• Peter Geissert, Sara Hallvik, Joshua Van Otterloo, Nicole OʼKane, Lindsey Alley, Jody Carson, Gillian Leichtling, Christi Hildebran, Wayne Wakeland, and Richard A Deyo.
• Department of Systems Science, Portland State University, Portland, OR, USA.
• Pain. 2018 Jan 1; 159 (1): 150-156.

AbstractTo develop a simple, valid model to identify patients at high risk of opioid overdose-related hospitalization and mortality, Oregon prescription drug monitoring program, Vital Records, and Hospital Discharge data were linked to estimate 2 logistic models; a first model that included a broad range of risk factors from the literature and a second simplified model. Receiver operating characteristic curves, sensitivity, and specificity of the models were analyzed. Variables retained in the final model were categories such as older than 35 years, number of prescribers, number of pharmacies, and prescriptions for long-acting opioids, benzodiazepines or sedatives, or carisoprodol. The ability of the model to discriminate between patients who did and did not overdose was reasonably good (area under the receiver operating characteristic curve = 0.82, Nagelkerke R = 0.11). The positive predictive value of the model was low. Computationally simple models can identify high-risk patients based on prescription history alone, but improvement of the predictive value of models may require information from outside the prescription drug monitoring program. Patient or prescription features that predict opioid overdose may differ from those that predict diversion.

13 keywords...

hide…