ABSTRACT
INTRODUCTION: Inflammatory bowel disease (IBD) is a chronic idiopathic inflammatory disorder that arises from complex interactions between genetics, environment and gut microbiota. It encompasses Crohn's disease, ulcerative colitis and IBD-unclassified. The protracted course of IBD imposes a significant burden on patients' quality of life, economic productivity, social functioning, as well as treatment, hospitalisation and surgery. This study aims to conduct an umbrella review of meta-analyses to systematically evaluate the methodology's quality, potential biases and validity of all epidemiological evidence focused on risk factors for IBD while providing an overview of the evidence concerning IBD risk factors. METHODS AND ANALYSIS: We will systematically search, extract and analyse data from reported systematic reviews and meta-analyses that specifically focus on the risk factors of IBD, following the guidelines outlined in Preferred Reporting Items for Overviews of Reviews. Our search will encompass PubMed, Embase, Web of Science and the Cochrane Database of Systematic Reviews from the initial period up until April 2023 (last update), targeting systematic reviews and meta-analyses based on non-interventional studies. Inclusion criteria allow for systematic reviews and meta-analyses evaluating IBD risk factors across all countries and settings, regardless of ethnicity or sex. The identified risk factors will be categorised according to the health ecological model into innate personal traits, behavioural lifestyles, interpersonal networks, socioeconomic status and macroenvironments. To assess methodological quality for each meta-analysis included in our study, two authors will employ a measurement tool to assess the methodological quality of systematic reviews (AMSTAR)-2, Grading of Recommendations, Assessment, Development and Evaluation (GRADE) criteria along with evidence classification criteria. ETHICS AND DISSEMINATION: Ethical approval is not required for this umbrella review. We will seek to submit the results for publication in a peer-reviewed journal or present it at conferences. PROSPERO REGISTRATION NUMBER: CRD42023417175.
Subject(s)
Inflammatory Bowel Diseases , Meta-Analysis as Topic , Research Design , Systematic Reviews as Topic , Humans , Risk FactorsABSTRACT
Background: Predicting the severity of acute pancreatitis (AP) early poses a challenge in clinical practice. While there are well-established clinical scoring tools, their actual predictive performance remains uncertain. Various studies have explored the application of machine-learning methods for early AP prediction. However, a more comprehensive evidence-based assessment is needed to determine their predictive accuracy. Hence, this systematic review and meta-analysis aimed to evaluate the predictive accuracy of machine learning in assessing the severity of AP. Methods: PubMed, EMBASE, Cochrane Library, and Web of Science were systematically searched until December 5, 2023. The risk of bias in eligible studies was assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST). Subgroup analyses, based on different machine learning types, were performed. Additionally, the predictive accuracy of mainstream scoring tools was summarized. Results: This systematic review ultimately included 33 original studies. The pooled c-index in both the training and validation sets was 0.87 (95 % CI: 0.84-0.89) and 0.88 (95 % CI: 0.86-0.90), respectively. The sensitivity in the training set was 0.81 (95 % CI: 0.77-0.84), and in the validation set, it was 0.79 (95 % CI: 0.71-0.85). The specificity in the training set was 0.84 (95 % CI: 0.78-0.89), and in the validation set, it was 0.90 (95 % CI: 0.86-0.93). The primary model incorporated was logistic regression; however, its predictive accuracy was found to be inferior to that of neural networks, random forests, and xgboost. The pooled c-index of the APACHE II, BISAP, and Ranson were 0.74 (95 % CI: 0.68-0.80), 0.77 (95 % CI: 0.70-0.85), and 0.74 (95 % CI: 0.68-0.79), respectively. Conclusions: Machine learning demonstrates excellent accuracy in predicting the severity of AP, providing a reference for updating or developing a straightforward clinical prediction tool.
