The use of fractional polynomials to model continuous risk variables in epidemiology

Background The traditional method of analysing continuous or ordinal risk f actors by categorization or linear models may be improved. Methods We propose an approach based on transformation and fractional polyn omials which yields simple regression models with interpretable curves. We suggest a way of presenting the results from such models which involves tab ulating the risks estimated from the model at convenient values of the risk factor. We discuss how to incorporate several continuous risk and confound ing variables within a single model. The approach is exemplified with data from the Whitehall I study of British Civil Servants. We discuss the approa ch in relation to categorization and non-parametric regression models. Results We show that non-linear risk models fit the data better than linear models. We discuss the difficulties introduced by categorization and the a dvantages of the new approach. Conclusions Our approach based on fractional polynomials should be consider ed as an important alternative to the traditional approaches for the analys is of continuous variables in epidemiological studies.