Interpreting higher-order dependence in multimorbidity using cohort data: A partial information decomposition approach (opens in new tab)

Author summary When older adults live with multiple health conditions simultaneously - a situation called multimorbidity - their symptoms and health indicators can interact in surprising ways. Two variables that seem unremarkable on their own may together carry important predictive information about a third. Standard statistical approaches, which typically examine variables in pairs, cannot detect these joint effects. We apply a method from information theory called Partial Information Decomp...