Assessing geospatial and predictive relationships between drug overdoses and hepatitis A case rates in North Carolina, 2018-2022

Hepatitis A outbreaks in the United States have shifted in recent years from foodborne and waterborne transmission to sustained person-to-person spread, disproportionately affecting people who use drugs, people experiencing homelessness, and men who have sex with men. From 2016 through 2023, nearly 44,896 cases were reported nationally, representing a substantial increase over a historical baseline of 1,600 annual cases. In North Carolina (NC), 1,818 cases occurred between 2018 and 2022, with field investigations suggesting a potential role of methamphetamine use, though this relationship had not been systematically evaluated. Using statewide surveillance data, we aimed to: 1) quantify geospatial relationships between drug overdoses and hepatitis A case rates across NC, and 2) develop and evaluate predictive models using routinely collected population-level data to identify counties at increased risk for hepatitis A outbreaks. We integrated hepatitis A case data, non-fatal overdose data, fatal overdose data, vaccination data, and demographic and socioeconomic indicators. Spatiotemporal analyses revealed consistent spatial clustering between methamphetamine-related overdoses and hepatitis A case rates during the outbreak peak (Global Moran’s I: 0.034–0.258), while other substances showed weak or inverse relationships. Joint spatial clusters of elevated methamphetamine overdoses and hepatitis A cases expanded across elevated methamphetamine overdoses and hepatitis A cases expanded across central and western NC from 2019 through 2021, indicating co-occurring geographic risk. Predictive models using non-fatal methamphetamine overdoses alone showed strong performance (area under the receiver-operating characteristic curve: 0.80–0.86), with modest improvements when vaccination and socioeconomic variables were added. Predicted spatial patterns aligned with observed case distributions and accurately identified high-burden counties. These findings indicate that methamphetamine-related overdose patterns are strongly associated with hepatitis A case rates and may serve as an early indicator of outbreak risk. More broadly, this work highlights the value of integrating substance use surveillance with infectious disease monitoring to support timely, focused public health response. Embedding geospatial and predictive approaches into routine surveillance systems could improve early outbreak detection, guide resource allocation, and inform proactive vaccination and harm-reduction strategies.