A Change Point Regression Model for Treatment Effects Estimation of Functional Foods in a Pre-Post Design

Abstract

In an intervention study for functional foods, the study population is often a mixed group of healthy individuals and/or those who are not so healthy but are not definitely disease. For such heterogeneous populations, we propose an alternative approach for evaluating the efficacy of dietary supplements and functional foods based on a changepoint regression model (CPRM). The CPRM can be performed without stratified or subgroup analysis. By using profile likelihood methods based on the Akaike information criterion, inferences can made easily using standard statistical software. The proposed method was applied to skeletal muscle ratio study data, and the merit of the proposed method was demonstrated by comparing it with the conventional method. Keywords: Change point regression model Clinical trial; Dietary supplements; AIC