A Multiple Baseline Trial of an Electronic ICU Discharge Summary Tool for Improving Quality of Care

 

by Stelfox, Henry T.; Brundin-Mather, Rebecca; Soo, Andrea; Whalen-Browne, Liam; Kashyap, Devika; Sauro, Khara M.; Bagshaw, Sean M.; Fiest, Kirsten M.; Taljaard, Monica; Parsons Leigh, Jeanna 

 

Critical Care Medicine: August 16, 2022 - Volume - Issue - 10.1097

 

Objective: 

Effective communication between clinicians is essential for seamless discharge of patients between care settings. Yet, discharge summaries are commonly not available and incomplete. We implemented and evaluated a structured electronic health record–embedded electronic discharge (eDischarge) summary tool for patients discharged from the ICU to a hospital ward.

Design: 

Multiple baseline trial with randomized and staggered implementation.

Setting: 

Adult medical-surgical ICUs at four acute care hospitals serving a single Canadian city.

Patients: 

Health records of patients 18 years old or older, in the ICU 24 hours or longer, and discharged from the ICU to an in-hospital patient ward between February 12, 2018, and June 30, 2019.

Intervention: 

A structured electronic note (ICU eDischarge tool) with predefined fields (e.g., diagnosis) embedded in the hospital-wide electronic health information system.

MEASUREMENTS AND MAIN RESULTS 

: We compared the percent of timely (available at discharge) and complete (included goals of care designation, diagnosis, list of active issues, active medications) discharge summaries pre and post implementation using mixed effects logistic regression models. After implementing the ICU eDischarge tool, there was an immediate and sustained increase in the proportion of patients discharged from ICU with timely and complete discharge summaries from 10.8% (preimplementation period) to 71.1% (postimplementation period) (adjusted odds ratio, 32.43; 95% CI, 18.22–57.73). No significant changes were observed in rapid response activation, cardiopulmonary arrest, death in hospital, ICU readmission, and hospital length of stay following ICU discharge. Preventable (60.1 vs 5.7 per 1,000 d; p = 0.023), but not nonpreventable (27.3 vs 40.2 per 1,000d; p = 0.54), adverse events decreased post implementation. Clinicians perceived the eDischarge tool to produce a higher quality discharge process.

Conclusions: 

Implementation of an electronic tool was associated with more timely and complete discharge summaries for patients discharged from the ICU to a hospital ward.