Objective: To evaluate the efficacy of computer-tailored nutrition education with objective outcome measures. Design: A 3-group randomized, controlled trial with posttests at 1 and 6 months post-intervention. Setting: Worksites and 2 neighborhoods in the urban area of Rotterdam. Participants: A convenience sample of healthy Dutch adults (n = 442). Interventions: A computer-tailored intervention delivered on CD-ROM; a computer-tailored intervention delivered in print; and a generic information condition. Main Outcome Measures: Blood lipids (total, high-density lipoprotein, and low-density lipoprotein cholesterol, and triacylglycerol) were measured by analyzing venous blood samples. Analysis: Linear mixed model procedure. Results: There were no significant differences among the 3 intervention groups in total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and triacylglycerol. Conclusions and Implications: Contrary to results based on self-report data, no effects of the computer-tailored interventions were found based on objective outcomes. This contradiction calls for a critical reflection on the use of computer-tailored nutrition education interventions and the need to improve those interventions. Furthermore, this study indicates that feasible methods are needed to objectively assess the impact of computer-tailored nutrition education interventions in free-living subjects. (Contains 2 tables and 1 figure.)