To assess the association between in-hospital use of calcium antagonists and incident reduction in hemoglobin levels, as well as the impact of individual baseline risk for gastrointestinal bleeding on such an association.The association between calcium antagonists and hemoglobin decrease > 1.2 g/dL was examined in 6,721 patients enrolled in a collaborative pharmacoepidemiology study who did not take calcium antagonists before admission and with baseline hemoglobin > or =12 g/dL. Among these participants, 1,076 patients started taking calcium antagonists during their hospital stays. Demographic variables, comorbid conditions, medications, and objective tests that were associated with incident hemoglobin loss in separate age- and sex-adjusted logistic regression models were examined as potential confounders in a summary model. Higher risk for gastrointestinal bleeding was defined by diagnosis, treatment for peptic disease, or both.Hemoglobin decrease was detected in 24% of participants who started treatment with calcium antagonists and in 19% of other patients (P < .0001). In addition, use of calcium antagonists was independently associated with increased probability of hemoglobin loss (odds ratio [OR], 1.22; 95% confidence interval [CI], 1.03 to 1.45; P = .018) after adjusting for potential confounders. Treatment with calcium antagonists was associated with hemoglobin loss in patients with higher baseline risk for gastrointestinal bleeding (OR, 1.67; 95% CI, 1.26 to 2.22; P < .0001) but not among other participants (OR, 1.02; 95% CI, 0.82 to 1.25).Starting treatment with calcium antagonists is associated with a reduction in hemoglobin levels during a hospital stay. However, the increased risk of hemoglobin loss seems to be limited to patients with diagnosis or symptoms of peptic disease.