Prediction modelling - Part 1 - Regression modelling

Abstract Risk prediction models are statistical models which estimate the probability of individuals having a certain disease or clinical outcome based on a range of characteristics, and can be used in clinical practice to stratify disease severity and to characterise the risk of disease or disease prognosis. With technological advancements and the proliferation of clinical and biological data, prediction models are increasingly being developed in many areas of nephrology practice. This article will guide the reader through the process of creating a prediction model, including: 1) Defining the clinical question and type of model 2) Data collection and data cleaning 3) Model building and variable selection 4) Model performance 5) Model validation 6) Model presentation and reporting, and 7) Impact evaluation. An example of developing a prediction model to predict mortality after intensive care unit admission for patients with end-stage kidney disease will also be provided to illustrate the model development process.