633. Prediction Model for 30-day Hospital Readmissions among Patients Discharged with Outpatient Parenteral Antibiotic Therapy (OPAT)
Session: Oral Abstract Session: Outpatient Antimicrobial Therapy
Friday, October 4, 2013: 8:45 AM
Room: The Moscone Center: 300
Background: Factors associated with readmission for patients discharged with outpatient parenteral antibiotics therapy (OPAT) have not been definitively identified.  Our study’s aim was to develop a predictive model of 30-day readmissions for patients discharged with OPAT.

Methods: We analyzed 784 patients treated between 2009 and 2011 in a single academic center OPAT program.  Variables collected included patient demographics, length of stay, comorbidities, prior non-OPAT admissions, drug resistant organisms, infections, antibiotics classes.  Factors whose univariable analysis resulted in significance level p<0.2 were candidate variables for the multivariable model.  The final model was assessed by receiver operator curve.

Results: The average age of our patient population was 58 years, 43% were women.  The most common diagnoses were bacteremia (24%), osteomyelitis (20%), pyelonephritis (13%) and intra-abdominal infection (11%).  The overall unplanned hospital readmission rate was 26% with one-third worsening infection and one-third non-infectious issues.  Final regression model consisted of patients’ age (Odds Ratio 1.09 per decade, 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 discharged without intravenous antibiotics in the past 12 months (OR 1.2 per prior admission, 95% CI 1.09-1.32).   Model discrimination was moderate (c-statistic 0.61), likely reflecting heterogeneity of the underlying population and post-discharge events. 

Conclusion: This study highlights the high rate of readmissions in a real-world OPAT setting and the need to develop evidence-based interventions to prevent OPAT readmissions.  Age and prior admission were expected to be associated with readmissions given prior non-OPAT readmission literature. Aminoglycosides and drug resistant organisms are important factors in OPAT care meriting further study. This model demonstrates several factors readily available at time of discharge that could be used in future OPAT outcome research.  Including these factors could ensure that patient characteristics highly associated with readmissions are included in future OPAT research designs.

Geneve Allison, MD, FACP1, Eavan Muldoon, MB, BCh, MD1, David Kent, MD, MS2, Jessica Paulus, ScD2, Robin Ruthazer, MPH2, Aretha Ren, B.A.1 and David Snydman, MD, FIDSA1, (1)Geographic Medicine and Infectious Diseases, Tufts Medical Center, Boston, MA, (2)Tufts Clinical Translational Science Institute, Tufts Medical Center, Boston, MA


G. Allison, Coram/Apria Healthcare: Consultant, Consulting fee
Merck: Investigator, Grant recipient
Merck: Speaker in Peer-to-Peer series - no slides, Speaker honorarium
National Institute of Health: Grant Investigator, Educational grant, Educational support, Grant recipient, Research grant, Research support and Salary

E. Muldoon, None

D. Kent, None

J. Paulus, None

R. Ruthazer, None

A. Ren, None

D. Snydman, None

Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.