Modeling the relationship of cotinine and self-reported measures of maternal smoking during pregnancy: a deterministic approach.

Studies of effects of prenatal exposure to cigarettes frequently acquire both self-report and biological assays of maternal smoking. However, little attention has been paid to methods for combining information from both sources to enhance the precision of exposure measurement. This paper analyzes the relationship between the two commonly used measures of smoking exposure during pregnancy: Maternal self-report and urinary cotinine. We present a deterministic method for combining the two measures and examine its robustness under different assumptions. We apply the method to a dataset from the Family Health and Development Project. In addition, we propose an approach for calibrating the self-report measures for individual women based on both sources of information. Enhancing the quality of exposure measurement may substantially advance studies of the teratological effects of exposure on offspring.