Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): The TRIPOD statement

摘要: Prediction models are developed to aid health-care providers in estimating the probability or risk that a specific disease condition is present (diagnostic models) event will occur future (prognostic models), inform their decision making. However, overwhelming evidence shows quality of reporting prediction model studies poor. Only with full and clear information on all aspects can bias potential usefulness be adequately assessed. The Transparent Reporting multivariable for Individual Prognosis Or Diagnosis (TRIPOD) Initiative set recommendations developing, validating, updating model, whether diagnostic prognostic purposes. This article describes how TRIPOD Statement was developed. An extensive list items based review literature created, which reduced after Web-based survey revised during 3-day meeting June 2011 methodologists, professionals, journal editors. refined several meetings steering group e-mail discussions wider contributors. resulting checklist 22 items, deemed essential transparent study. aims improve transparency study regardless methods used. best used conjunction explanation elaboration document. To editorial process readers studies, it recommended authors include completed submission (also available at www.tripod-statement.org).