Grandmothers' smoking in pregnancy and grandchildren's birth weight: comparisons by grandmother birth cohort.

This study examined whether grandmothers' smoking behavior during pregnancy was associated with birth weights in their grandchildren, considering possible birth cohort effects in the grandmothers' generation. The birth weights of 935 singleton children were compared by grandmothers' and mothers' smoking status during pregnancy. In 2008, women (n = 397) from the Michigan Bone Health and Metabolism Study were interviewed about their own birth history, including whether their own mother smoked while pregnant with them, and the birth histories of their offspring. While also accounting for family clustering, linear mixed models were used to evaluate whether birth weight differences in the grandchildren were associated with grandmothers' and mothers' smoking behavior during pregnancy. Associations were compared among grandmothers born from 1904 to 1928 versus grandmothers born from 1929 to 1945 to determine potential birth cohort effects. Forty-six (5 %) grandchildren had grandmothers and mothers who smoked while pregnant, while 455 (49 %) had grandmothers and mothers who did not smoke during pregnancy. After adjustment, birth weight was an average of 346 (95 % confidence interval 64-628) grams higher in grandchildren whose grandmother and mother both smoked during pregnancy relative to grandchildren whose grandmother and mother both did not smoke during pregnancy, but only among grandmothers who were born from 1929 to 1945. For grandmothers born from 1904 to 1928, grandchildren birth weights did not differ by grandmother and mother smoking status. Birth weight may be associated with grandmother and mother smoking behaviors during pregnancy, but birth cohort effects should be considered.