Prediction of Prolonged Length of Hospital Stay After Cancer Surgery Using Machine Learning on Electronic Health Records: Retrospective Cross-sectional Study

Abstract

Background Postoperative length of stay is a key indicator in the management of medical resources and an indirect predictor of the incidence of surgical complications and the degree of recovery of the patient after cancer surgery. Recently, machine learning has been used to predict complex medical outcomes, such as prolonged length of hospital stay, using extensive medical information. Objective The objective of this study was to develop a prediction model for prolonged length of stay after cancer surgery using a machine learning approach. Methods In our retrospective study, electronic health records (EHRs) from 42,751 patients who underwent primary surgery for 17 types of cancer between January 1, 2000, and December 31, 2017, were sourced from a single cancer center. The EHRs included numerous variables such as surgical factors, cancer factors, underlying diseases, functional laboratory assessments, general assessments, medications, and social factors. To predict prolonged length of stay after cancer surgery, we employed extreme gradient boosting classifier, multilayer perceptron, and logistic regression models. Prolonged postoperative length of stay for cancer was defined as bed-days of the group of patients who accounted for the top 50% of the distribution of bed-days by cancer type. Results In the prediction of prolonged length of stay after cancer surgery, extreme gradient boosting classifier models demonstrated excellent performance for kidney and bladder cancer surgeries (area under the receiver operating characteristic curve [AUC] >0.85). A moderate performance (AUC 0.70-0.85) was observed for stomach, breast, colon, thyroid, prostate, cervix uteri, corpus uteri, and oral cancers. For stomach, breast, colon, thyroid, and lung cancers, with more than 4000 cases each, the extreme gradient boosting classifier model showed slightly better performance than the logistic regression model, although the logistic regression model also performed adequately. We identified risk variables for the prediction of prolonged postoperative length of stay for each type of cancer, and the importance of the variables differed depending on the cancer type. After we added operative time to the models trained on preoperative factors, the models generally outperformed the corresponding models using only preoperative variables. Conclusions A machine learning approach using EHRs may improve the prediction of prolonged length of hospital stay after primary cancer surgery. This algorithm may help to provide a more effective allocation of medical resources in cancer surgery.

Factors associated with prolonged postoperative length of hospital stay after laparoscopic colorectal cancer resection: a secondary analysis of a randomized controlled trial

“…We found that PPLOS was more likely in patients undergoing Mile's operation and left hemicolectomy than in those undergoing other surgical approaches. This might be related to more complex surgical procedures and greater trauma in these two surgical approaches compared with other surgical approaches [14,16]. We found that tumor metastasis was a protective factor against PPLOS.…”

“…However, multivariate analysis found that other factors were also related to PPLOS. These results indicated that PPLOS was the result of a combination of surgical factors and multiple factors [13,14]. Herein, variables related to anesthesia management, such as intraoperative lung-protective ventilation, urine volume, and lactic acid level, were not found to be related to PPLOS.…”

Textbook Oncologic Outcome Summarizes the Perioperative Cancer Journey, but Should it be used to Judge Hospitals?

“…They can be reproduced with machine learning, where multifactorial prediction models for 30-day mortality and length of stay have an area under the receiver operating characteristic curve (AUC) of [0.8 for some cancer surgeries. 8 Methodological excellence is apparent in the current study by Aquina et al, where additional risk adjustment, controlling for patient demographics, comorbidity status, tumor characteristics, and pathological stage, was undertaken because of evidence of residual confounding on bivariate analysis of patient and oncologic factors by adjusted hospital TOO rate quintile. While this additional adjustment controls for those factors captured in the NCDB data, it still falls short of acknowledging the complexity of technical, social, and biological factors that have all been correlated to cancer outcomes.…”

Technology Innovations in Anesthesiology

“…Those for mortality, hospital length of stay, and intermediate outcomes including acute kidney injury, deep venous thrombosis, re-intubation, and delirium, among others, have all been developed. [11][12][13][14] Models predicting which patients require therapy, and the effect of treatment, have more recently been reported as well as those predicting adverse events. 15,16 Perhaps the best example is intraoperative hypotension, a common adverse event with a clear relationship to multiple organ system injuries.…”