Abstract
Background and Objective: Hospitalist practices around the country switch service on different days of the week. It is unclear whether switching clinical service later in the week is associated with an increase in length of stay (LOS). The aim of this study was to examine the association between service switch day for hospitalists at an academic medical center and LOS.
Methods: A single-center, cross-sectional study examined 4284 discharges from hospitalist staffed general internal medicine ward teams over a 1-year period between July 2018 and June 2019. Hospitalist service switch day changed from Tuesday to Thursday on January 1, 2019. The period between July 1, 2018, and December 31, 2018, was defined as the pre-switch time, while January 1, 2019, to June 30, 2019, was defined as the post-switch period. We calculated the LOS in days for patients discharged from hospitalist general internal medicine teams in the 2 periods. Generalized linear models were used to examine the association between attending switch day and LOS while adjusting for demographic factors, payer status, markers of severity of illness, and hospital or discharge-level confounders.
Results: There was no difference in mean LOS for patients discharged in the pre-switch time (6 days) period versus patients discharged in the post-switch time (6.03 days) (difference of means 0.03 days, 95% confidence interval -0.04 to 0.09, P value .37).
Conclusions: Change in attending switch day from earlier in the week to later in the week is not associated with an increase in LOS. Other factors such as group preference and institutional needs should drive service switch day selection for hospitalist groups.