Dataset Information

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.

SUBMITTER: Jo YY

PROVIDER: S-EPMC7939945 | biostudies-literature | 2021 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Jo Yong-Yeon YY Han JaiHong J Park Hyun Woo HW Jung Hyojung H Lee Jae Dong JD Jung Jipmin J Cha Hyo Soung HS Sohn Dae Kyung DK Hwangbo Yul Y

JMIR medical informatics 20210222 2

<h4>Background</h4>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.<h4>Objective</h4>The objective of this study was to develop a prediction model for prolonged length of ...[more]

PMID: 33616544

Similar Datasets

Project description:ObjectivesDengue imposes substantial economic, societal and personal burden in terms of hospital stay, morbidity and mortality. Early identification of dengue cases with high propensity of increased hospital stay and death could be of value in isolating patients in need of early interventions. The current study was aimed to determine the significant factors associated with dengue-related prolonged hospitalisation and death.DesignCross-sectional retrospective study.SettingTertiary care teaching hospital.ParticipantsPatients with confirmed dengue diagnosis were stratified into two categories on the basis of prolonged hospitalisation (≤3 days and >3 days) and mortality (fatal cases and non-fatal cases). Clinico-laboratory characteristics between these categories were compared by using appropriate statistical methods.ResultsOf 667 patients enrolled, 328 (49.2%) had prolonged hospitalisation. The mean hospital stay was 4.88±2.74 days. Multivariate analysis showed that dengue haemorrhagic fever (OR 2.3), elevated alkaline phosphatase (ALP) (OR 2.3), prolonged prothrombin time (PT) (OR 1.7), activated partial thromboplastin time (aPTT) (OR 1.9) and multiple-organ dysfunctions (OR 2.1) were independently associated with prolonged hospitalisation. Overall case fatality rate was 1.1%. Factors associated with dengue mortality were age >40 years (p=0.004), secondary infection (p=0.040), comorbidities (p<0.05), acute kidney injury (p<0.001), prolonged PT (p=0.022), multiple-organ dysfunctions (p<0.001), haematocrit >20% (p=0.001), rhabdomyolosis (p<0.001) and respiratory failure (p=0.007). Approximately half of the fatal cases in our study had prolonged hospital stay of greater than three days.ConclusionsThe results underscore the high proportion of dengue patients with prolonged hospital stay. Early identification of factors relating to prolonged hospitalisation and death will have obvious advantages in terms of appropriate decisions about treatment and management in high dependency units.

Project description:BackgroundThe remarkable advancements in surgical techniques over recent years have shifted the clinical focus from merely reducing mortality to enhancing the quality of postoperative recovery. The duration of a patient's hospital stay serves as a crucial indicator in evaluating postoperative recovery and surgical outcomes. This study aims to identify predictors of the length of hospital stay for children who have undergone corrective surgery for Ebstein Anomaly (EA).MethodsWe conducted a retrospective cohort study on children (under 18 years of age) diagnosed with EA who were admitted for corrective surgery between January 2009 and November 2021 at Fuwai Hospital. The primary outcome was the Time to Hospital Discharge (THD). Cox proportional hazard models were utilized to identify predictors of THD. In the context of time-to-event analysis, discharge was considered an event. In cases where death occurred before discharge, it was defined as an extended THD, input as 100 days (exceeding the longest observed THD), and considered as a non-event.ResultsA total of 270 children were included in this study, out of which three died in the hospital. Following the Cox proportional hazard analysis, six predictors of THD were identified. The hazard ratios and corresponding 95% confidence intervals were as follows: age, 1.030(1.005,1.055); C/R > 0.65, 0.507(0.364,0.707); Carpentier type C or D, 0.578(0.429,0.779); CPB time, 0.995(0.991,0.998); dexamethasone, 1.373(1.051,1.795); and transfusion, 0.680(0.529,0.875). The children were categorized into three groups based on the quartile of THD. Compared to children in the ≤ 6 days group, those in the ≥ 11 days group were associated with a higher incidence of adverse outcomes. Additionally, the duration of mechanical ventilation and ICU stay, as well as hospital costs, were significantly higher in this group.ConclusionWe identified six predictors of THD for children undergoing corrective surgery for EA. Clinicians can utilize these variables to optimize perioperative management strategies, reduce adverse complications, improve postoperative recovery, and reduce unnecessary medical expenses.

Dataset Information

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

Background

Objective

Methods

Results

Conclusions

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets