Abstract
In large-scale disease etiology studies, epidemiologists often need to use multiple imperfect binary measures of the unobserved true causes of disease to estimate the cause-specific case fractions, or “population etiologic fractions” (PEFs). Despite recent advances in statistical methods, the scientific need of estimating the effect of explanatory variables upon the PEFs in the presence of control data remains unmet. In this paper, we start with and extend the nested partially-latent class model (npLCM, Wu et al., 2016b) to a general framework for etiology regression analysis in case-control studies. Data from controls provide requisite information about measurement specificities and covariations to correctly assign cause-specific probabilities for each case given her measurements. We estimate the distribution of the controls’ diagnostic measures given the covariates via a separate regression model and a priori encourage simpler dependence structures. We use Markov chain Monte Carlo for posterior inference of the PEF functions, cases’ latent classes and the overall PEFs of policy import. We illustrate the regression analysis via simulations and show less biased estimation and more valid inference of the overall PEFs than an npLCM analysis omitting covariates. Regression analysis of data from a childhood pneumonia study site reveals the dependence of pneumonia etiology upon season, age, disease severity and HIV status.