Purpose: Understanding the association between attending physicians' workload and teaching is critical to preserving residents' learning experience. The authors tested the association between attending physicians' self-reported workload and perceptions of time for teaching before and after the 2003 resident duty hours regulations.
Method: From 2001 to 2008, the authors surveyed all inpatient general medicine attending physicians at a teaching hospital. To measure workload, they used a conceptual framework to create a composite score from six domains (mental demand, physical demand, temporal demand, effort, performance, frustration). They measured time for teaching using (1) open-ended responses to hours per week spent doing didactic teaching and (2) responses (agree, strongly agree) to the statement "I had enough time for teaching." They conducted multivariate logistic regression analyses, controlling for month, year, and clustering by attending physicians, to test the association between workload scores and time for teaching.
Results: Of 738 eligible attending physicians, 482 (65%) completed surveys. Respondents spent a median of three hours per week dedicated to teaching. Less than half (198; 43%) reporting enough time for teaching. The composite workload scores were normally distributed (median score of 15) and demonstrated a weak positive correlation with actual patient volume (r = 0.25). The odds of an attending physician reporting enough time for teaching declined by 21% for each point increase in composite workload score (odds ratio = 0.79 [95% confidence interval 0.69-0.91]; P = .001).
Conclusions: The authors found that attending physicians' greater self-perceived workload was associated with decreased time for teaching.