We thank Dr. Lawson for his interest in our Research Report and would like to clarify several points of apparent misunderstanding. Dr. Lawson incorrectly notes that we reported residents “alleged to have made a harmful error scored 7.1 on depression . . . and those who did not scored 4.8.” In our analysis, the 20% of residents identified as screening positive for depression scored a 9 or greater on the Harvard Department of Psychiatry/National Depression Screening Day Scale (HANDS). The numbers Dr. Lawson is referring to were mean HANDS scores from a small subanalysis of residents who had 2 screens available; some of these residents were depressed and some were not.
In addition, the author suggests that the differences in the medical error rates we detected could be linked to the error collection methodology relying on resident self-report, which he proposes may lead to a falsely elevated number of reports from burned out or depressed residents. We agree that self-report has been a limitation of most prior studies. Our study, however, did not rely on self-report. We used a well-described 2-step surveillance process.1 First, a research nurse conducted a multipronged surveillance for possible medical errors that included daily reviews of medical records and orders, formal incident reports, solicited reports from nurses, and daily postshift surveys from residents. If a resident was primarily responsible for the suspected error, the research nurse noted the resident ID number. Second, 2 physician reviewers independently categorized each incident as an (1) adverse event, (2) nonharmful medical error, or (3) exclusion.1 Neither our research nurses nor our physician reviewers were aware of whether residents had screened positive for depression.
As Dr. Lawson notes, there are many potential contributors to medical errors. However, we believe he is wrong to dismiss depression, a factor associated with tripling a resident’s risk of making an error. Quite the contrary, we believe that our research substantiates the many emerging concerns about resident physician mental health, and the urgency with which we as a community should address them. Our study was not designed to discover why there is an association between depression and harmful medical errors, but further research will be important to evaluate this relationship. The next steps, if this relationship is further substantiated, would be to develop and rigorously test more effective approaches to mental health screening and treatment in the interest of patient safety and our resident physicians themselves.2
Sharon Calaman, MD
Professor of pediatrics and director of the pediatric residency program, Department of Pediatrics, Drexel University College of Medicine and St. Christopher’s Hospital for Children, Philadelphia, Pennsylvania; Sc493@drexel.edu.
Katherine A. Brunsberg, MD
Pediatric hospitalist, Children’s Hospitals and Clinics of Minnesota, Minneapolis and St. Paul, Minnesota, and adjunct assistant professor, Department of Pediatrics, University of Minnesota, Minneapolis, Minnesota.
Christopher P. Landrigan, MD, MPH
Professor of pediatrics, Harvard Medical School, chief of general pediatrics, Boston Children’s Hospital, and director of the sleep and patient safety program, Brigham and Women’s Hospital, Boston, Massachusetts.
1. Starmer AJ, Spector ND, Srivastava R, et al.; I-PASS Study Group. Changes in medical errors after implementation of a handoff program. N Engl J Med. 2014;371:1803–1812.
2. Goldman ML, Bernstein CA, Summers RF. Potential risks and benefits of mental health screening of physicians. JAMA. 2018;320:2527–2528.