BACKGROUND/RATIONALE:
Reducing suicide is a national and a VA priority. VA patients in depression treatment are at high risk for suicide. To develop effective suicide prevention programs, VA policymakers and clinicians need basic information about absolute suicide rates among these patients and information about suicide risk factors within the depression treatment population.

OBJECTIVE(S):
Our primary objectives were to determine overall suicide rates among the VA depression treatment population and patient characteristics associated with higher suicide rates. Additional goals included determining suicide rates among important subgroups of depressed patients and suicide rates during potentially high-risk treatment periods. An exploratory goal was to assess the level of monitoring provided to depressed patients during high-risk periods and to examine whether higher levels of monitoring were associated with reduced suicide risks.

METHODS:
We conducted a retrospective cohort study using data from a comprehensive sample of VA patients receiving depression treatment. We described suicide rates, patient risk factors for suicide, levels of clinical monitoring during high-risk treatment
periods and relationships between patient risk factors, monitoring, and suicide risks. Our primary outcome was suicide as documented in National Death Index data. We calculated unadjusted suicide rates and median time to suicide since the date of the initial depression diagnosis, by age group, gender, race/ethnicity, and specific psychiatric and substance abuse comorbidities. We used Cox’s proportional hazards regression models to assess the relationships between suicide and the exposures of interest, adjusting for other covariates. In exploratory analyses, we used propensity scoring and instrumental variable techniques to examine the relationship between levels of clinical monitoring and suicide, while accounting for likely treatment selection.

FINDINGS/RESULTS:
The overall suicide rate for VA patients in depression treatment (excluding individuals with unknown race) was 88/100,000 person-years during the observation period; cohort patients were followed for an average of 2.5 years during the study period.
Suicide rates per 100,000 person-years were higher for men (90) than for women (29) and for whites (95) than for African Americans (27). Veterans of Hispanic origin had lower suicide rates (46) than those not of Hispanic origin (87). Patient aged 18-44 years had a suicide rate of 95 suicides per 100,000 person-years, compared with 78 for patients aged 45-64 and 90 for patients aged 65 and older. Concurrent PTSD and service connection were associated with lower suicide rates. In the first 60 weeks following antidepressant starts, dose changes, and psychiatric hospitalizations, the suicide rate was 114/100,000 person-years. Suicide rates were 568/100,000 p-y in the first 12 weeks following psychiatric hospitalizations and 210/100,000 p-y following new antidepressant starts. Suicide rates remained above the base rate for 48 weeks following inpatient discharge and 12 weeks following antidepressant treatment events. Adults aged 61-80 years were at highest risk in these 12-week periods. (Further results addressing other specific aims and results from additional project related studies are reported in the text and the attached articles/manuscripts.)

IMPACT:
This project was the largest and most comprehensive examination of suicide among patients receiving VA treatment for depression. Suicide rates were documented for the entire VA depression treatment population and for important subgroups of these patients. We also documented suicide rates during high-risk treatment periods and the monitoring offered to patients during these high-risk periods, overall and by age, ethnicity, and psychiatric status. Finally, we examined the relationship between monitoring and suicide. Supplemental methodological and content relevant studies were completed. We believe study results will inform the field about these critical issues and will have practical implications for VA clinicians treating depressed patients and for VA administrators organizing preventive services for these patients.

---

External Links for this Project

---

PUBLICATIONS:

None at this time.

DRA: Military and Environmental Exposures, Mental, Cognitive and Behavioral Disorders
DRE: none
Keywords: Depression, PTSD
MeSH Terms: none