Predictive patterns of suicidal behavior: The United States armed services versus the civilian population

Abstract

Suicide is the eighth leading cause of death in the United States and the third leading cause of death in the military population. The purpose of this study was to compare the suicide rates of the Army, Navy, Marine Corps, and Air Force to the U.S. population for the years 1980 through 1992. Multivariate linear regression analytic techniques were used to test the specific effects of year, service membership, and gender on suicide rates. A hierarchical linear regression equation resulted in R2 = 0.6737, a highly predictive finding (F112,118 = 13.600, p < 0.O00). Further controlled hypothesis testing suggested that each independent variable provided a unique contribution to the total variance accounted for in suicide rates. A follow- on set of time series regressions was used to assess the effects of gender and specific military service on suicide rates over the 13-year period. Findings revealed that suicide rates for civilian males are increasing, whereas civilian female rates are declining. No significant trends by year were shown among the services. Military suicide rates were consistently lower than civilian rates, and military female rates tended to be lower yet than military male rates.