Predicting future outcomes of patients is essential to clinical practice, with many prediction models published each year. Empirical evidence suggests that published studies often have severe methodological limitations, which undermine their usefulness. This article presents a step-by-step guide to help researchers develop and evaluate a clinical prediction model. The guide covers best practices in defining the aim and users, selecting data sources, addressing missing data, exploring alternative modelling options, and assessing model performance. The steps are illustrated using an example from relapsing-remitting multiple sclerosis. Comprehensive R code is also provided. Clinical prediction models aim to forecast future health outcomes given a set of baseline predictors to facilitate medical decision making and improve people’s health outcomes.1 Prediction models are becoming increasingly popular, with many new ones published each year. For example, a review of prediction models identified 263 prediction models in obstetrics alone2; another review found 606 models related to covid-19.3 Interest in predicting health outcomes has been heightened by the increasing availability of big data,4 which has also led to the uptake of machine learning methods for prognostic research in medicine.56 Several resources are available to support prognostic research. The PROGRESS (prognosis research strategy) framework provides detailed guidance on different types of prognostic research.789 The TRIPOD (transparent reporting of a multivariable prediction model for individual prognosis or diagnosis) statement gives recommendations for reporting and has recently been extended to address prediction model research in clustered datasets.1011121314 PROBAST (prediction model risk-of-bias assessment tool) provides a structured way to assess the risk of bias in a prediction modelling study.15 Several papers further outline good practices and provide software code.161718 Despite these resources, published prediction modelling studies often have severe methodological limitations. For instance, …

中文翻译：

---

开发临床预测模型：分步指南


预测患者未来的结果对于临床实践至关重要，每年都会发布许多预测模型。经验证据表明，已发表的研究通常存在严重的方法学局限性，从而削弱了其实用性。本文提供了分步指南，帮助研究人员开发和评估临床预测模型。该指南涵盖了定义目标和用户、选择数据源、解决缺失数据、探索替代建模选项以及评估模型性能的最佳实践。使用复发缓解型多发性硬化症的例子来说明这些步骤。还提供了全面的 R 代码。临床预测模型旨在根据一组基线预测因子来预测未来的健康结果，以促进医疗决策并改善人们的健康结果。1预测模型正变得越来越流行，每年都会发布许多新模型。例如，对预测模型的回顾仅在产科领域就确定了 263 个预测模型2；另一项综述发现了 606 个与 covid-19.3 相关的模型。3 随着大数据可用性的不断增加，人们对预测健康结果的兴趣也越来越大，4 这也导致机器学习方法被用于医学预后研究。 56 有多种资源可用于支持预后研究。 PROGRESS（预后研究策略）框架为不同类型的预后研究提供了详细指导。789 TRIPOD（个人预后或诊断多变量预测模型的透明报告）声明给出了报告建议，并且最近已扩展到解决预测模型研究在聚类数据集中。1011121314 PROBAST（预测模型偏倚风险评估工具）提供了一种结构化方法来评估预测建模研究中的偏倚风险。15 有几篇论文进一步概述了良好实践并提供了软件代码。161718 尽管有这些资源，但已发表的预测建模研究通常具有严重的方法学局限性。例如， …