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 |review
Conditional logistic regression models were developed for all risk factors of interest and their potential confounding variables.

Potential cofounders included:

  • Maternal level of education
  • Maternal/paternal hours of passive smoke
  • Maternal/paternal time reviewing exposure lists
  • Per capita income