The role of medicolegal systems in producing geographic variation in suicide rates.

Objectives In this analysis, we ask whether there is systematic variation in the reporting of suicide by medicolegal system and if so whether this biases estimated effects of social correlates on suicide. Methods With cause of death records (1999-2002) and 2000 Census data, we use negative binomial regression to analyze the effects of medicolegal system on suicide and nonsuicide mortality aggregated at county of occurrence. Results We find that elected coroners have slightly lower official suicide rates than medical examiners (MEs; all of whom are appointed) and appointed coroners. In addition, we find that omitting medicolegal system does not bias estimates of the social determinants of suicide. Conclusion Contrary to arguments that MEs’ greater scientific training makes them more likely to underreport suicides, we conclude that appointed death investigators (MEs and appointed coroners) underreport suicide to a lesser degree than elected coroners, who are subject to greater public pressures that result in the misclassification of suicides.