Analysis of Longitudinal Multivariate Outcome Data From Couples Cohort Studies: Application to HPV Transmission Dynamics

Abstract

We consider a specific situation of correlated data where multiple outcomes are repeatedly measured on each member of a couple. Such multivariate longitudinal data from couples may exhibit multi-faceted correlations which can be further complicated if there are polygamous partnerships. An example is data from cohort studies on human papillomavirus (HPV) transmission dynamics in heterosexual couples. HPV is a common sexually transmitted disease with 14 known oncogenic types causing anogenital cancers. The binary outcomes on the multiple types measured in couples over time may introduce inter-type, intra-couple, and temporal correlations. Simple analysis using generalized estimating equations or random effects models lacks interpretability and cannot fully utilize the available information. We developed a hybrid modeling strategy using Markov transition models together with pairwise composite likelihood for analyzing such data. The method can be used to identify risk factors associated with HPV transmission and persistence, estimate difference in risks between male-to-female and female-to-male HPV transmission, compare type-specific transmission risks within couples, and characterize the inter-type and intra-couple associations. Applying the method to HPV couple data collected in a Ugandan male circumcision (MC) trial, we assessed the effect of MC and the role of gender on risks of HPV transmission and persistence.

HPV Infections in Heterosexual Couples: Mechanisms and Covariates of Virus Transmission

“…Transmission from man to woman or woman to man has been evaluated in six of the recent studies shown in Table 1. Three of the studies have detected that transmission of α-HPV occurs more likely from man to woman [4, 19, 20] and two from woman to man [16, 17]. The only study detecting β- and γ-HPV infection within a 6-week follow-up, so far, has not found any difference in transmission between sexes [6].…”

A Distributed and Integrated Method of Moments for High-Dimensional Correlated Data Analysis

“…In this way, CL is free of the computational burden of inverting high-dimensional correlation matrices and benefits from an objective function that facilitates model selection. Pairwise CL has been used with longitudinal (Kuk and Nott (2000), Kong et al (2015)), spatial (Heagerty and and (1998), Arbia (2014)), spatiotemporal (Bai et al (2012), Bevilacqua et al (2012)), and genetic (Larribe and Fearnhead (2011)) data.…”

Marginal Logistic Regression for Spatially Clustered Binary Data

“…This fitting method alternates between optimal estimating equations for estimation of regression coefficients and pairwise likelihood (Varin et al, 2011) for estimation of dependence parameters. Recent applications of hybrid pairwise likelihood include pairwise comparison models (Cattelan and Varin, 2013) and multivariate longitudinal data (Kong et al, 2015). Pairwise likelihood estimation for the analysis of binary spatial data is discussed, for example, in Heagerty and Lele (1998), Varin et al (2005), Apanasovich et al (2008), Bai et al (2014) and .…”