Individual-level data on suicide deaths were obtained from the Multiple Cause of Death files for 1990–2010, collected by the National Center for Health Statistics. Customized files including geographic data were obtained through the National Association for Public Health Statistics and Information Systems (NAPHSIS). From the complete set of death records, we selected observations for which suicide was either the underlying cause or among the contributing causes of death, using codes from the International Classification of Disease, versions 9 and 10 (codes E950-E959 and X60-X84, Y87, respectively). These records were combined with data on the living population obtained from the annual American Community Survey (ACS) for the years 2001–2010. For living population data prior to 2001 (when the ACS was initiated), we used data from 1% samples of the 1990 and 2000 Census. In order to estimate data for years 1991 through 1999, we used a linear interpolation procedure described elsewhere (Grucza et al., 2014, 2012). Briefly, this was done by determining the weights for records representing each possible combination of covariate parameters in each Census data set (i.e., each combination
