similar rates of secondary outcomes, including mortality (3.8% of peer and 3.7% of clinician recipients died within 12 months of the index call) and health care use (69.0% of peer recipients and 68.5% of clinician recipients had at least 1 outpatient mental health visit in the 12 months following the index call).
When comparing outcomes among those who received either type of caring letters to those who did not receive caring letters, 6,801 caring letter recipients (7.7%) and 10,910 nonrecipients (7.8%) had a suicide attempt in the 12 months following the index call. There was no evidence of an association between receipt of caring letters and mortality. However, Veterans who received caring letters were significantly more likely to have subsequent inpatient and outpatient care (e.g., mental health care, ED visits).
Conclusion
Caring letters were not associated with suicide attempts or all-cause mortality, but they were associated with higher probabilities of outpatient and inpatient mental health care use, ED visits, and any outpatient or inpatient VA care use. No differences in outcomes were identified when cards were sent from a peer Veteran vs a clinician the recipient had not met. These results will be used by the VCL to optimize the caring letters intervention.
PEPReC, within the Veterans Health Administration and funded in large part by the Quality Enhancement Research Initiative (QUERI), is a team of health economists, health services and public health researchers, statistical programmers, and policy analysts who support VA efforts to improve Veterans’ lives through evidence-driven innovations using advanced quantitative methods.
