Combining Mixed Effects Hidden Markov Models with Latent Alternating Recurrent Event Processes to Model Diurnal Active-Rest Cycles

TL;DR

This paper introduces a novel modeling approach combining mixed effects hidden Markov models with latent recurrent event processes to analyze diurnal activity-rest cycles from wearable device data, incorporating covariates and providing robust parameter estimation.

Contribution

It develops an EM algorithm that simplifies to a hidden Markov model with logistic regression transition probabilities, enabling improved modeling of behavioral state transitions with covariates.

Findings

The proposed model accurately captures diurnal activity patterns.

Simulation studies show improved parameter estimation.

Application to adolescent smartphone data reveals meaningful behavioral insights.

Abstract

Data collected from wearable devices and smartphones can shed light on an individual's pattern of behavioral and circadian routine. Phone use can be modeled as alternating event process, between the state of active use and the state of being idle. Markov chains and alternating recurrent event models are commonly used to model state transitions in cases such as these, and the incorporation of random effects can be used to introduce diurnal effects. While state labels can be derived prior to modeling dynamics, this approach omits informative regression covariates that can influence state memberships. We instead propose an alternating recurrent event proportional hazards (PH) regression to model the transitions between latent states. We propose an Expectation-Maximization (EM) algorithm for imputing latent state labels and estimating regression parameters. We show that our E-step simplifies to the hidden Markov model (HMM) forward-backward algorithm, allowing us to recover a HMM with logistic regression transition probabilities. In addition, we show that PH modeling of discrete-time transitions implicitly penalizes the logistic regression likelihood and results in shrinkage estimators for the relative risk. We derive asymptotic distributions for our model parameter estimates and compare our approach against competing methods through simulation as well as in a digital phenotyping study that followed smartphone use in a cohort of adolescents with mood disorders.