Annex: The Keneba pregnancy supplementation study.

An unmatched case-control study (with historical controls) was performed in a village in Gambia among all mothers of known parity who gave birth to a singleton live-born from April 1976 to May 1984 (with the exception of January to April 1978). The study tested the effect of dietary supplementation (biscuits and tea with a mean energy intake of 430 kcal per day) during pregnancy on birth weight and gestational age. Infant variables tested were birth weight gestational age and head circumference. Maternal variables were age parity height weight upper arm measurement and triceps skinfold measurement. The sample consisted of 182 historical control infants 197 cases and 187 mothers. A linear regression analysis of the data was performed to relate birth weight and head circumference to maternal anthropometry during pregnancy. A direct correlation was found between birth weight and maternal height weight in late pregnancy and weight gain in the third trimester. The correlation was weakened only slightly when gestational age was added to the regression. Maternal size rather than shape was found to be important to birth weight. Maternal height as well as both mid- and late-pregnancy weight were significantly related to head circumference. Head circumference and birth weight were both influenced by maternal anthropometry. A World Health Organization (WHO) meta-analysis of these data (which are small and show only weak significance) found the maternal body mass index to be a better predictor of birth weight than maternal weight. Unlike the analysis reported in this study the WHO data do not make a seasonal adjustment to reflect the fact that the women are thinner and birth weight less during the rainy season. Earlier reports of this study demonstrated a positive effect of supplementation during the rainy season but not the dry season.