Can “deaths of despair” serve as a focus for planning and evaluating clinical and preventive services for Veterans?

• Published in: Psychiatry research 2020-03, Vol.285, p.112841-112841, Article 112841
• Main Authors: Katz, Ira R., Dent, Kallisse R., Morley, Sybil W, Hein, Tyler C., Hoff, Rani A., McCarthy, John F.
• Format: Article
• Language: English
• Subjects: Alcohol use disorder; Mortality; Substance use disorder; Suicide
• ISSN: 0165-1781; 1872-7123
• DOI: 10.1016/j.psychres.2020.112841

•“Deaths of despair” is a composite of suicide, drug-, and alcohol-related deaths.•Rates are higher in Veteran men in VA health care and lower in other Veteran men.•Rates for Veterans and others are associated with characteristics of the states.•Rates for suicide are associated with rates for alcohol- but not drug-related causes.•It may be useful to modify current strategies for prevention to address the full range of these outcomes. Deaths of despair, a composite outcome including suicide and drug- and alcohol-related deaths, have been increasing, especially in subpopulations and geographic areas sensitive to economic and social hardships. The Veterans Health Administration (VHA) has begun evaluating the utility of this concept to guide planning and evaluations of clinical and preventive services for Veterans. In this study, mortality rates for middle-aged American men for 2013 to 2017 were from CDC WONDER, and rates for all Veterans, those using VHA healthcare services (VHA-utilizers), and other (non-VHA) Veterans were derived from National Death Index data. Findings demonstrated that rates for the composite were higher in VHA-utilizers and lower in non-VHA Veterans than middle-aged American men, consistent with use of VHA services by Veterans with the greatest needs. State rates in Veteran men were significantly and positively correlated with state rates for American men, and both were correlated with other characteristics of the social environment. The lack of correlation between rates for suicide and drug-related deaths indicates that deaths of despair cannot be modeled by assuming parallel paths from reactions to community-based stressors to the component outcomes; models should allow for an impact of community characteristics on partitioning between outcomes.