Whether, and to what degree, dietary calcium is inversely associated with blood lead levels was examined in 2,926 black and white children, 1 to 11 years of age, from the Second National Health and Nutrition Examination Survey (NHANES II). Blood lead levels were significantly higher in black than in white children, whereas white children had significantly higher dietary calcium intake (P less than .0001). Using covariance analysis and multiple regression analysis, determinants of blood lead levels were assessed as follows. The dependent variable was log10 lead, with independent variables age, sex, race, height, Quetelet index, dietary intake of calcium, phosphorous, fat, carbohydrate, and calories, community size index, poverty index ratio, geographic region, urbanization index, and all two-way interactions of the preceding. In the multiple regression analysis, the following independent variables were significant, P less than .05. Race (black) and sex (male) were positively associated with blood lead level (P less than .0001 for both). The lower the family income and the more urban the family residence, the higher were the blood lead levels (P less than .0001, less than .008, respectively). A significant independent inverse association of blood lead levels with year of examination was noted, reflecting a downward secular trend in blood lead levels. Height was inversely associated with blood lead level (P less than .0001). Dietary calcium intake was also inversely associated with blood lead level (P = .028). Dietary intake of phosphorous, fat, carbohydrate, and total calories were not significantly associated with blood lead levels. The most direct strategy for prevention of childhood lead poisoning involves primary prevention to reduce exposure.(ABSTRACT TRUNCATED AT 250 WORDS)