Disease modification is a primary therapeutic aim when developing novel treatments. One of many challenges for establishing disease modification relates to the identification of adequate analytic tools to show differences in a disease course. Traditional approaches rely on the comparisons of slopes or noninferiority margins. However, conclusively demonstrating disease modification using such approaches has proven difficult.
We propose a novel adaptation of the delayed-start design that incorporates posterior probabilities identified by hierarchical Bayesian inference to establish evidence for disease modification. Our models compare the size of the treatment differences at the end of the delayed-start period with those at the end of the early-start period. These include: (i) general linear models, (ii) repeated-measures models, (iii) spline models, and (iv) model-averaging over a library of splines. Our work supports the superiority of model averaging for accurately characterizing complex data.
This novel approach has been applied to the design of an on-going, doubly-randomized, matched-control study that aims to show disease modification in young persons with schizophrenia.