Background: The reason for the high prevalence of mild cobalamin (vitamin B-12) deficiency in the elderly is poorly understood, Objective: We aimed to determine the reason for this high prevalence, Design: We examined cobalamin intake, the presence and severity of atrophic gastritis, the presence of Helicobacter pylori infection, and plasma cobalamin and methylmalonic acid (MMA) concentrations in 105 healthy, free-living, older subjects aged 74-80 y. Results: Mild cobalamin deficiency, ie, low to low-normal plasma cobalamin concentrations (<260 pmol/L) and elevated plas ma MMA concentrations (>0.32 mu mol/L), were found in 23.8% of subjects; 25.7% of subjects were not cobalamin deficient (plasma cobalamin greater than or equal to 260 pmol/L and plasma MMA less than or equal to 0.32 mu mol/L). Six subjects (5.8%), including 1 with mild cobalamin deficiency, had dietary cobalamin intakes below the Dutch recommended dietary intake of 2.5 mu g/d. Mildly cobalamin-deficient subjects had lower total (diet plus supplements) cobalamin intakes (median: 4.9 mu g/d: 25th and 75th percentiles: 3.9, 6.4) than did non-cobalamin-deficient subjects (median: 6.3 mu g/d: 25th and 75th percentiles: 5.4, 7.9) (P = 0.0336). mainly because of less frequent use of cobalamin supplements (8% compared with 29.6%; chi(2) = 3.9. P = 0.048). Atrophic gastritis was found in 32.4% of the total study group: mild to moderate in 19.6% and severe in 12.7%. The prevalence of severe atrophic gastritis, but not mild-to-moderate atrophic gastritis, was higher in mildly cobalamin-deficient subjects (25%) than in non-cobalamin-deficient subjects (3.7%) (chi(2) = 4.6, P = 0.032). The prevalence of immunoglobulin G antibodies to H. pylori was similar in mildly cobalamin-deficient subjects (54.2%) and in non-cabalamin-deficient subjects (44.4%) (chi(2) = 0.5, P = 0.5). Conclusions: The high prevalence of mild cobalamin deficiency in healthy, free-living, older Dutch subjects could be explained by inadequate cobalamin intake or severe atrophic gastritis in only 28% of the study population. Other mechanisms explaining mild cobalamin deficiency in older people must be sought.