Previous studies examining the hypocholesterolemic effects of high-soluble-fiber diets have not been designed to control for dietary fat intake. Serum cholesterol reductions may therefore be accounted for by differences in consumption of fat. Moderately hypercholesterolemic, nonobese, Caucasian men and women, 30-50 y old were randomly assigned to low-fat, low-fat plus high-fiber, or usual-diet groups and followed for 12 mo. At 12 mo the high-fiber group consumed significantly more soluble fiber than both the low-fat and usual-diet groups (P = 0.0063 and P = 0.0001); the high-fiber group did not differ from the low-fat group in quantity of dietary fat consumed. The high-fiber group experienced a greater average reduction (13%) in serum cholesterol than did the low-fat (9%) and usual-diet (7%) groups. After adjustment for relevant covariates, the reduction in the high-fiber group was significantly greater than that in the low-fat group (P = 0.0482). Supplementation with soluble fiber reduces serum cholesterol beyond the reduction observed with low-fat diet alone.