Development of a reliable marker of adherence to high-fiber diets is essential for accurately assessing dietary fiber intake in community interventions and clinical trials. The objective of this study was to evaluate the feasibility of using a riboflavin tracer incorporated into wheat bran cereal to determine fiber intake and compare results to the more traditional methodology of measuring stool weight. The inpatient phase of the study established that the excretion of urinary riboflavin was highly correlated with the dose of the riboflavin-spiked wheat bran cereal (r = 0.95, P < 0.005) and could be used as a biomarker to validate fiber supplement intake. The outpatient clinical intervention included a group of seven African-American men and women, who were asked to incorporate 1/2 cup of wheat bran cereal (11.6 g of dietary fiber) into their daily diet for a 6-week period. The cereal was spiked with a 28-mg dose of riboflavin. Baseline measurements of urinary riboflavin and stool weight were compared to postintervention levels. Comparison of pre- and postintervention measures of riboflavin excretion showed a significant increase (0.8 +/- 0.1 versus 6.0 +/- 0.6 mg/day, P < 0.02), which confirmed a high level of adherence to the dietary intervention. Although wet stool weights at baseline were significantly lower than postintervention (106 +/- 20 versus 146 +/- 23 g/day; P < 0.03), differences in dry stool weights did not reach significant levels (28 +/- 4 versus 33 +/- 5 g/day, P < 0.30). Furthermore, pre- and poststool measurements overlapped and could not provide definitive data on participant adherence. These results indicate that the riboflavin tracer was a more sensitive biomarker of wheat bran fiber supplementation than stool weight and provided an accurate method for validating adherence to the dietary intervention. A riboflavin marker provides a valid technique for adherence assessment in large-scale community trials, in which collection of 3-day fecal samples is not a manageable option.