Spline Regression Modeling

Abstract

Spline modeling is a mathematically refined modeling tool, that adequately fits complex data, even if they do not fit the traditional mathematical models. It is called the digital clay of the twenty-first century, although, so far, little used in clinical research spline modeling, but this is a matter of time. The usual model for the outcome analysis of clinical trials is the linear or log linear model. However, when the linear model is not significant, or when the data plots suggest nonlinearity, the quadratic and cubic models often produce a better fit. The current chapter shows, that spline modeling can produce curves that even better fit the outcome data of such clinical trials, than these traditional models do. Also, patterns in the data can be detected, that may go unobserved with traditional linear or curvilinear modeling. Spline modeling can adequately assess the relationships between an exposure and outcome variable in a clinical trial. It can detect patterns in a trial that are relevant, but go unobserved with simpler regression models. In clinical research spline modeling has great potential, given the presence of many nonlinear effects in this research field, and given its sophisticated mathematical refinement to fit any nonlinear effect in the most accurate way. Spline modeling should enable to improve making predictions from clinical research for the benefit of health decisions and health care. We hope that this brief introduction to spline modeling will stimulate clinical investigators to start using this wonderful method.