Determining compliance with a dietary fiber supplement.

The purpose of this pilot study was to evaluate possible ways to determine compliance with a dietary fiber supplement. A wheat bran supplement (30 g daily) significantly increased mean daily wet and dry stool weights (SW) in 7 adults, when compared to SW during an ad libitum low-fiber diet (paired t-test, P less than .01). Because unpaired data would be used during a clinical trial, distribution of the 7 ad libitum low-fiber mean SW observations was used to establish a reference distribution and an upper confidence limit against which the bran supplement SW could be compared. Only one of the seven bran supplement mean SW was above the confidence limit of the low-fiber period, independent of the number of days of collection (2-10) used to calculate the individual mean daily SW. Total fecal output over varying periods of time (2-10 days) suffered the same intersubject and intrasubject variability. Most (5-6) of the bran mean daily SW were above the group mean SW of the low-fiber period. However, this dose of bran was large enough to significantly decrease calcium absorption, and differences in SW produced by lower doses of wheat bran would probably not be as great. The bulk (greater than or equal to 80%) of a single dose of a fecal marker, chromium sesquioxide, which could be incorporated into a specific day's fiber supplement, was recovered in 5 days of excretion during the control period and in 4 days during the bran period. However, the blue color of the chromium before ingestion is clearly a negative feature. Another marker, polyethylene glycol, could not be recovered in excreta when transient time was 4 days or more. In a separate study, demonstration of very little overlap in the concentration of fecal neutral detergent fiber between the control and bran periods suggests that fecal fiber may be a marker of compliance with a fiber supplement.