Objective: To determine the effect of nonesterified, nonhydrogenated plant sterols solubilized in a partly vegetable oil-filled low-fat milk on serum low-density lipoprotein cholesterol (LDL) in mildly hypercholesterolemic patients. Design: Double-blind, randomized, placebo-controlled three-arm crossover study. Setting: Outpatient clinical trial. Subjects: A total of 138 patients were screened, providing 81 patients for randomization; 71 patients completed the study. Interventions: The study product was a 500 ml milk blend with or without nonesterified, nonhydrogenated sterols. The daily consumption of sterols in the three groups was 0 g/day, control group ( C); 1.2 g/day, (Lo); or 1.6 g/day, ( Hi), respectively. The patients were randomly assigned to one of three different treatment sequences. Each intervention period was 4 weeks. The total study duration was 12 weeks. Results: The milk product was well tolerated. The placebo-adjusted mean reduction in LDL was 7.13 +/- 12.31 and 9.59 +/- 12.44% (mean +/-s.d.) for Lo and Hi groups, respectively ( P<0.0001); there was no statistically significant difference in LDL lowering for the Lo and Hi groups. There were no significant changes in serum vitamin E or carotenoid concentrations after standardization with LDL cholesterol during the study period. Conclusion: The present study shows for the first time a substantial reduction in LDL cholesterol with a new, partly vegetable oil-filled 1.2% low-fat milk product, containing nonesterified plant sterols from soybean oil, in a randomized, placebo-controlled trial. This result encourages further development of novel low-fat dairy products containing free plant sterols for future use in cholesterol-lowering initiatives. Sponsorship: The study was financed by Nestec Ltd, Switzerland.