Increases from 2002 to 2015 in prescription opioid overdose deaths in combination with other substances

Background: Prescription opioid (PO) overdose deaths increased sharply over the last decade. Changes in PO deaths in combination with other psychoactive substances may provide a partial explanation. Methods: PO deaths from the National Multiple-Cause-of-Death Files for 2002-03 (N = 15,973) and 2014-15 (N = 41,491) were analyzed. We calculated (1) changes in proportions of deaths in combination with benzo-diazepines, antidepressants, heroin, alcohol, cocaine between the two periods, and (2) proportions of increase in deaths attributable to each substance among PO and synthetic opioids other than methadone (SO-M) deaths, by age, gender, race/ethnicity. Results: Between 2002-03 and 2014-15, PO deaths increased 2.6 times; SO-M deaths 5.6 times, especially for ages 18-34, males, African-Americans. For PO deaths, most frequent combinations at both periods were with benzodiazepines; for SO-M, benzodiazepines, antidepressants in 2002-03, heroin, benzodiazepines in 2014-15. The largest increases occurred in combination with heroin among all PO (4.6% to 15.4%, change ratio = 3.3[95%CI = 3.1-3.6]), but especially SO-M deaths (1.2% to 24.5%, change ratio = 21.3[95%CI = 15.0-30.3]). Deaths involving cocaine decreased among PO, increased among SO-M deaths. One-fifth of increased PO or SO-M deaths were attributable to any of the five substances. Increased PO deaths were equally attributable to benzodiazepines and heroin; deaths attributable to heroin were higher among ages 18-49, males, and non-Hispanic whites. Increased SO-M deaths were attributable mostly to heroin among all groups. Conclusions: Increased PO overdose deaths over the last decade may be partially explained by increased deaths in combination with other psychoactive substances. Use of other substances should be considered in efforts toward reducing prescription opioid overdoses.