Abstract
Statistical analysis to evaluate mechanistic pathways can be limited by non-causal associations as well as co-linearity of high-dimensional data. Here, we present a protocol evaluating statistical associations between multiple exposure variables (sociodemographic and behavioral), immune biomarkers, and HIV acquisition. We describe steps for study setup, combining Least Absolute Shrinkage and Selective Operator with the standard regression approach, and building nested models. This approach can determine to what extent associations between risks for exposure contributes to HIV acquisition with or without associated changes in immune activation. For complete details on the use and execution of this protocol, please refer to Bender Ignacio et al.(1).