The general impact of blood donation on iron status was studied in a population survey comprising 1359 nonpregnant Danish women in age cohorts of 30, 40, 50, and 60 years; 809 were premenopausal and 550 postmenopausal; 180 (13%) were blood donors. Iron stores were assessed by serum (S-)ferritin and hemoglobin (Hb). Hb levels were not significantly different in donors: mean 137 +/- 10 (SD) g/l (8.5 +/- 0.6 mmol/l) compared with nondonors, 139 +/- 11 g/l (8.6 +/- 0.7 mmol/l). Values < 121 g/l (7.5 mmol/l) were observed in 3.3% of donors vs 3.8% of nondonors. Correlations between S-ferritin and Hb were without practical relevance: r(s) = 0.29, p < 0.0001 in donors vs r(s) = 0.22, p < 0.0001 in nondonors. Blood donation had a profound influence on iron status, especially in the premenopausal women population. Donors had lower S-ferritin than nondonors in all age-groups and in pre- and postmenopausal groups (p < 0.001 in all groups). Premenopausal donors had a median S-ferritin of 31-mu-g/l vs 39-mu-g/l in nondonors, postmenopausal donors of 47-mu-g/l vs 72-mu-g/l in nondonors. S-ferritin values < 15-mu-g/l (i.e., depleted iron stores) were observed in 31.7% of premenopausal donors vs 15.2% of nondonors, and in 7.0% of postmenopausal donors vs 2.9% of nondonors. Iron deficiency anemia (i.e., S-ferritin < 15-mu-g/l and Hb < 121 g/l) was seen in 2.8% of donors vs 1.5% of nondonors. Donors using oral contraceptives had higher S-ferritin, median 33-mu-g/l compared with nonusers, 22-mu-g/l, and a lower frequency of depleted iron stores, 29% vs 39%. Ideally, the frequency of phlebotomy should be adjusted according to S-ferritin as well as Hb levels. If Hb is used as single criterion for donation, only donors with predonation values greater-than-or-equal-to 124-125 g/l should be allowed to undergo phlebotomy.