We used linear models to generate mean adjusted birth weights for women with different levels of tobacco exposure, first using cigarettes per day, and then using urine cotinine concentration. We adjusted for maternal age, race, education, parity, body mass index, Women, Infants, and Children (WIC) enrollment, caffeine and alcohol consumption, marital status, state of residence, infant's sex and gestational age.

We then plotted adjusted birth weight against cigarettes per day and against urine cotinine concentration in order to learn more about how birth weight is affected by level of tobacco exposure.