Ecologic studies are important for hypothesis generation, e.g. dietary intact of calories in countries correlated with the prevalence of diabetes.  However they are always plagued by the ecologic fallacy that one is not taking measures on an individual basis by rather on groups (Supercourse)