Poor handling of continuous predictors in clinical prediction models using logistic regression: a systematic review

Background When developing a clinical prediction model, assuming a linear relationship between the continuous predictors and outcome is not recommended. Incorrect specification of the functional form of continuous predictors could reduce predictive accuracy. We examine how continuous predictors are handled in studies developing a clinical prediction model. 

Methods We searched PubMed for clinical prediction model studies developing a logistic regression model for a binary outcome, published between 01/07/2020 and 30/07/2020.

Results 118 studies were included in the review (18 studies (15%) assessed the linearity assumption or used methods to handle nonlinearity and 100 studies (85%) did not). Transformation and splines were commonly used to handle nonlinearity, used in 7 (n=7/18,39%) and 6 (n=6/18, 33%) studies respectively. Categorisation was most often used method to handle continuous predictors (n=67/118, 56.8%) where most studies used dichotomisation (n=40/67,60%). Only ten models included nonlinear terms in the final model (n=10/18,56%).

Conclusion Though widely recommended not to categorise continuous predictors or assume a linear relationship between outcome and continuous predictors, most studies categorise continuous predictors, few studies assess the linearity assumption, and even fewer use methodology to account for nonlinearity. Methodological guidance is provided to guide researchers on how to handle continuous predictors when developing a clinical prediction model.