Factors associated with incomplete toxicology reporting in drug overdose deaths, 2010-2016

Classification of overdose deaths is often geographically and demographically inconsistent. Incomplete surveillance records may distort estimates of drug overdose rates across time and place. We examined incomplete toxicology reporting among drug overdose decedents by demographic and geographic characteristics, measuring changes in missingness rates and their associations with decedent characteristics over time.

We estimated the percentage of overdose deaths reported in the National Vital Statistics System with missing toxicology results from 2010 to 2016, overall and by decedents' demographic and geographic characteristics. Multilevel logistic regression models evaluated prevalence of missingness by decedent characteristics, accounting for geographic clustering.

Overall, 20.3% of death certificates did not indicate a specific drug, declining from 24.4% in 2010 to 14.6% in 2016. Deaths were less likely to have missing information if they occurred in counties with medical examiners versus coroners. Female decedents were more likely to have missing information than males, as were non-Hispanic whites compared with Hispanics and non-Hispanic blacks.

The percentage of deaths with missing toxicology information declined over time, but demographic and geographic differences in missingness persist. This yields detection biases that skew temporal trends and understanding of groups impacted by the opioid epidemic.