Richard Drew, Ethan Brenneman, Jason Funaro, Hui-Jie Lee, Michael Yarrington, Kristen Dicks, David Gallagher
Abstract
Outpatient Parenteral Antibiotic Therapy (OPAT) provides coordinated services to deliver parenteral antibiotics outside of the acute care setting. However, the reduction in monitoring and supervision may impact the risks of readmission to the hospital. While identifying those at greatest risk of hospital readmission through use of computer decision support systems could aid in its prevention, validation of such tools in this patient population is lacking.
Introduction
Outpatient Antibiotic Therapy (OPAT) is a structured program that coordinates the administration of parenteral antibiotics outside of the acute care setting. The OPAT program at our institution was established in 2016, and has grown to include a core group of pharmacists, nurses, and infectious diseases providers. OPAT nurses and pharmacists review weekly laboratory work and adjust medications. OPAT patients are seen by infectious diseases providers during their OPAT course.
Materials and method
This retrospective cohort study (reviewed and exempted by the Duke University Hospital Institutional Review Board) was a secondary analysis of data obtained for the evaluation of a published predictive tool for the determination of risk for unplanned hospital readmission in OPAT patients within our academic, tertiary care hospital system [6]. Patients were included if they were Duke University Health System (DUHS) inpatients ≥ 18 years of age enrolled in the DUHS OPAT program (OPAT care initiated inpatient and coordinated by the DUHS OPAT program) discharged to a home or skilled nursing facility between July 1, 2019 –February 1, 2020 and with a planned duration of OPAT at hospital discharge ≥ 7 days, and with an Epic readmission score available during the index hospitalization. Patient populations not routinely cared for by the OPAT service (patients with solid organ and hematopoietic stem cell transplants patients, and those with cystic fibrosis, or left ventricular assist device (LVAD), receiving OPAT administered in a long-term acute care facility (LTAC), or ongoing renal replacement therapy) and those lost to follow-up were excluded.
Results
From the 606 distinct OPAT encounters that were identified, 115 episodes were excluded for not meeting eligibility criteria. After randomization and removal of an additional 24 episodes for multiple OPAT encounters with the same patient, 467 unique encounters (representing 467 unique patients) were included in the analysis.
Discussion
In contrast to previous reports evaluating the utility of the EPIC Unplanned Readmission Model 1 in predicting 30-day unplanned hospital readmissions conducted within DUHS on a variety of clinical service cohorts, we found poor discrimination ability among models utilizing maximum and discharge Epic readmission scores in predicting both all-cause and OPAT-related 30-day unplanned readmission in the DUHS OPAT cohort [10,11]. We believe several factors may influence such differences, including the method in which the score is calculated, performance variability in different patient cohorts, and differences in study endpoints.
Conclusions
The models incorporating maximum or discharge Epic readmission scores showed poor discrimination ability in predicting 30-day unplanned readmission in the DUHS OPAT cohort. Incorporating additional DUHS-specific variables, including age, vancomycin use before index discharge, IV drug abuse, and mode of OPAT delivery in a skilled nursing facility did not improve the discrimination ability. The models for predicting 30-day unplanned OPAT-related readmission performed slightly better but the discrimination ability was still poor. There remains something unique in the OPAT population influencing readmissions that is not captured in the standard readmission risk model (such as the Epic model) despite our attempts to improve the model’s discriminating ability by adding variables identified by other investigators. This may include factors such as such as the patients’ subjective needs and/or cost-of-care reimbursement. We believe that more research needs to be done to investigate what unique socioeconomic or clinical factors may be responsible for readmissions in the OPAT population.
Citation: Drew R, Brenneman E, Funaro J, Lee H-J, Yarrington M, Dicks K, et al. (2023) Electronic health record-based readmission risk model performance for patients undergoing outpatient parenteral antibiotic therapy (OPAT). PLOS Digit Health 2(8): e0000323. https://doi.org/10.1371/journal.pdig.0000323
Editor: Benjamin P. Geisler, Massachusetts General Hospital, UNITED STATES
Received: March 3, 2023; Accepted: July 10, 2023; Published: August 2, 2023
Copyright: © 2023 Drew et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Study data are deposited at https://zenodo.org/record/8007308 via zenodo (a cross-discipline generalist repository acceptable under journal policy).
Funding: This study was supported by a grant from the Campbell University College of Pharmacy & Health Sciences (RD). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Research reported in this publication was supported by the National Center For Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR002553. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Competing interests: The authors have declared that no competing interests exist.
https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0000323#sec010
