Citation

Cutright P, Fernquist RM. Arch. Suicide Res. 2004; 8(3): 271-285.

Affiliation

Indiana University, Bloomington, Indiana, USA.

Copyright

(Copyright © 2004, International Academy of Suicide Research, Publisher Informa - Taylor and Francis Group)

DOI

10.1080/13811110490436936

PMID

16081393

Abstract

A multivariate model of 1996-1998 male and female suicide rates in the 50 states using Durkheim's societal integration concepts, spatial integration, and the culture of suicide was the focus of our research. Using multiple regression analysis, state divorce rates, church membership, approval of suicide, sex ratios, population density, and lagged 1981-82 suicide rates appeared to be statistically significant, which explained 90 percent of male and 79 percent of female suicide rate variation. Previous reports based on the hypothesis that western states have high suicide rates due to the high suicide rates of Native Americans residing in the West was rejected, as was the second hypothesis that the high western rates are due to higher western than non-western use of firearms in suicide. The usefulness of macro-level predictors of suicide rates was discussed and we concluded that macro-level analysis was needed to explain trends in suicide, to understand differences in suicide levels among states and nations, as well as differences in the age structures of suicide among populations.