Background Advances in machine learning (ML) provide great opportunities in the prediction of hospital readmission. This review synthesizes the literature on ML methods and their performance for predicting hospital readmission in the US. Methods This review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Extension for Scoping Reviews (PRISMA-ScR) Statement. The extraction of items was also guided by the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS). Electronic databases PUBMED, MEDLINE, and EMBASE were systematically searched from January 1, 2015, through December 10, 2019. The articles were imported into COVIDENCE online software for title/abstract screening and full-text eligibility. Observational studies using ML techniques for hospital readmissions among US patients were eligible for inclusion. Articles without a full text available in the English language were excluded. A qualitative synthesis included study characteristics, ML algorithms utilized, and model validation, and quantitative analysis assessed model performance. Model performances in terms of Area Under the Curve (AUC) were analyzed using R software. Quality in Prognosis Studies (QUIPS) tool was used to assess the quality of the reviewed studies. Results Of 522 citations reviewed, 43 studies met the inclusion criteria. A majority of the studies used electronic health records (24, 56%), followed by population-based data sources (15, 35%) and administrative claims data (4, 9%). The most common algorithms were tree-based methods (23, 53%), neural network (NN) (14, 33%), regularized logistic regression (12, 28%), and support vector machine (SVM) (10, 23%). Most of these studies (37, 85%) were of high quality. A majority of these studies (28, 65%) reported ML algorithms with an AUC above 0.70. There was a range of variability within AUC reported by these studies with a median of 0.68 (IQR: 0.64–0.76; range: 0.50–0.90). Conclusions The ML algorithms involving tree-based methods, NN, regularized logistic regression, and SVM are commonly used to predict hospital readmission in the US. Further research is needed to compare the performance of ML algorithms for hospital readmission prediction.

No abstract

Background Fondaparinux sodium has been compared with low‐molecular‐weight heparins ( LMWH ) in randomized controlled trials for perioperative surgical thromboprophylaxis. However, the results from these studies are inconsistent in terms of efficacy and safety to reach a clinical decision. The objective of this study was to systematically review the randomized controlled trials comparing the efficacy and safety of fondaparinux and LMWH for perioperative surgical thromboprophylaxis. Methods and Results Systematic search in various databases was done to identify randomized controlled trials comparing fondaparinux and LMWH published during the years 2000 to 2017. Outcomes of interest in this study included venous thromboembolism up to day 15, all‐cause mortality up to day 90, major bleeding, and minor bleeding during the treatment period. Analyses were performed with the relative odds based on a random‐effects model using Mantel‐Haenszel statistics. Results were presented as odds ratios with their 95% CIs. The assessment of study quality was performed as per Cochrane collaboration. After screening 10 644 articles, 12 randomized controlled trials including 14 906 patients were included in the final analyses. Pooled analyses showed the odds of venous thromboembolism in the fondaparinux group were 0.49 times the odds in LMWH group ( OR =0.49 [0.38–0.64]). However, the odds of major bleeding in the fondaparinux group were 1.48 times the odds in the LMWH group ( OR =1.48 [1.15–1.90]). Conclusions Fondaparinux was associated with a superior efficacy in terms of reduction of venous thromboembolism in this meta‐analysis. However, it was also associated with increased odds of major bleeding.

No abstract