Association of Procedures and Patient Factors With 30-Day Readmission Rates After Pediatric Surgery | Pediatrics | JAMA Surgery | JAMA Network
[Skip to Navigation]
Sign In
Table 1.  Operations With the Highest 30-Day Unplanned Readmission Rates
Operations With the Highest 30-Day Unplanned Readmission Rates
Table 2.  Multivariable Analysis Examining Risk Factors of 30-Day Unplanned Readmission
Multivariable Analysis Examining Risk Factors of 30-Day Unplanned Readmission
1.
American College of Surgeons National Surgical Quality Improvement Program – Pediatrics Data User Guide. 2013. https://www.facs.org/~/media/files/quality%20programs/nsqip/acs_nsqip_puf_user_guide_2013.ashx. Accessed September 8, 2016.
2.
Brown  EG, Anderson  JE, Burgess  D, Bold  RJ, Farmer  DL.  Pediatric surgical readmissions: are they truly preventable?  J Pediatr Surg. 2017;52(1):161-165.PubMedGoogle ScholarCrossref
Research Letter
October 2017

Association of Procedures and Patient Factors With 30-Day Readmission Rates After Pediatric Surgery

Author Affiliations
  • 1Division of Pediatric Surgery, University of California, Davis Medical Center
JAMA Surg. 2017;152(10):980-981. doi:10.1001/jamasurg.2017.1722

Efforts to create verified pediatric surgical centers require rigorous standards to ensure quality across surgical specialties. However, risk factors for readmission after pediatric surgery are poorly understood. This study aims to identify surgical procedures and patient factors associated with increased rates of readmission from the American College of Surgeons Pediatric National Surgical Quality Improvement Program (NSQIP-P) database.

Methods

A retrospective review of data from the American College of Surgeons NSQIP-P Participant Use Data File from 2012 to 2014 was performed. The NSQIP-P collects data on 147 variables, including preoperative risk factors and intraoperative variables, and tracks 30-day outcomes.1 The data include children younger than 18 years at 56 children’s hospitals in the United States. This research was deemed exempt by the University of California, Davis institutional review board, which waived the need for patient consent because this was an analysis of a national, deidentified database.

We examined all readmissions within 30 days, both related and unrelated to the index operation. Procedures with the highest unplanned readmission rates were identified. Unplanned readmission rates were calculated by specialty. A multivariable logistic regression identified factors predicting unplanned readmission. The analysis of unplanned readmissions is limited to operations with case volumes of at least 39 patients (corresponding to the 25th percentile) to avoid skewing data toward rarely performed procedures. Analysis was performed in Stata, version 14.1 (StataCorp). Significance was defined as P <.05 (1-sided).

Results

Of 183 233 patients captured by NSQIP-P from 2012 to 2014, 8838 patients (4.8%) were readmitted within 30 days, and 712 patients (8.1% of readmitted patients) had more than 1 readmission within 30 days. Almost half of readmissions (n = 4256; 47.0%) were unrelated to the index operation.

Most readmissions (84.3%) were unplanned. Operations with the highest overall unplanned readmission rates were laparoscopic ileostomy/jejunostomy, portoenterostomy, and diagnostic thoracoscopy (Table 1). By specialty, unplanned readmission rates were highest among neurosurgical operations (n = 1598; 11.2%) compared with other specialties (n = 6-3630; 1.3%-5.4%). The median time to readmission (within 30 days) was 13 days (25th percentile, 8 days; 75th percentile, 21 days).

The most common reason for related readmissions was surgical site infection including superficial and deep incisional surgical site infection, organ/space surgical site infection, and wound disruption (n = 1374; 29.4%). The next most common reason was urinary tract infection (n = 174; 3.7%). Of patients with surgical site infections, the most common index procedures were laparoscopic appendectomy, posterior arthrodesis, craniectomy, and operations related to ventriculo-peritoneal shunts. The most common reasons for unrelated readmissions were seizure (n = 30; 1.7%), pneumonia (n = 29; 1.6%), and urinary tract infection (n = 24; 1.3%).

On multivariable logistic regression, predictors of 30-day unplanned readmission included patients undergoing neurosurgical procedures (vs pediatric general surgery: odds ratio [OR], 2.52; 95% CI, 1.96-3.23; P < .001), female sex (OR, 1.24; 95% CI, 1.06-1.44; P = .008), Asian race/ethnicity (vs white: OR, 1.64; 95% CI, 1.10-2.44; P = .02), higher American Society of Anesthesiologists classification (American Society of Anesthesiologists class 2-4 vs 1: class 2 OR, 2.01; 95% CI, 1.47-2.75; class 2 OR, 3.68; 95% CI, 2.66-5.09; class 4 OR, 2.92; 95% CI, 1.98-4.32; P < .001), current cancer or active cancer treatment (OR, 2.06; 95% CI, 1.24-3.42; P = .005), patients with cerebral palsy (OR, 3.69; 95% CI, 1.77-7.69; P < .001), patients with clean-contaminated cases (OR, 1.28; 95% CI, 1.05-1.56; P = .01), and laparoscopic procedures (OR, 1.55; 95% CI, 1.28-1.87; P < .001; Table 2). Patients had lower odds of unplanned readmission if they underwent urologic procedures (vs pediatric general surgery: OR, 0.70; 95% CI, 0.50-0.98; P = .04) or were neonates (OR, 0.75; 95% CI, 0.60-0.93; P = .01). Lower odds of readmission in neonates was likely because length of stay was often longer than 30 days.

Discussion

In strengthening pediatric surgical centers, it is important to reduce unplanned hospital readmissions. This study identifies procedures with particularly high rates of readmission and finds surgical site infections as the most common reason for readmission related to the operation. However, nearly half of readmissions were related to chronic medical conditions and other nonmodifiable risk factors, similar to other research.2 Patients with cancer, patients with cerebral palsy, and patients undergoing neurosurgical procedures are at particularly high risk of readmission. The main limitation of this study is that, while robust, the NSQIP-P database is a sample of children’s surgery centers and may not be nationally representative of all pediatric surgery in the United States. These findings, along with internal analysis of hospital-specific data, will help inform quality improvement measures to decrease readmission rates in pediatric surgery.

Back to top
Article Information

Corresponding Author: Jamie E. Anderson, MD, MPH, Division of Pediatric Surgery, University of California, Davis Medical Center, 2315 Stockton Blvd, OP 512, Sacramento, CA 95817 (jeanderson@ucdavis.edu).

Accepted for Publication: April 4, 2017.

Published Online: June 28, 2017. doi:10.1001/jamasurg.2017.1722

Author Contributions: Dr Anderson had full access to the data and takes responsibility for the integrity of the data and accuracy of the analysis.

Concept and design: All authors.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Anderson, Stark.

Critical revision of the manuscript for important intellectual content: Anderson, Saadai, Hirose.

Statistical analysis: Anderson.

Administrative, technical, or material support: Hirose.

Supervision: Stark, Saadai, Hirose.

Conflict of Interest Disclosures: None reported.

Additional Contributions: American College of Surgeons National Surgical Quality Improvement Program and hospitals participating in the American College of Surgeons National Surgical Quality Improvement Program are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.

References
1.
American College of Surgeons National Surgical Quality Improvement Program – Pediatrics Data User Guide. 2013. https://www.facs.org/~/media/files/quality%20programs/nsqip/acs_nsqip_puf_user_guide_2013.ashx. Accessed September 8, 2016.
2.
Brown  EG, Anderson  JE, Burgess  D, Bold  RJ, Farmer  DL.  Pediatric surgical readmissions: are they truly preventable?  J Pediatr Surg. 2017;52(1):161-165.PubMedGoogle ScholarCrossref
×