Given the potential urgency of acting rapidly in the case of opioid overdoses, this study compared the performance of alternative predictive models with risk information from the past 3, 6, 9, and 12 months.
The study was composed of 1,014,033 Maryland residents ages 18–80 with at least one opioid prescription and no recorded death in 2015. Maryland’s 2015 prescription drug monitoring data were used to identify risk factors for non-fatal opioid overdoses from hospital discharge records and investigated fatal opioid overdoses from medical examiner data in 2016. Predictors derived from the drug monitoring program included demographics, payment sources for opioid prescriptions, count of unique opioid prescribers and pharmacies, and quantity and types of opioids and benzodiazepines filled. The study estimated a series of logistic regression models that included 3, 6, 9, and 12 months of prescription drug monitoring program data and compared model performance, using bootstrapped C-statistics and associated 95-percent confidence intervals. For hospital-treated nonfatal overdose, the C-statistic increased from 0.73 for a model including only the fourth quarter to 0.77 for a model with 4 quarters of data. For fatal overdose, the area under the curve increased from 0.80 to 0.83 over the same models. The strongest predictors of overdose were prescription fills for buprenorphine and Medicaid and Medicare as sources of payment. Models that predicted opioid overdose using 1 quarter of data were nearly as accurate as models using all 4 quarters. Models with a single quarter may be more timely and easier to identify persons at risk of an opioid overdose. (publisher abstract modified)