Background:
Information is frequently lost during in‐hospital transitions of care. An effective and timely sign‐out between providers relaying critical information may reduce errors. However, little is known about provider preferences and practices (or sign‐out. We endeavored to determine the desired content and process for end‐of‐shift sign‐out.
Methods:
A 23‐question survey was e‐mailed once to practicing hospitalists via the Society of Hospital Medicine distribution list. Current format and content of sign‐out documents, nature of sign‐out episodes, and health information technology use were queried. Answers were rated on a Likert scale. For preferred sign‐out elements and modalities, 1 was rated “irrelevant,” 3 “desired,” and 5 “essential.” For provider practice in a sign‐out document, 1 was rated “never,” 3 “often,” and 5 “always.”
Results:
The response rate was 9% (503 of 5512). Both teaching (52%) and nonteaching (47%) hospitalists were represented. Eighty percent of hospitalists signed out 10–20 patients per episode, and 17% of patients were signed out 3 or more times in a 24‐hour period. Forty‐nine percent of respondents covered more than 40 patients overnight. Verbal (48%) and written (35%) communication were the most common methods of sign‐out. Additional modalities included IT‐supported programs (21%), billing systems (17%), e‐mail (10%), and personal digital assistants (3%). Thirty‐three percent of physicians used an electronic medical record for their sign‐out. “To‐do” lists (62%), code status (51%), plan of action for pending studies (44%), brief history of presenting illness (HPI; 43%), and anticipated change in clinical status (41%) were most frequently rated as “essential” components of preferred sign‐out. Brief HPI (41%) and to‐do lists (41%) were most frequently rated as “always” included in sign‐out practice. Significant discrepancies were reported between preferred sign‐out and routine practice for: code status (51% preferred vs. 32% practice), to‐do list (62% vs. 41%), list of pending studies (38% vs. 28%), plan of action for pending studies (44% vs. 28%), updated problem list (34% vs. 20%), and anticipated change in clinical status (41% vs. 25%). On a weekly basis, 47% and 69% of sign‐out documents contained inaccurate information or were incomplete, respectively, requiring chart review. Respondents reported that 18% of sign‐outs contained misinformation or were missing information that had a perceived negative impact on clinical outcome at least weekly.
Conclusions:
Hospitalist sign‐out often requires multiple episodes per patient, per day. There is heterogeneity in sign‐out modalities, with underutilization of verbal and written communication as well as information technology–based solutions. Sign‐out documents are frequently incomplete or inaccurate, contributing to perceived increases in work and negative clinical outcomes. Significant discrepancies exist between provider preferences and practice in sign‐out document content.
Author Disclosure:
Jeffrey E. Carter, none; Gregory J. Misky, none; Jeffrey J. Glasheen, none.