graded doses that kept the ratio of thiamin to energy constant for those studied who had different energy requirements. Other studies provided total amounts of thiamin (and sometimes energy) that were the same for all individuals. Sauberlich and colleagues (1979) adjusted activity levels rather than energy intake to maintain weight in their subjects. Several investigators examined their data to assess whether it would be better to express thiamin as an absolute value or in relation to energy. For example, Dick and colleagues (1958) reported that the coefficient of variation of the estimated thiamin requirement for adolescent boys was 14.2 percent/person, 15.5 percent/1,000 kcal, 27.5 percent/kg body weight, 19.5 percent/m² surface area, and 19.2 percent/mg of creatinine excretion. Elsom and coworkers (1942) noted that they could not distinguish whether it was better to express thiamin in absolute values or per 1,000 kcal but that thiamin intake expressed per body weight did not discriminate between those who were deficient and those who were not. Anderson and colleagues (1986) presented evidence that expressing the thiamin requirements in absolute terms is more useful for predicting biochemical thiamin status than expressing it in relation to energy intake, and data from individuals presented by Henshaw and coworkers (1970) appear supportive.

Despite the lack of direct experimental data, the known biochemical function of thiamin as thiamin pyrophosphate (TPP) in the metabolism of carbohydrate suggests that at least a small (10 percent) adjustment to the estimated requirement to reflect differences in the average energy utilization and size of men and women, a 10 percent increase in the requirement to cover increased energy utilization during pregnancy, and a small increase to cover the energy cost of milk production during lactation may be necessary. It has been observed that during periods of starvation such as in war, larger individuals present signs of beriberi more rapidly than do those with smaller body builds, indicating their greater needs for thiamin and other energy-related nutrients (Burgess, 1946). Many studies report thiamin intake per 1,000 kcal; others report total intake. Thus, the evidence below is presented as it was done in the studies and not because the ratio is considered important.

Heavy exercise under certain conditions may increase the requirement for thiamin as well as other vitamins, but the observations on the effects of physical activity on the thiamin requirement have been inconsistent, the effects small, and the experimental conditions