Introduction The ongoing overdose crisis in the United States is increasingly characterized by polysubstance involvement, shifting beyond single-drug fatalities to complex combinations that present novel public health challenges. However, most epidemiological literature describing the US overdose crisis has still primarily relied on traditional data sources, especially the CDC WONDER system that is limited to assessing drugs with established ICD-10 codes using a maximum of n=2 ICD-10 codes. In this analysis we leverage complete medical examiner records Los Angeles County examine the limitations of CDC WONDER data for detecting polysubstance drug deaths.

Methods Deaths counts describing drug-involved deaths occurring in Los Angeles County were described using the CDC WONDER system, and medical examiner data from Los Angeles County. The average number of drugs, and the proportion of polysubstance deaths (involving 2 or more) involving 3 or more substances was summarized by year. Head-to-head comparisons were conducted of single-drug involvement, and UpSet plot visualization was employed for complex set analysis.

Results The average number of drugs present per drug-involved death increased from 1.72 in 2012 to a peak of 2.29 in 2023. The percentage of polysubstance deaths (2+ substances) that would not be fully characterizable using the CDC WONDER system given a 2 drug limit), increased from 51% in 2012 to a peak of 74% in 2023. The total number of unique polysubstance combinations with >2 substances increased from n=11 in 2012 to n=57 in 2023 and was n=39 in 2024. Overall concordance in single-substance death counts between CDC WONDER and medical examiner records was high among n=6 substances of interest. Across n=42 drug and year pairs observed between 2018 and 2024, the overall R2 according to a Pearson correlation analysis was 0.99.

Discussion Although concordance is very high for trends describing single-substance drug deaths, CDC WONDER fails to capture the majority of polysubstance deaths adequately. We find that a very large fraction of polysubstance deaths (about three quarters) are incompletely described using the 2 drug limit employed by CDC WONDER. This limitation has increased over time, as the complexity of polysubstance overdose deaths has increased. We illustrate a huge variety of polysubstance deaths that can be seen in medical examiner deaths. As the overdose crisis grows increasingly polysubstance in nature, improving the epidemiological tracking of deaths involving multiple drugs, and drugs not captured by ICD-10 codes currently, is of paramount importance. Here we highlight the gaps in existing traditional epidemiological sources which may serve as a motivating example for future improvements.

The authors have declared no competing interest.

This work was supported by the National Institute on Drug Abuse (R01DA057630). CLS received support from the National Institute on Drug Abuse (K01DA050771). JRF received funding from the National Institute on Drug Abuse (DA049644) and the National institute of Mental Health (MH101072). Funders played no role in data analysis or decision to publish.

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

Study protocols were approved by the University of California, Los Angeles IRB (IRB-22-1273).

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.

Yes

The CDC data used in this study are available at: https://wonder.cdc.gov/ The medical examiner data used in this study cannot be shared by the authors, but summary statistics may be shared by the authors upon reasonable request.

https://github.com/ROSLA-UCLA/ROSLA-HEAL-Data