The role of prescribed controlled substance acquisition as potential triggers of opioid overdose: A case-crossover study

Background: The role of prescribed opioids and benzodiazepines as risk factors for opioid overdose are well established, however, their role as potential ‘triggers’ of opioid overdose has not been formally investigated. Objective: The objective of this study was to evaluate the temporal relationship between controlled substance acquisition and opioid overdose utilizing a case-crossover design. Methods: This study utilized Arkansas statewide data between 2014 and 2020. Prescription Drug Monitoring Program (PDMP) data were used to assess controlled substance acquisition and fatal and non-fatal opioid overdose were assessed using linked death certificate, inpatient discharge, and emergency department (ED) data. All persons residing in Arkansas who experienced an opioid overdose or had ≥ 1 Arkansas PDMP prescription fill(s) were included. Controlled substance characteristics were described in the 7 days prior to overdose and compared to the controlled substance characteristics in 11 weekly (7-day) control windows prior to overdose. Binary controlled substance variables indicating presence or absence of: any controlled substance, opioid, benzodiazepine, stimulant, sedative, carisoprodol, opioid and benzodiazepine, and opioid and benzodiazepine and carisoprodol were created. Additionally, total morphine milliequivalents were calculated for each time window. Conditional logistic regression models were estimated and adjusted odds ratios for each controlled substance characteristic after accounting for other controlled substance, and prior overdose, and clinical characteristics derived from ED and inpatient data are reported. Results: A total of 2,818,135 individuals with ≥1 Arkansas PDMP record(s) (45.10 % male; 39.94 mean age) were included, of which 28,670 (1.02 %) experienced ≥1 opioid overdose. There was a significant association between opioid overdose and the acquisition of a controlled substance (OR=1.785; p < 0.001), opioid (OR=1.992; p < 0.001), benzodiazepine (OR=1.379; p < 0.001), carisoprodol (OR=1.744; p < 0.001), opioid and benzodiazepine (OR=2.203; p < 0.001), and opioid and benzodiazepine and carisoprodol (OR=2.503; p < 0.001), in the 7 days prior to an opioid overdose event. Conclusion: Controlled substance prescription acquisition, particularly opioids in combination with carisoprodol and/or benzodiazepines, are potential triggers of opioid overdose.