Comorbidity Correlates of Death Among New Veterans of Iraq and Afghanistan Deployment

Med Care. 2016 Dec;54(12):1078-1081. doi: 10.1097/MLR.0000000000000588.

Abstract

Background: Veterans of the wars in Iraq and Afghanistan who receive care in the Veterans Health Administration (VA) have high disease burden. Distinct comorbidity patterns have been shown to be differentially associated with adverse outcomes, including death. This study determined correlates of 5-year mortality.

Materials and methods: VA demographic, military, homelessness, and clinical measures informed this retrospective analysis. Previously constructed comorbidity classifications over 3 years of care were entered into a Cox proportional hazards model of death.

Results: There were 164,933 veterans in the cohort, including African Americans (16%), Hispanics (11%), and whites (65%). Most were in their 20s at baseline (60%); 12% were women; 4% had attempted suicide; 4% had been homeless. Having clustered disorders of pain, posttraumatic stress disorder, and traumatic brain injury was associated with death [hazard ratio (HR)=2.0]. Mental disorders including substance abuse were similarly associated (HR=2.1). Prior suicide attempt (HR=2.2) or drug overdose (HR=3.0) considerably increased risk of death over 5 years.

Conclusions: As congressional actions such as Veterans Choice Act offer more avenues to seek care outside of VA, coordination of care, and suicide prevention outreach for recent veterans may require innovative approaches to preserve life.

MeSH terms

  • Adolescent
  • Adult
  • Afghan Campaign 2001-*
  • Age Factors
  • Brain Injuries, Traumatic / mortality
  • Comorbidity
  • Female
  • Humans
  • Ill-Housed Persons / statistics & numerical data
  • Iraq War, 2003-2011*
  • Male
  • Mental Disorders / mortality
  • Middle Aged
  • Mortality*
  • Pain / mortality
  • Proportional Hazards Models
  • Retrospective Studies
  • Risk Factors
  • Sex Factors
  • Stress Disorders, Post-Traumatic / mortality
  • Suicide, Attempted / statistics & numerical data
  • Veterans / statistics & numerical data*
  • Young Adult