Unexplained decline in the prevalence of anemia among US children and women between 1988-1994 and 1999-2002.

The current anemia burden among US preschool children and women of childbearing age has not been documented. We used data from National Health and Nutrition Examination Surveys 1988-1994 and 1999-2002 to examine recent anemia changes. We calculated the prevalence of anemia (hemoglobin < 11.0 g/dL at <24 mo, <11.1 g/dL at 24-59 mo, and <12.0 g/dL for women), iron deficiency anemia (anemia plus abnormal value >or=2: serum ferritin, transferrin saturation, and erythrocyte protoporphyrin), and high blood lead (>or=10 microg/dL) with anemia among children 12-59 mo and women 20-49 y in both surveys. Among women, we also calculated the prevalence of folate deficiency (erythrocyte folate < 317.2 nmol/L) with anemia and high C-reactive protein (>10 mg/L) with anemia. Multiple logistic regression was used to compare anemia prevalence between surveys, with control for race and age. Anemia declined significantly in children (from 8.0% to 3.6%; OR: 0.4; 95% CI: 0.3, 0.7) and women (10.8% to 6.9%; OR: 0.6; CI: 0.4, 0.7), but the prevalence of iron deficiency anemia did not change significantly in children (1.5% compared with 1.2%; OR: 0.7; 95% CI: 0.4, 1.5) or women (4.9% compared with 4.1%; OR: 0.8; 95% CI: 0.6, 1.1). Folate deficiency with anemia declined significantly in women (from 4.1% to 0.5%; OR: 0.1; 95% CI: 0.1, 0.2), but logistic regression models and standardization indicated that none of the known possible anemia causes could account for the decline in total anemia in children or women. The prevalence of anemia declined significantly among US women and children between 1988-1994 and 1999-2002, but this decline was not associated with changes in iron or folate deficiency, inflammation, or high blood lead.