Model-Twin Randomization (MoTR) for Estimating One's Own Recurring Individual Treatment Effect

Temporally dense single-person "small data" have become widely available thanks to mobile apps and wearable sensors. Many caregivers and self-trackers want to use these data to help a specific person change their behavior to achieve desired health outcomes. Ideally, this involves discerning possible causes from correlations using that person's own observational time series data. In this paper, we estimate within-individual average treatment effects of physical activity on sleep duration, and vice-versa. We introduce the model twin randomization (MoTR; "motor") method for analyzing an individual's intensive longitudinal data. Formally, MoTR is an application of the g-formula (i.e., standardization, back-door adjustment) under serial interference. It estimates stable recurring effects, as is done in n-of-1 trials and single case experimental designs. We compare our approach to standard methods (with possible confounding) to show how to use causal inference to make better personalized recommendations for health behavior change, and analyze up to almost eight years of the authors' own Fitbit steps and sleep data.

Bio

https://www.ericjdaza.com/

Eric J. Daza has been a biostatistician and health data scientist for 21+ years (BA neurobiology, MPS applied statistics, DrPH biostatistics). Dr. Daza created Stats-of-1, a newsletter/podcast (featured in Forbes, Fortune) on n-of-1 trials (switchback experiments), single-case designs, and personal AI for digital health/medicine. He invented a patent-pending method using his time series causal inference framework. Daza is a DEI leader at the American Statistical Association, Filipino American immigrant, and trained musician.