Faculty Collaboration Database

Comparison of a Potential Hospital Quality Metric With Existing Metrics for Surgical Quality-Associated Readmission. JAMA Netw Open 2019 Apr 05;2(4):e191313

Date

04/20/2019

Pubmed ID

31002316

Pubmed Central ID

PMC6481441

DOI

10.1001/jamanetworkopen.2019.1313

Scopus ID

2-s2.0-85065015141 (requires institutional sign-in at Scopus site)   18 Citations

Abstract

IMPORTANCE: The existing readmission quality metric does not meaningfully distinguish readmissions associated with surgical quality from those that are not associated with surgical quality and thus may not reflect the quality of surgical care.

OBJECTIVE: To compare a quality metric that classifies readmissions associated with surgical quality with the existing metric of any unplanned readmission in a surgical population.

DESIGN, SETTING, AND PARTICIPANTS: Cohort study using US nationwide administrative data collected on 4 high-volume surgical procedures performed at 103 Veterans Affairs hospitals from October 1, 2007, through September 30, 2014. Data analysis was conducted from October 1, 2017, to January 24, 2019.

MAIN OUTCOMES AND MEASURES: Hospital-level rates of unplanned readmission (existing metric) and surgical readmissions associated with surgical quality (new metric) in the 30 days following hospital discharge for an inpatient surgical procedure.

RESULTS: The study population included 109 258 patients who underwent surgery at 103 hospitals. Patients were majority male (94.1%) and white (78.2%) with a mean (SD) age of 64.0 (10.0) years at the time of surgery. After case-mix adjustment, 30-day surgical readmissions ranged from 4.6% (95% CI, 4.5%-4.8%) among knee arthroplasties to 11.1% (95% CI, 10.9%-11.3%) among colorectal resections. The new surgical readmission metric was significantly correlated with facility-level postdischarge complications for all procedures, with ρ coefficients ranging from 0.33 (95% CI, 0.13-0.51) for cholecystectomy to 0.52 (95% CI, 0.38-0.68) for colorectal resection. Correlations between postdischarge complications and the new surgical readmission metric were higher than correlations between complications and the existing readmission metric for all procedures examined (knee arthroplasty: 0.50 vs 0.48; hip replacement: 0.44 vs 0.18; colorectal resection: 0.52 vs 0.42; and cholecystectomy: 0.33 vs 0.10). When compared with using the existing readmission metric, using the new surgical readmission metric could change hip replacement-associated payment penalty determinations in 28.4% of hospitals and knee arthroplasty-associated penalties in 26.0% of hospitals.

CONCLUSIONS AND RELEVANCE: In this study, surgical quality-associated readmissions were more correlated with postdischarge complications at a higher rate than were unplanned readmissions. Thus, a metric based on such readmissions may be a better measure of surgical care quality. This work provides an important step in the development of future value-based payments and promotes evidence-based quality metrics targeting the quality of surgical care.

Author List

Graham LA, Mull HJ, Wagner TH, Morris MS, Rosen AK, Richman JS, Whittle J, Burns E, Copeland LA, Itani KMF, Hawn MT

Author

Jeffrey Whittle MD Professor in the Medicine department at Medical College of Wisconsin

MESH terms used to index this publication - Major topics in bold

Aged
Arthroplasty, Replacement, Hip
Arthroplasty, Replacement, Knee
Benchmarking
Cohort Studies
Colorectal Surgery
Female
Hospitals
Humans
Male
Middle Aged
Patient Discharge
Patient Readmission
Postoperative Complications
Prospective Studies
Quality of Health Care
Surgical Procedures, Operative
United States
United States Department of Veterans Affairs