We have undertaken a randomized trial to confirm the ability of a class of phenolics, flavonoids, to increase urinary anti-mutagenicity in smokers. Ninety heavy smokers were recruited and randomly assigned to three groups, who were given three different diets. One diet was rich in flavonoids, but not based on supplementation ('flavonoid'), one was a normal iso-caloric diet with an adequate administration of fruit and vegetables ('normal'), and one was based on supplementation of flavonoids in the form of green tea and soy products ('supplement'). The urinary anti-mutagenicity-as inhibiting effect of the urinary extracts on the mutations induced by MeIQx-was measured in Salmonella typhimurium YG1024 in the presence of liver S9 from male Sprague-Dawley rats treated with Aroclor 1254. The amount of total phenolics in the urinary extracts was measured by use of spectrometric analysis. We found that important dietary modifications can be achieved through special recipes and instructions given by a cook during an intensive course. The intervention was focused on increasing the flavonoid intake, and it was successful in that respect. In fact, differences in flavonoid intake were appreciated mainly between the first group (normal diet) and the other two (flavonoid-rich and supplemented diet), suggesting that dietary modification can be as effective as supplementation. However, both urinary anti-mutagenicity and the amounts of urinary phenolics did not change as a consequence of the trial. These results suggest that only a small fraction of urinary phenolics is influenced by dietary changes in the intake of flavonoids, and that most urinary anti-mutagens and phenolics are metabolites of dietary flavonoids, whose formation is more affected by the activity and diversity of bacterial flora in the colon than by the quantity and type of intake. A strong correlation was found between urinary phenolics and anti-mutagenicity in all the groups involved in the trial. Such correlation was not explained by dietary variables.