Response to Letter to the Editor: The Contribution of Prescribed and Illicit Opioids to Fatal Overdoses in Massachusetts,2013-2015

We are grateful for the interest of Bhullar and Gilson from the Cuyahoga County Medical Examiner’s Office in our article, “The Contribution of Prescribed and Illicit Opioids to Fatal Overdoses in Massachusetts, 2013-2015.” ¹ They expressed 3 concerns that we would like to address.

First, they were concerned about the proportion of deaths that were excluded because of inadequate toxicology. As we described in the Methods section, toxicology reports were available for analysis through the Chapter 55 database from June 1, 2013, through December 31, 2015. We excluded 21% of deaths from analysis because the toxicology results available to us in the database did not specify an opioid. For those reports that we excluded, we do not know whether further detail was absent because further testing was not conducted or because details on further testing were not entered into the database. In Table 1 of our article, we did compare the demographic characteristics of decedents who were included and decedents who were excluded to determine significant differences. Early during the study period, toxicology testing for fentanyl was included at the discretion of the medical examiner’s office; later, during the study period, fentanyl was routinely included. Therefore, some samples earlier in the sampling window may not have been tested for fentanyl, and so our study may have underestimated the true prevalence of fentanyl-positive toxicology tests.

Second, the medical examiner’s office decided when a full autopsy was done. Full autopsies were completed in 68% of the cohort in our study.

Third, Bhullar and Gilson were concerned about our focus on decedents with active opioid prescriptions at the time of death, because the focus may have minimized the relative importance of opioid analgesics in the opioid crisis. We acknowledged in our article that most persons (58%) who died of opioid-related overdose during our study period had received an opioid prescription in the year before death. Our analysis was not designed to identify the role of previous opioid prescriptions in the development of opioid use disorder but, rather, to examine the opioids present in toxicology at the time of death. A major strength of our study was that we explicitly looked at the proportion of decedents with prescribed opioids at the time of death and whether or not those opioids were present in toxicology. We found that actively prescribed opioid prescriptions were often not detected in toxicology. Furthermore, toxicology tests showed that non-prescribed substances were present in >98% of opioid overdose decedents. These data support a focus on illicit opioids as the opioids most likely involved in the overdoses of Massachusetts decedents. Although opioid analgesic stewardship is a worthwhile focus to reduce unnecessary and risky opioid exposure, our data highlight the need to amplify harm-reduction services and the identification and treatment of persons with opioid use disorder to reduce the risk of dying from opioid overdose.