CLINICAL PRACTICE Substance Use in Emergency Medicine Training Programs Brian D. McBeth, MD, Felix K. Ankel, MD, Louis J. Ling, MD, Brent R. Asplin, MD, MPH, Emily J. Mason, MD, Thomas J. Flottemesch, PhD, Robert M. McNamara, MD Abstract Objectives: To explore the prevalence of substance use among emergency medicine (EM) residents and compare to a prior study conducted in 1992. Methods: A voluntary, anonymous survey was distributed in February 2006 to EM residents nationally in the context of the national in-service examination. Data regarding 13 substances, demographics, and perceptions of personal patterns of substance use were collected. Results: A total of 133 of 134 residencies distributed the surveys (99%). The response rate was 56% of the total EM residents who took the in-service examination (2,397 / 4,281). The reported prevalence of most illicit drug use, including cocaine, heroin, amphetamines, and other opioids, among EM residents are low. Although residents reporting past marijuana use has declined (52.3% in 1992 to 45.0% in 2006; p < 0.001), past-year use (8.8%–11.8%; p < 0.001) and past-month use (2.5%–4.0%; p < 0.001) have increased. Alcohol use appears to be increasing, including an increase in reported daily drinkers from 3.3% to 4.9% (p < 0.001) and an increase in number of residents who indicate that their consumption of alcohol has increased during residency (from 4% to 12.6%; p < 0.001). Conclusions: Self-reported use of most street drugs remains uncommon among EM residents. Mari- juana and alcohol use, however, do appear to be increasing. Educators should be aware of these trends, and this may allow them to target resources for impaired and at-risk residents. ACADEMIC EMERGENCY MEDICINE 2008; 15:45–53 ª 2008 by the Society for Academic Emergency Medicine Keywords: physician impairment, graduate medical education P hysician impairment by substance use continues to be an issue for the health care system. 1–3 It has been identified as a source of medical errors, increased liability, and a detriment to patient safety, as well as a cause of personal and family breakdown for the affected physician. 1,4,5 Over the past few decades there has been a growing awareness of the importance of stu- dent and resident wellness and its effects on education and patient care. 6–8 There have been fundamental changes in graduate medical education, including institu- tion of the 80-hour work week and increasing require- ments for faculty supervision. Although numerous state medical societies and physician health programs have published their outcomes for treatment of physician sub- stance impairment and these results are generally favor- able, 9–15 resident impairment by substance use remains largely undiagnosed and untreated. McNamara and coworkers 16 published a large study of emergency medicine (EM) residents in 1994, utilizing an anonymous self-report survey administered follow- ing the American Board of Emergency Medicine’s (ABEM) annual in-service examination. While active use of most illicit drugs was uncommon, 7.6% of EM residents in this study had one positive response by CAGE screening for alcohol dependence, and an addi- tional 4.9% had two or more positive screening responses. Past-year use rates were 8.8% for marijuana, 1.0% for cocaine, and 1.5% for opioids. 16 ª 2008 by the Society for Academic Emergency Medicine ISSN 1069-6563 doi: 10.1111/j.1553-2712.2007.00008.x PII ISSN 1069-6563583 45 From the Department of Emergency Medicine, San Francisco General Hospital / University of California (BDM), San Fran- cisco, CA; the Department of Emergency Medicine, Regions Hospital (FKA, BRA, EJM, TJF), St. Paul, MN; the Department of Emergency Medicine, Hennepin County Medical Center (LJL), Minneapolis, MN; and the Department of Emergency Medicine, Temple University Medical Center (RMM), Philadel- phia, PA. Received July 13, 2006; revision received August 31, 2007; accepted September 2, 2007. Address for correspondence: Brian D. McBeth, MD; e-mail: bmcbeth@sfghed.ucsf.edu Presented at the ACEP Scientific Assembly, New Orleans, LA, October 2006. This study was funded by an internal grant from the Health- partners Research Foundation, Bloomington, MN. No conflicts of interest were identified.