A Multivariable Prediction Model for 30-day Readmissions for Patients Discharged with Outpatient Intravenous Antibiotic Therapy.
Allison, Geneve.
2013
-
Abstract: Thirty-day unplanned hospital readmissions contribute to patient morbidity and healthcare costs, and are increasingly scrutinized as a quality measure. Outpatient intravenous antibiotics are used by 250,000 patients per year in the U.S. Awareness of the patient and healthcare associated factors at the time of hospital discharge associated with 30-day readmissions could facilitate targete... read mored approaches to reduce readmissions and improve care. However, factors associated with readmission for patients prescribed intravenous antibiotics at hospital discharge have not been definitively identified to our knowledge. Studies of readmissions for other patient groups have shown conflicting results and predictive models of readmissions have been fair to moderate in their ability to discriminate readmission risk. In this thesis, we describe a new predictive model for patients discharged with outpatient intravenous antibiotic therapy. We conducted a retrospective cohort analysis of 784 patients treated in an outpatient intravenous antibiotic program at a single academic center. We used clinical judgment and statistical criteria to develop a multivariable model of patient characteristics associated with 30-day readmissions. Overall readmission rate was 26%. Our final model included: age by decades (odds ratio 1.09, 95% confidence interval 0.99, 1.21), aminoglycoside use (OR 2.33, 95% CI 1.17, 4.57), presence of resistant organisms (OR 1.57, 95% CI 1.03, 2.36), and number of prior admissions in the past 12 months (OR 1.2, 95% CI 1.09-1.32). Model discrimination was fair (c-statistic 0.61), likely reflecting heterogeneity of the underlying population and post-discharge events. Further studies of outpatient intravenous antibiotics should focus on post-discharge factors that contribute to readmission and are potentially modifiable.
Thesis (M.S.)--Tufts University, 2013.
Submitted to the Dept. of Clinical & Translational Science.
Advisors: David Kent, and David Snydman.
Committee: Robin Ruthazer, and Jessica Paulus.
Keyword: Health sciences.read less - ID:
- nz806b71g
- Component ID:
- tufts:20209
- To Cite:
- DCA Citation Guide EndNote
