Novel Statistical Methods Applied in Clinical Trials and Gut Microbiota

Ethical clinical trials need both societal and personal equipoise. Recently, personal equipoise has been disturbed by the introduction of interim analyses; after an interim analysis has been performed the study administrators have additional information about the treatments, which is withheld from new recruits. For true informed consent, this information should be given to new study recruits to use in making a personal decision about their desired treatment. We present a method (and the rationale behind the method) that provides unbiased estimates of hazard ratios when new recruits are given information from interim analyses and allowed to choose their own treatments. We then developed a novel procedure that allows for the identification of longitudinal gut microbiota patterns (corresponding to the gut ecosystem evolving), which are associated with an outcome of interest, while appropriately controlling for the false discovery rate. Finally, using novel statistical models, we investigated the impact of POPs (in particular, non-dioxin-like polychlorinated biphenyl, IUPAC no.: 153; "PCB153") on human health through the disruption of natural gut microbiota establishment in infants. We created novel distributed lag two-part models to account for the cumulative exposure of POPs.