Health Economics / Outcomes Research

4D-04 - Development of a Rule-based Definition of Clinically Ready for Discharge Status in Hospitalized Patients with Complicated Urinary Tract Infection Treated with Intravenous Carbapenems

Monday, August 25, 2025

5:00 PM - 5:15 PM ET

Co-Author(s)

SK

Serena P. Koenig

Brigham and Women’s Hospital, Boston, MA, USA and Harvard Medical School, Boston, MA, USA, United States
JE

Jeffrey J. Ellis

US RWE/HOR, GSK, Collegeville, PA, USA, United States
DB

Douglas Boettner

US Medical Affairs, GSK, Collegeville, PA, USA, United States
LP

Lindsey Parker

US Medical Affairs, GSK, Collegeville, PA, USA, United States

Presenting Author(s)

RC

Rose Chang

Analysis Group, Inc., Boston, MA, USA, United States

Co-Author(s)

LY

Louise Yu

Analysis Group, Inc., Boston, MA, USA, United States
ER

Emily Reichert

Analysis Group, Inc., Boston, MA, USA, United States
JC

Joanne Chukwueke

Analysis Group, Inc., Boston, MA, USA, United States
ZC

Zheyi Cao

Analysis Group, Inc., Boston, MA, USA, United States
YH

Yichuan Grace Hsieh

Harvard Medical School, Boston, MA, USA and Research Information Science and Computing, Mass General Brigham, Somerville, MA, USA, United States
CH

Christopher Herrick

Research Information Science and Computing, Mass General Brigham, Somerville, MA, USA, United States
MD

Mei Sheng Duh

Analysis Group, Inc., Boston, MA, USA, United States
SM

Shawn Murphy

Harvard Medical School, Boston, MA, USA; Research Information Science and Computing, Mass General Brigham, Somerville, MA, USA; and Neurology, Massachusetts General Hospital, Boston, MA, USA, United States

Background: Patients hospitalized with complicated urinary tract infection (cUTI) caused by multidrug-resistant uropathogens often need intravenous (IV) carbapenems, requiring extended inpatient (IP) treatment or outpatient administration.

Objectives: To develop a clinically ready for discharge (CRFD) definition for patients hospitalized with cUTI and determine the difference in duration between CRFD and actual hospital discharge dates.

Methods: Electronic health record (EHR) data from the Mass General Brigham (MGB) Research Patient Data Registry (January 2019–June 2024) was used. Among patients aged ≥18 years who received IV carbapenem for cUTI with a positive urine culture, structured EHR and unstructured clinical notes for 20 randomly-selected patients were assessed. An initial rule-based CRFD definition, developed using clinical criteria from literature and expert input, was used to assess patient data on each IP day and finalized via an iterative process. Lenient (missing data indicates CRFD criterion met) and stringent (last observation of CRFD criterion carried forward) rules were used to impute missing criterion on a given IP day. An MGB infectious disease physician verified the date when CRFD status was achieved.

Results: Patients had a median (interquartile range [IQR]) age of 69 (63–76) years, 70% were female and 85% were White. The median (IQR) duration of IP stay was 6.5 (5.8–9.0) days. Among all patients, 89% (16/18) had extended-spectrum β-lactamase-producing Enterobacterales; 93% (14/15) and 63% (12/19) had isolates not-susceptible to fluoroquinolones and trimethoprim-sulfamethoxazole, respectively. Patients were CRFD if they met all criteria on an IP day: temperature < 101˚F; normal white blood cell count; no urinary tract infection-related pain, acute urinary symptoms, or cUTI-related complications; mobility, ability to eat/drink, physical and psychological stability. All patients had data for ≥1 criterion, with the average proportion of IP days of complete data ranging from 61% (cUTI-related complications) to 98% (temperature). Using a stringent definition, 11 (55%) patients were CRFD before being actually discharged, with a mean difference between the CRFD and actual discharge dates of 3 days (range: 1–6 days). Using the lenient definition, 18 (90%) patients were CRFD by a mean of 3 days (range: 1–9 days) before actual discharge.

Conclusions: We developed a clinically relevant, rule-based CRFD definition based on data from structured EHR and unstructured clinical notes. Our results suggest that >50% of patients with cUTI remained hospitalized longer than clinically necessary, possibly due to the IV carbapenem treatment, underscoring the unmet need for effective oral antibiotic therapy options.