To evaluate the concept of a predetermined "setpoint" for iron stores, 20 healthy elderly individuals (12 blood donors and 8 nondonors) were studied to examine the effects of changes in iron stores on iron absorption. Oral iron-absorption tests revealed a statistically significant increase (P < 0.001) in iron absorption in donors, 7.4 +/- 3.6% (mean +/- SD) compared with 2.5 +/- 1.4% in nondonors. In a comparison of percent iron absorption with changes in baseline iron stores, a statistically significant correlation was noted (r2 = 0.702, P < 0.001). Thus, reductions in iron stores were correlated with increases in iron absorption. Basal gastric acid output was found to be within normal limits (1.62-10.20 mmol/h) in all elderly subjects and unrelated to iron stores. These findings are consistent with iron absorption being regulated according to the degree of depletion of iron stores from predonation values and is consistent with a proposed setpoint theory of iron stores.