Chart of the progression of patients through the trial. All TLGS participants were residents of district 13 of Tehran. We divided this district into three subareas, of which one was further away from the two other areas and was considered as the case area, consisting of 5,000 subjects. All people in the case area with serum lipid or glucose abnormalities or high blood pressure were invited to the nutrition clinic. Among the invitees who accepted, subjects with metabolic syndrome were randomly divided to two groups: DASH and weight reducing. From the two other areas that constituted the control group, we invited those subjects with metabolic syndrome who were matched with cases for confounder variables. All patients completed the entire study protocol. All of the patients were included in the data analysis, and there were no deviations from the intended protocol.