Yuxing Xia

Background: Physician burnout is increasingly recognized as a problem in physician well-being and may negatively affect patient care outcomes. Burnout can begin at any point of training or practice, potentially as early as the first year of medical school. Thus, there is a need to characterize possible burnout in medical students as the first step to optimizing strategies for mitigation. Traditionally, burnout has been studied using survey-based variables; however, identifying novel physiological and molecular biomarkers could allow for the expansion of screening and intervention strategies. Methods: In this pilot prospective cohort study, we followed a group of preclinical 1st and 2nd year medical students (n = 9) at the University of Florida over one academic year of medical school. We collected survey responses (Maslach Burnout Inventory [MBI], Patient Health Questionnaire-9 [PHQ-9], and Perceived Stress Scale [PSS]) and measured a panel of candidate physiological biomarkers of burnout (Inflammatory Cytokine Panel, Heart Rate Variability [HRV], and Leukocyte Telomere Length). Results: In the study participants, MBI composite scores and PHQ-9 scores showed a statistically significant increase over the course of an academic year, indicating higher levels of medical student burnout. Additionally, respondents reported a statistically significant decrease in time devoted to exercise, and we measured a significant increase in body mass index (BMI) during the academic year. PSS scores showed an upward trend which was not statistically significant. Likewise, average leukocyte telomere length trended downward, but the change was not statistically significant. There were no measured changes in the serum concentration of pro-inflammatory cytokines, and time-domain heart rate variability metrics did not differ significantly between timepoints. Conclusions: This pilot study supports the notion that burnout can begin early in medical school and is detectable via survey instruments in first-year and second-year medical students even with a small sample size. Additionally, leukocyte telomere length could potentially be a useful biomarker of burnout with supporting data, but we did not observe any statistically significant changes in inflammatory cytokines or heart rate variability. Further investigation into these potential biomarkers with larger cohort sizes is required to fully characterize their clinical utility.