Prenatal effects of maternal smoking on daughters' smoking: nicotine or testosterone exposure?

OBJECTIVES: The purpose of this study was to specify the effect of prenatal fetal exposure to maternal cotinine and testosterone on daughters' smoking in adolescence and adulthood.
METHODS: Longitudinal causal models were estimated among 240 White mother-daughter pairs from the Child Health and Development Study. Mothers and daughters were reinterviewed when daughters were aged 15 to 17 years, and daughters were interviewed at 27 to 30 years of age. Blood samples were obtained from both parents during pregnancy and from adult daughters.
RESULTS: Testosterone and smoking were positively correlated among mothers during their pregnancy and among adult daughters. Maternal prenatal cotinine had no direct effect on daughters' smoking; self-reported smoking in pregnancy did have a direct effect. Smoking among daughters during adolescence was determined by maternal prenatal testosterone and self-reported maternal smoking during pregnancy and postnatally. Smoking among adult daughters reflected chronic smoking since adolescence and the continuing effect of postnatal maternal smoking. Prenatal maternal testosterone affected adult daughters' testosterone.
CONCLUSIONS: Estimates of the impact of prenatal maternal smoking depend on the measure of smoking. Prenatal testosterone exposure is a previously unrecognized risk factor for smoking among female offspring.