Depression among Emergency Medicine Residents over an Academic Year

Objectives: To describe the frequency of depression among emergency medicine (EM) residents by month, gender, rotation type, postgraduate year (PGY), and number of hours worked.
Methods: This was a prospective, nonblinded, cohort study of consenting EM residents in a four-year, 51-resident EM residency program from July 2003 to June 2004. Participants received an anonymous monthly survey via Web site that consisted of the Center for Epidemiologic Studies Depression Scale (CESD) and the resident's gender, PGY, number of hours worked in the previous week (≤40, 41–60, 61–80 and >80), and rotation type (EM, intensive care unit, non-EM clinical, or other). Residents were excluded from analysis if they did not complete at least one survey during each season. For each resident, the peak score for each three-month period was recorded and analyzed with a mixed-model analysis of variance to account for a repeated-measures effect.
Results: Fifty of 51 (98.0%) residents consented for participation. Nineteen (38%) were excluded because of incomplete data. The prevalence of depression was 12.1% (95% confidence interval 95% CI] = 7.5% to 19.0%; 15 of 124 scores). The women had numerically, but not statistically, significantly lower mean ± standard deviation CESD scores than the men (6.4 ± 6.8 vs. 8.7 ± 8.6, p = 1.0). There was no significant difference in mean CESD score by month, PGY, rotation type, or number of hours worked.
Conclusions: Season, number of hours worked, rotation type, PGY, and gender all failed to predict depression among EM residents in this single-center trial. The prevalence of depression was comparable to that of the general population.