The social gradient in birthweight at term: quantification of the mediating role of maternal smoking and body mass index.

BACKGROUND Maternal education is associated with the birthweight of offspring. We sought to quantify the role of maternal body mass index (BMI) and smoking as intermediary variables between maternal education and birthweight at term. METHODS We examined the association between maternal education, BMI, smoking and offspring's birthweight among women who delivered at term in the Danish National Birth Cohort (n = 75,085). RESULTS Compared with mothers with more than 12 years of education, women with 10-12 years of education had babies that were 12 (4-19) g lighter. Mothers with 9 years of education had babies that were 51 (95% CI; 39-62) g lighter. BMI and smoking affected the association between maternal education and birthweight, albeit in different directions. If all mothers had the BMI of the highest educated mothers, the deficits would be larger: -20 (-22 to -19) and -74 (-80 to -68) g, respectively. If all mothers smoked like the highest educated mothers, mothers with a shorter education would have the heaviest babies: the difference would be 9 (2-16) and 23 (11-36) g, respectively. In combination, smoking and BMI all but explained the educational gradient in birthweight at term. CONCLUSION Maternal smoking and BMI are important intermediates of the educational gradient in birthweight at term. As the prevalence of smoking is dropping and the prevalence of obesity is increasing the educational gradient will likely reverse, but it seems unlikely that this will translate into a health advantage for the children of the least educated mothers.