Background: The optimal duration of hospitalist clinical blocks is unknown. Longer block lengths may offer greater continuity and discontinuity between providers has been adversely associated with 30-day mortality and readmissions. However, extending days worked may lead to fatigue and negatively impact hospitalist satisfaction or contribute to burnout. We conducted a randomized trial comparing two different block lengths, measuring both patient outcomes and provider experience.

Methods: This study was conducted in a single academic medical center staffed by ~95 hospitalists who work on various teaching and non-teaching services, including a direct-care general medicine service. Solo hospitalist physicians on the direct-care service manage patients admitted to general acute care and transitional care units. At the start of the study, we aimed to schedule half of each hospitalist’s rotations to a “5-day block” model and half to a “7-day block” model. Admissions from the previous evening (e.g., holdovers) are randomly distributed to teams every morning; for this study, we studied only patients assigned to the 6 direct-care hospitalist teams. Due to the nature of the study, block assignments could not be blinded. We collected the following patient outcomes: length of stay, discharge by noon, 30-day mortality, and 30-day readmissions. We also administered a 2-question survey at the end of each clinical block to assess hospitalist satisfaction and burnout. Differences between 5 vs. 7-day block length on outcomes were assessed using logistic regression for patient outcomes and t-tests for provider experience.

Results: The trial was conducted between 7/11/2020 – 7/2/2021. A total of 1100 patients were managed on the 5-day block model and 1090 patients on the 7-day model, with no significant differences in baseline patient characteristics in the two groups. The average number of patients per team was 8.5. We found no differences in patient outcomes between the 5-day and 7-day models, including length of stay, discharge before noon, 30-day readmission, and 30-day mortality (Table 1). For the assessment of provider experience, hospitalists completed 274 of 380 end-of-block surveys (72% completion rate). Analysis was done for all responses, which included providers who did partial blocks (such as in situations where hospitalists found replacement coverage), as well as for those that completed the full 5 or 7-day blocks. For providers that completed either a full 5 or 7-day block, mean satisfaction (scale 1-10), did not differ between 5-day blocks and 7-day blocks (7.57 vs 7.35, respectively p=0.33). However, mean physician burnout (scale 1-5) was significantly higher with 7-day blocks than 5-day blocks (2.37 vs 2.15, respectively p=0.02) (Table 2).

Conclusions: A 5 vs. 7-day work schedule for hospitalists on a direct care service at an academic medical center had no impact on patient outcomes or hospitalist satisfaction, but was associated with higher physician burnout. While the optimal block length is unknown, shorter block lengths may mitigate provider burnout.

IMAGE 1: Table 1: Patient Outcomes for 5 vs. 7 Day Block Models

IMAGE 2: Table 2: Provider Experience for 5 vs. 7 Day Block Models