Interactions between growth and nutrient status in school-age children of urban Bangladesh
Ahmed, F.; Barua, S.; Mohiduzzaman, M.; Shaheen, N.; Bhuyan, M.A.; Margetts, B.M.; Jackson, A.A.
American Journal of Clinical Nutrition 58(3): 334-338
1993
ISSN/ISBN: 0002-9165 PMID: 8237842 DOI: 10.1093/ajcn/58.3.334Document Number: 336310
Between February and March 1990, health workers interviewed and took anthropometric measurements and blood samples from 242 children 5-12 years old, attending 3 primary schools in affluent areas and 2 primary schools in poorer areas around a university campus in Dhaka, Bangladesh. Researchers wanted to study the interaction between anthropometric and biochemical measures of nutritional status in seemingly healthy school children. As the height-for-age increased, so did the mean serum levels of protein, hemoglobin, and vitamin A (p = 0.001 for protein and 0.01 for hemoglobin and vitamin A). The rising trends were strongest for hemoglobin and vitamin A. Children in the group with the highest serum vitamin A level ( 1.05 mcmol/L) had significantly higher hemoglobin levels than did those with lower vitamin A levels (adjusted, 138.9 g/L vs. 133.8 g/L for 0.7-1.07 mcmol/L and 132.8 g/L for 0.7 mcmol/L; p = 0.002). For every unit change in vitamin A, a 4.92 unit change in hemoglobin existed. Children who had the lowest serum vitamin A levels had the highest serum copper levels (22.8 mcmol/L vs. 22.3 mcmol/L middle vitamin A group and 19.8 mcmol/L highest vitamin A group; p = 0.05). Multiple regression analysis showed that family income, age, weight-for-age, and hemoglobin and serum copper levels were significantly associated with serum vitamin A levels. These findings reveal that short children who were light for their age had lower serum vitamin A and hemoglobin levels and higher serum copper levels than their taller and heavier counterparts. They demonstrate a complex interaction between serum levels of biochemical indexes of nutritional status and other anthropometric, biochemical, and sociodemographic variables.