Statistical models for respiratory disease diagnosis and prognosis

Risk prediction equations are used in a variety of healthcare settings to provide prognosis for patients with various respiratory conditions. This article provides a review of statistical methods for the development, evaluation and implementation of respiratory disease prediction models. We also consider a second, closely related application of these methods: the creation of equations that describe normal lung function in a particular population and the use of such equations in the diagnosis of abnormal lung function. The methods are illustrated with examples of models that have been developed for use in respiratory medicine and research.