OBJECTIVES : This study aims to identify sociodemographic and clinical risk factors related to an opioid-related overdose event among patients enrolled in Mississippi Medicaid.

METHODS : A matched case-control study was conducted using linked data from Mississippi Medicaid administrative claims and the Mississippi Prescription Drug Monitoring Program (PDMP) database. Cases were identified as Medicaid beneficiaries who had an overdose event between July 1, 2014-December 31, 2015. The overdose event date was the index date. Every case was matched to 3 control beneficiaries, who did not have an overdose event, on age and sex. The corresponding index date was assigned to the controls. Various sociodemographic, clinical, and opioid use-related characteristics were assessed in the 6-month period prior to the index date. A conditional logistic regression model was conducted in SAS to identify the risk factors associated with overdose.

RESULTS : A total of 341 cases and 1023 controls were included in the study. Overdose cases included a higher proportion of beneficiaries who were female, Caucasian, ages 21-44 years and diagnoses of depression, opioid or other substance dependence. Also correlated was the use of a single prescriber and a single pharmacy, chronic opioid use, paying cash for prescriptions and concomitant sedative use. Logistic regression identified major predictors of overdose to be beneficiaries who had opioid dependence (Odds ratio [OR] 9.9, 95% CI 3.57-27.80), other dependence (OR 4.4, 95% CI 2.20-8.91), depression (OR 4.8, 95% CI 2.87-8.29), and those who paid in cash before reaching their monthly prescription quota (OR 2.8, 95% CI 1.86-4.14). Beneficiaries who had multiple prescribers (OR 0.1, 95% CI 0.04-0.26) and African Americans (OR 0.3, 95% CI 0.21-0.45) had lower odds of an overdose.

CONCLUSIONS : Results from this study can help in the identification of individuals at risk of opioid-related overdose events for implementation of targeted preventative interventions.