Model justification and stratification for confounding of Chlamydia trachomatis disease

Abstract

This study involves statistical analysis of reported cases of sexually transmitted diseases (STDs) of Chlamydia infection in the United States. The data are collected from 2007 to 2016. The research studies incidence of sexually transmitted diseases and survival among different age groups and gender and race factors which influence the incidence in the target population. In this work, log-binomial, logit model, probit model and complementary log–log model are used to establish a suitable model (using different criteria) that can predict the survival of infected people with STDs based on their age, gender and race. Here we have also focused on stratification: a statistical technique that allows to control for confounding by creating two or more categories. The Mantel–Haenszel formula allows to calculate an overall, unconfounded, that is adjusted, effect estimate for a specific outcome by combining (pooling) stratum-specific relative risks and odds ratios. Simulation is based on R-software.