Transparent reporting of multivariable prediction models for individual prognosis or diagnosis: checklist for systematic reviews and meta-analyses (TRIPOD-SRMA)

Clinical prediction models are used to predict health related outcomes in individual patients.12 These models focus on predicting whether outcomes are either already present (diagnostic models) or will occur in the future (prognostic models).34 Examples include the 4C Deterioration model for estimating risk of in-hospital clinical deterioration in adults with covid-19,5 the QRISK3 model for calculating risk of cardiovascular disease onset within 10 years,6 and the Wells’ score for estimating the risk of a deep vein thrombosis in individuals admitted to hospital with suspected deep vein thrombosis.7Box 1 includes a glossary of terms relating to prediction models. Prediction model research includes primary studies of model development and model evaluation (validation).89 Model development studies typically use statistical or machine learning methods to derive a model (eg, regression equation, random forest, or neural network) for predicting a specific outcome (eg, cardiovascular disease within 10 years) based on multiple variables (predictors) such as…
Read Original Article: Transparent reporting of multivariable prediction models for individual prognosis or diagnosis: checklist for systematic reviews and meta-analyses (TRIPOD-SRMA) »