Supplemental Data Content to 'Data from 81 cases of subtotal cholecystectomy used to generate a multiple logistic regression model to predict postoperative bile leak'
Description
This is the supplemental data content for the original paper titled “Multiple logistic regression model to predict bile leak associated with subtotal cholecystectomy” (Surg Endosc 2023 Apr 4:1–9. DOI: 10.1007/s00464-023-10049-2. Epub ahead of print. PMID: 37016083; PMCID: PMC10072799), and data article titled “Data from 81 cases of subtotal cholecystectomy used to generate a multiple logistic regression model to predict postoperative bile leak”. The study was reported according to the preferred reporting of case series in surgery (PROCESS) and transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) guidelines. The data are presented by FAIR (Findable, Accessible, Interoperable, Reusable) principles. Multivariable logistic regression revealed two independent predictors of bile leak associated with subtotal cholecystectomy: open-tract STC (odds ratio [OR], 7.07; 95% confidence interval [CI], 2.191–25.89; P = 0.0170) and acute cholecystitis (OR, 5.449; 95% CI, 1.584–23.48; P = 0.0121). The area under the receiver-operating characteristic curve was 82.11% (95% CI, 72.87–91.34; P < 0.0001). Tjur’s pseudo-R2 was 0.3189, and the Hosmer–Lemeshow goodness-of-fit statistic was 4.916 (P = 0.7665). This supplement for the Mendeley repository provides further details to reproduce the study results. It is organised into sections that follow the structure of the articles. It contains nine tables.
Files
Steps to reproduce
The authors analysed a prospectively managed clinical data database of 81 adult patients who underwent an urgent or elective subtotal cholecystectomy by a single consultant surgeon at the University Hospital between 14 May 2013 and 21 December 2021. The database with data points is not shown in this supplement. However, the original paper published in ‘Surgical Endoscopy and Other Interventional Techniques’ and this supplement with processed data provide information sufficient to reproduce a multiple logistic regression model to predict bile leakage associated with subtotal cholecystectomy and estimate its overall performance.