Fatal Drug Overdoses Affect Health Care Workers in Large Numbers

Registered nurses, social workers and other behavioral health workers, as well as those in health care support are at significantly greater risk for drug overdose death compared to non-healthcare workers, according to a soon-to-be released study led by researchers at Columbia University Mailman School of Public Health and Columbia University Irving Medical Center. Despite the unparalleled increase in recent U.S. drug overdose deaths, until now the risks faced by health care workers were not known. The results are published online in Annals of Internal Medicine.

Using a U.S. nationally representative cohort from the 2008 American Community Survey (ACS) followed for mortality through 2019, the researchers compared the hazards of drug overdose death between health care workers and non–health care workers.  To estimate risks for drug overdose death among health care workers relative to non–health care workers, the researchers analyzed the data to determine drug overdose deaths for six health care worker groups including physicians, registered nurses, other diagnosing or treating health care workers, health technicians, health care support workers, and social or behavioral health workers versus non-health care workers.

After accounting for age and sex differences among the groups, social workers and other behavioral health workers were 112 percent more likely, health care support workers were 100 percent more likely, and registered nurses were 51 percent more likely than non–health care workers to die of a drug overdose. Results were generally similar for opioid-related overdose deaths and unintentional overdose deaths. For health care workers overall, 84.5 percent of drug overdose deaths were considered opioid-related and 76.0 percent unintentional. For non–health care workers, 76 percent were opioid-related and 85 percent were unintentional.

“Several factors could place health care workers at increased risk for drug overdose death,” said Mark Olfson, MD, MPH, professor of Epidemiology at Columbia Public Health and professor of Psychiatry, Vagelos College of Physicians & Surgeons. “Those who prescribe or administer medications have ready access to opioids and other controlled prescription drugs and also frequently have job stress and occupational burnout—all of which have been associated with increased risk for opioid use disorder, which in turn can increase the risk for overdose. Many health care workers also routinely do physically strenuous tasks that can lead to injuries that can result in opioid prescriptions to manage pain and attendant risks.”

Despite the increased risks of health care worker groups, there is a shortage of programs dedicated to treating health care workers with substance use problems.

“Although recent progress has been made in developing specialized programs to provide substance use treatment for physicians, few programs exist for other health care professionals with substance use problems who were found in this study to be at increased risk,” observed Olfson, who is also a research psychiatrist at New York State Psychiatric Institute. “More generally, these new findings draw attention to the importance of identifying health care workers who may be struggling with substance use problems, helping them to access confidential and effective treatment. We further urge that subsequent research studies consider deeper, more granular details, including the specific nature of work within each occupational group that could contribute to overdose risk.”

Co-authors are Candace M. Cosgrove, U.S. Census Bureau, Mortality Research Group; Melanie M. Wall, Columbia Mailman School of Public Health and Vagelos College of Physicians and Surgeons, Columbia University; and Carlos Blanco, National Institute on Drug Abuse, part of the National Institutes of Health.

Regarding Dr. Cosgrove’s contribution, this material should not be interpreted as representing the viewpoint of the U.S. Census Bureau. Regarding Dr. Blanco’s contribution, this material should not be interpreted as representing the viewpoint of the U.S. Department of Health and Human Services, the National Institutes of Health, or the National Institute on Drug Abuse.

The study was supported by a National Heart, Lung, and Blood Institute interagency agreement with U.S. Census Bureau.