The purpose of this study was to investigate the association between disability and one year mortality in more than four million veterans receiving compensation for service-connected conditions, which are illnesses or injuries incurred or aggravated during military service. To do this, we used administrative data from the Department of Veterans Affairs (VA). The VETSNET file contained a snapshot of demographic, military service, and disability characteristics of veterans receiving compensation in October 2016 and the vital status file had the date of death or last follow-up as of September 2017. The overall disability rating, ranging from 10% to 100%, was the major independent variable, and mortality at 1 year was the dependent variable. We used standard bivariate and multivariate statistical methods and found that mortality was highest in the 100% rating group. This study involved no direct contact with human subjects and required skills in data management and data analysis as well as familiarity with the VA computing environment. The major challenge was to understand the limitations of the VETSNET file; this understanding was acquired by careful consideration of documentation as well as exploration of study data. The major limitations of the file were the lack of an award date for compensation as well as no indicator of combat exposure. The first limitation resulted in a very limited exposure time of 1 year. Combat exposure is likely an important predictor of the overall disability rating as well as a predictor of mortality. The importance of understanding one’s data cannot be overemphasized, particularly when that data are collected for purposes other than research.