The primary purpose of the proposed study is to determine the effects of nutritional exposures present in women before and during pregnancy on infant birth weight for gestational age. It is postulated that maternal nutrition exposures before and during pregnancy, and changes in them over time, influence birth weight. Other lifestyle, biological, and psychosocial exposures may also influence birth weight, or they may interact with nutrition exposures. The effects of these exposures on birth weight and their interactions with the nutrition exposures will be assessed and accounted for in the analyses. Birth weight has been chosen as the primary outcome variable because it is the measure most closely associated with infant morbidity and mortality in the U.S. Nutritional factors have been chosen as the exposure variables because they are often modifiable. The long-term objective of the study is to increase the availability of information that could be applied toward the prevention of poor pregnancy outcomes. The study may contribute particularly useful information as it would be the first prospective, analytical investigation that accounts for the influence of prepregnancy exposure on birth weight. Enrollment of a large number of subjects needed prior to pregnancy is currently possible due to the collaborative arrangement established with a health maintenance organization that serves a large number of women of child bearing age who tend to plan their pregnancies. A system for repeated contacts with potential participants will be set up to ensure successful recruitment. The study design calls for enrolling a cohort of 1,000 eligible women over a 26 month period. A total of 700 deliveries are needed to test nutrition exposures - birth weight hypotheses. Following standardized protocols, repeated assessments of nutritional, biological, lifestyle, and psychosocial exposures would be undertaken before and during pregnancy. Prepregnancy data collection will be concentrated in a two-month period after enrollment to avoid a length bias in women who conceive relatively quickly. Women who fail to conceive within 12 months, and those who experience an early fetal loss would be exited from the study. Women who conceive and maintain a pregnancy for over 20 weeks will be followed through parturition. After controlling for confounders, the effects of the exposures on birth weight for gestational age would be determined by regression methods and by comparing exposed to unexposed women.
