Background:

Resident duty hour restrictions have resulted in an increase in the number of new overnight admissions that are transferred to day teams. These “holdovers” make up about 40% of all admissions to our teaching medicine service, however, little is known about the efficiency, safety and educational objectives of holdover signout. Our goal is to describe baseline characteristics of holdover signouts in an internal medicine residency program.

Methods:

We performed an observational cross sectional study using a convenience sample of morning holdover signouts between overnight admitting residents and receiving day teams on our inpatient medicine service. Receiving day teams consisted of medical students, interns, a resident, and an attending. Signout included the overnight resident presentation, closing the loop by the day team, and any teaching or constructive feedback by team members. Closing the loop consisted of after-presentation questions, clarifications, and to-dos, while constructive feedback included specific statements designed to acknowledge or improve overnight management.

We audited the duration of each holdover signout, and the frequency of signouts with closing the loop, teaching moments, and immediate constructive feedback. A subset of holdover observations also measured the frequency of presentations with interruptions by team members and the duration of the overnight resident presentation, including patient data (history of present illness, review of systems, physical exam, social/family history, medications and laboratory data) and assessment and plan. We summarized the data using descriptive statistics.

Results:

We observed 61 holdover signouts performed by 47 residents, which are detailed in Table 1. The median holdover signout duration was 14.3 minutes. The receiving team closed the loop for all observed signouts. Teaching moments from team members occurred at a rate of 32.8% while immediate constructive feedback did not occur.

The subset data consisted of 34 holdovers signouts, with the median resident presentation lasting 13.4 minutes. The patient data portion, all of which can be found in our electronic medical record (EMR), made up a median 62.0% of the presentation time. Three presentations had interruptions by team members.

Conclusions:

This is the first known reported data on holdover signout. Holdover signout is a longer process than most other signout and patient data available in an EMR may not need to be a substantial portion of holdover presentations. In addition, overnight admissions are a source of missed teaching and feedback opportunities for overnight residents. These findings suggest there may be a role for creating standard best practices to improve efficiency and education during the holdover signout process.