prev next front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |19 |20 |21 |22 |23 |24 |25 |26 |27 |28 |29 |30 |31 |review
We studied the relationship between cigarettes smoked per day and urine cotinine concentration among women. We limited this analysis to self-reported active smokers in order to minimize the contribution of active smokers falsely reporting that they had quit. We first used crude data to generate a "box-and-whisker plot" of corresponding values of cigarettes smoked per day and urine cotinine concentration.

We then used general linear models to quantify the amount of cotinine variability explained by self-reported cigarettes smoked per day. We used r2 as the measure of variability and adjusted for maternal age, education, body mass index, state of residence, passive tobacco exposure, and weeks of gestation.