Estimating Time-Varying Causal Excursion Effect in Mobile Health with Binary Outcomes
Year: 2019
Type: article
Abstract: Advances in wearables and digital technology now make it possible to deliver\nbehavioral mobile health interventions to individuals in their everyday life.\nThe micro-randomized trial (MRT) is increasingly used to provide data to inform\nthe construction of these interventions. In an MRT, each individual is\nrepeatedly randomized among multiple intervention options, often hundreds or\neven thousands of times, over the course of the trial. This work is motivated\nby multiple MRTs that have been conducted, or are currently in the field, in\nwhich the primary outcome is a longitudinal binary outcome. The primary aim of\nsuch MRTs is to examine whether a particular time-varying intervention has an\neffect on the longitudinal binary outcome, often marginally over all but a\nsmall subset of the individual's data. We propose the definition of causal\nexcursion effect that can be used in such primary aim analysis for MRTs with\nbinary outcomes. Under rather restrictive assumptions one can, based on\nexisting literature, derive a semiparametric, locally efficient estimator of\nthe causal effect. We, starting from this estimator, develop an estimator that\ncan be used as the basis of a primary aim analysis under more plausible\nassumptions. Simulation studies are conducted to compare the estimators. We\nillustrate the developed methods using data from the MRT, BariFit. In BariFit,\nthe goal is to support weight maintenance for individuals who received\nbariatric surgery.
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7134719707