Accurate preoperative prediction of unplanned 30-day postoperative readmission using 8 predictor variables.
Adult
Aged
Datasets as Topic
Feasibility Studies
Female
Humans
Logistic Models
Male
Middle Aged
Patient Readmission
/ statistics & numerical data
Postoperative Complications
/ epidemiology
Preoperative Period
Risk Assessment
/ methods
Risk Factors
Surgical Procedures, Operative
/ adverse effects
Time Factors
Journal
Surgery
ISSN: 1532-7361
Titre abrégé: Surgery
Pays: United States
ID NLM: 0417347
Informations de publication
Date de publication:
11 2019
11 2019
Historique:
received:
21
01
2019
revised:
21
05
2019
accepted:
06
05
2019
pubmed:
6
7
2019
medline:
25
2
2020
entrez:
6
7
2019
Statut:
ppublish
Résumé
Unplanned postoperative readmissions are associated with high costs, may indicate poor care quality, and present a substantial opportunity for healthcare quality improvement. Patients want to know their risk of unplanned readmission, and surgeons need to know the risk to adequately counsel their patients. The Surgical Risk Preoperative Assessment System tool was developed from the American College of Surgeons National Surgical Quality Improvement Program dataset and is a parsimonious model using 8 predictor variables. Surgical Risk Preoperative Assessment System is applicable to >3,000 operations in 9 surgical specialties, predicts 30-day postoperative mortality and morbidity, and is incorporated into our electronic health record. A Surgical Risk Preoperative Assessment System model was developed using logistic regression. It was compared to the 28 nonlaboratory variables model from the American College of Surgeons National Surgical Quality Improvement Program 2012 to 2017 dataset using the c-index as a measure of discrimination, the Hosmer-Lemeshow observed-to-expected plots testing calibration, and the Brier score, a combined metric of discrimination and calibration. Of 4,861,370 patients, 188,150 (3.98%) experienced unplanned readmission related to the index operation. The Surgical Risk Preoperative Assessment System model's c-index, 0.728, was 99.3% of that of the full model's, 0.733; the Hosmer-Lemeshow plots indicated good calibration; and the Brier score was 0.0372 for Surgical Risk Preoperative Assessment System and 0.0371 for the full model. The 8 variable Surgical Risk Preoperative Assessment System model detects patients at risk for postoperative unplanned, related readmission as accurately as the full model developed from all 28 nonlaboratory preoperative variables in the American College of Surgeons National Surgical Quality Improvement Program dataset. Therefore, unplanned readmission can be integrated into the existing Surgical Risk Preoperative Assessment System tool providing moderately accurate prediction of postoperative readmission.
Sections du résumé
BACKGROUND
Unplanned postoperative readmissions are associated with high costs, may indicate poor care quality, and present a substantial opportunity for healthcare quality improvement. Patients want to know their risk of unplanned readmission, and surgeons need to know the risk to adequately counsel their patients. The Surgical Risk Preoperative Assessment System tool was developed from the American College of Surgeons National Surgical Quality Improvement Program dataset and is a parsimonious model using 8 predictor variables. Surgical Risk Preoperative Assessment System is applicable to >3,000 operations in 9 surgical specialties, predicts 30-day postoperative mortality and morbidity, and is incorporated into our electronic health record.
METHODS
A Surgical Risk Preoperative Assessment System model was developed using logistic regression. It was compared to the 28 nonlaboratory variables model from the American College of Surgeons National Surgical Quality Improvement Program 2012 to 2017 dataset using the c-index as a measure of discrimination, the Hosmer-Lemeshow observed-to-expected plots testing calibration, and the Brier score, a combined metric of discrimination and calibration.
RESULTS
Of 4,861,370 patients, 188,150 (3.98%) experienced unplanned readmission related to the index operation. The Surgical Risk Preoperative Assessment System model's c-index, 0.728, was 99.3% of that of the full model's, 0.733; the Hosmer-Lemeshow plots indicated good calibration; and the Brier score was 0.0372 for Surgical Risk Preoperative Assessment System and 0.0371 for the full model.
CONCLUSION
The 8 variable Surgical Risk Preoperative Assessment System model detects patients at risk for postoperative unplanned, related readmission as accurately as the full model developed from all 28 nonlaboratory preoperative variables in the American College of Surgeons National Surgical Quality Improvement Program dataset. Therefore, unplanned readmission can be integrated into the existing Surgical Risk Preoperative Assessment System tool providing moderately accurate prediction of postoperative readmission.
Identifiants
pubmed: 31272812
pii: S0039-6060(19)30284-3
doi: 10.1016/j.surg.2019.05.022
pii:
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, U.S. Gov't, P.H.S.
Langues
eng
Sous-ensembles de citation
IM
Pagination
812-819Subventions
Organisme : NCATS NIH HHS
ID : UL1 TR001082
Pays : United States
Commentaires et corrections
Type : CommentIn
Type : CommentIn
Informations de copyright
Copyright © 2019 Elsevier Inc. All rights reserved.