Any explanations for calculating relative risk (RR) for incidence and odds ratio (OR) for the longitudinal prevalence of diarrhea in randomized controlled trials to evaluate the effectiveness of intervention?
Clinical trials are longitudinal studies. They measure the incidence in the groups under study and then calculated the relative risk (RR). For example:
Incidence 1 = number of new cases in group 1 / total number of patients in group 1.
Incidence 2 = number of new cases in group 2 / total in group 2 patients.
Antonio is right, but in some cases, like in cluster randomized interventions, you could compare prevalences at followup, adjusted for baseline prevalences (see Lancet. 2012 Mar 24;379(9821):1120-8)