Variability of undetermined manner of death classification in the US.

OBJECTIVES To better understand variations in classification of deaths of undetermined intent among states in the National Violent Death Reporting System (NVDRS). DESIGN Data from the NVDRS and the National Vital Statistics System were used to compare differences among states. MAIN OUTCOME MEASURES Percentages of deaths assigned undetermined intent, rates of deaths of undetermined intent, rates of fatal poisonings broken down by cause of death, composition of poison types within the undetermined-intent classification. RESULTS Three states within NVDRS (Maryland, Massachusetts, and Rhode Island) evidenced increased numbers of deaths of undetermined intent. These same states exhibited high rates of undetermined death and, more specifically, high rates of undetermined poisoning deaths. Further, these three states evidenced correspondingly lower rates of unintentional poisonings. The types of undetermined poisonings present in these states, but not present in other states, are typically the result of a combination of recreational drugs, alcohol, or prescription drugs. CONCLUSIONS The differing classification among states of many poisoning deaths has implications for the analysis of undetermined deaths within the NVDRS and for the examination of possible/probable suicides contained within the undetermined- or accidental-intent classifications. The NVDRS does not collect information on unintentional poisonings, so in most states data are not collected on these possible/probable suicides. The authors believe this is an opportunity missed to understand the full range of self-harm deaths in the greater detail provided by the NVDRS system. They advocate a broader interpretation of suicide to include the full continuum of deaths resulting from self-harm.