An epidemiological systems analysis of diarrhea in children in Pakistan is presented. Applying additive Bayesian network (ABN) modeling to data from the Pakistan Social and Living Standards Measurement (PSLSM) survey reveals the
complexity of child diarrhea as a disease system. The key distinction between standard analytical approaches, such as multivariable regression, and Bayesian network analyzes is that the latter attempts not only to identify statistically associated variables, but to additionally, and empirically, separate these into those directly and indirectly dependent with the outcome variable. Such discrimination is vastly more ambitious but has the potential to reveal far more about key features of complex disease systems. Additive Bayesian network analyzes across 41 variables from the PSLSM identified 182 direct dependencies, but with only
three variables: Access to a dry pit latrine (protective: OR=0.67); Access to an atypical water source (protective: OR=0.49); and No formal garbage collection (unprotective: OR=1.32), supported as directly dependent with the presence of diarrhea. All but two of the remaining variables were also in turn directly or indirectly dependent with these three key variables. These results are contrasted with the use of a standard approach (multivariable regression).