![[ Visit Client Website ]](images/banner.jpg)
BACKGROUND: The Kansas Injury Prevention Program (KIPP) and other states have been working collaboratively to examine the specificity of coding in drug poisoning deaths. Data comparing the specificity of drug related poisoning deaths by state were highlighted at the 2013 CSTE conference and showed that Kansas reported a disproportionately large number of non-specific drug deaths. Extrapolation of national ratios estimated that only half of the actual opioid analgesics overdose deaths in Kansas were being reported. In light of these findings, KIPP and the Office of Vital Statistics (OVS) at the Kansas Department of Health and Environment recognized the need to investigate the underlying reasons for non-specific drug coding on Kansas death certificates and implement strategies to improve data quality.
METHODS: After high rates of non-specific drug coding were brought to their attention, OVS staff identified three coroners who certified deaths for a large proportion of the state as likely sources of non-specific coding. The OVS proceeded to convene coroners to identify barriers and provide guidance for improving specificity of drug poisonings on death certificates. OVS and KIPP staff then collaborated to assess whether or not OVS’ provision of guidance (intervention) improved data specificity over time among a subset of records specific to the three identified coroners that included all death records 12 months prior to the intervention and 12 months post intervention.
RESULTS: Preliminary analyses revealed no significant improvement in data specificity after OVS’ provision of guidance to coroners. Discussions between KIPP and OVS staff revealed important limitations with the guidance provided. OVS staff conveyed that some coroners believed that there were not enough fields to input all the drugs in the event of a poisoning due to multiple drugs; this led to indicating no specific drugs at all. OVS staff advised coroners to indicate “multi-drug toxicity” in these instances. Using “multi-drug toxicity” was not sufficient to meet KIPP’s goal of improving specificity of drug related poisoning deaths and prompted KIPP and OVS staff to implement a quality improvement process to systematically assess and enhance data specificity for drug related deaths in Kansas.
CONCLUSIONS: This ongoing intra-agency partnership illustrates the necessity of engaging data collectors, data stewards, analysts and program staff in efforts to improve data quality to ensure that technical assistance provided to coroners aligns with the needs of the injury program.