Bayesian analysis of ambulatory blood pressure dynamics with application to irregularly spaced sparse data

摘要

Ambulatory cardiovascular (CV) measurements provide valuable insights intoindividuals' health conditions in "real-life," everyday settings. Currentmethods of modeling ambulatory CV data do not consider the dynamiccharacteristics of the full data set and their relationships with covariatessuch as caffeine use and stress. We propose a stochastic differential equation(SDE) in the form of a dual nonlinear Ornstein--Uhlenbeck (OU) model withperson-specific covariates to capture the morning surge and nighttime dippingdynamics of ambulatory CV data. To circumvent the data analytic constraint thatempirical measurements are typically collected at irregular and much largertime intervals than those evaluated in simulation studies of SDEs, we adopt aBayesian approach with a regularized Brownian Bridge sampler (RBBS) and anefficient multiresolution (MR) algorithm to fit the proposed SDE. The MRalgorithm can produce more efficient MCMC samples that is crucial for validparameter estimation and inference. Using this model and algorithm to data fromthe Duke Behavioral Investigation of Hypertension Study, results indicate thatage, caffeine intake, gender and race have effects on distinct dynamiccharacteristics of the participants' CV trajectories.