Methods in most prediction studies do not follow guidelines

In this week's PLoS Medicine, Walter Bouwmeester of the University Medical Center Utrecht, the Netherlands and colleagues investigate the reporting and methods of prediction studies in 2008 in six top international general medical journals. The authors' findings reveal that the majority of prediction studies do not follow current methodological recommendations.

The authors state: "We identified poor reporting and poor methods in many published prediction studies, which limits the reliability and applicability of the published findings…Only a very small minority of the papers involved the most useful approaches in predicting participant clinical outcomes, namely, external validations or impact assessments of a previously developed prediction model."