Claire A Hoffmire1, Shannon K Barth1, Robert M Bossarte1. 1. U.S. Department of Veterans Affairs, Rocky Mountain Mental Illness, Research, Education and Clinical Center for Suicide Prevention, Aurora, Colorado (Hoffmire); Department of Physical Medicine and Rehabilitation, University of Colorado School of Medicine, Aurora (Hoffmire); U.S. Department of Veterans Affairs, Veterans Integrated Service Network 2 Center of Excellence for Suicide Prevention, Canandaigua, New York (Barth); Department of Epidemiology and Injury Control Research Center, West Virginia University School of Public Health, Morgantown (Barth, Bossarte).
Abstract
OBJECTIVE: The Department of Veterans Affairs and Department of Defense Mortality Data Repository (MDR) compiles National Death Index records for all veterans and military service members. This study aimed to compare MDR findings with those from a preexisting data source. METHODS: Veteran suicide rates estimated from death certificates were replicated with the MDR. Annual suicide rates were computed for veterans overall, by gender, and by Veterans Health Administration service utilization and compared with rates for adult nonveterans by using standardized mortality ratios (SMRs). RESULTS: Suicide rates and SMRs differed between the data sources. The 2010 MDR-derived veteran suicide rate was 27.4 deaths per 100,000 veterans, compared with an earlier estimate of 35.9. Differences were greater for females. Divergence was attributed to improved accuracy identifying veteran suicide decedents in the MDR. CONCLUSIONS: Conditions with low base rates can be major public health problems, and minor misclassification can substantially affect surveillance accuracy, prevention efforts, and the validity of study findings.
OBJECTIVE: The Department of Veterans Affairs and Department of Defense Mortality Data Repository (MDR) compiles National Death Index records for all veterans and military service members. This study aimed to compare MDR findings with those from a preexisting data source. METHODS: Veteran suicide rates estimated from death certificates were replicated with the MDR. Annual suicide rates were computed for veterans overall, by gender, and by Veterans Health Administration service utilization and compared with rates for adult nonveterans by using standardized mortality ratios (SMRs). RESULTS: Suicide rates and SMRs differed between the data sources. The 2010 MDR-derived veteran suicide rate was 27.4 deaths per 100,000 veterans, compared with an earlier estimate of 35.9. Differences were greater for females. Divergence was attributed to improved accuracy identifying veteran suicide decedents in the MDR. CONCLUSIONS: Conditions with low base rates can be major public health problems, and minor misclassification can substantially affect surveillance accuracy, prevention efforts, and the validity of study findings.
Authors: Frederic C Blow; Amy S B Bohnert; Mark A Ilgen; Rosalinda Ignacio; John F McCarthy; Marcia M Valenstein; Kerry L Knox Journal: Am J Public Health Date: 2012-03 Impact factor: 9.308
Authors: Matthew Miller; Deborah Azrael; Catherine Barber; Kenneth Mukamal; Elizabeth Lawler Journal: Am J Public Health Date: 2012-03 Impact factor: 9.308