AbstractSouth Asian populations are predisposed to early onset of the metabolic syndrome. Lifestyle intervention programmes have demonstrated a reduction in the metabolic syndrome and CVD risk; however, the most effective components of the multi-faceted lifestyle interventions are unknown. We studied 2637 Sri Lankan males (n 1237) and females (n 1380), with a mean BMI of 23·9 (sd 4·2) kg/m², aged 22·5 (sd 10·0) years, who had participated in a 5-year lifestyle-modification programme to examine the effect of dietary changes on distinct components of the metabolic syndrome. The dietary intervention comprised advice to replace polished starches with unpolished starches, high-fat meat and dairy products with low-fat products and high-sugar beverages and snacks with low-sugar varieties. For the purposes of this analysis, data from the control and intensive lifestyle groups were combined. Anthropometric and biochemical data were recorded, and a FFQ was completed annually. Multiple regression was used to determine the effect of the dietary changes on distinct components of the metabolic syndrome. The ratio unpolished:polished rice was inversely related to change in fasting glucose (β=−0·084, P=0·007) and TAG (β=−0·084, P=0·005) and positively associated with change in HDL-cholesterol (β=0·066, P=0·031) at the 5-year follow-up after controlling for relevant confounders. Red meat intake was positively associated with fasting glucose concentrations (β=0·05, P=0·017), whereas low-fat (β=−0·046, P=0·018) but not high-fat dairy products (β=0·003, P=0·853) was inversely related to glucose tolerance at the follow-up visit. Replacement of polished with unpolished rice may be a particularly effective dietary advice in this and similar populations.