Data collection in pregnancy (AIP study)

The pregnant women were interviewed, usually at home, before 24 weeks of gestational age. A standardized questionnaire included information on demographic and household characteristics including marital status, family income, health risk factors, medical conditions, and obstetric history. Pregnancy outcome data, which included prenatal, labor, and delivery information, including information of the newborn, was abstracted from medical records. A postpartum interview was conducted in the hospital or by telephone within 1 month after delivery.

Information on folic acid, iron, and vitamin use was obtained before 24 weeks of gestation from the following questions in the prenatal exposure questionnaire: “Have you used any of the following vitamin or mineral supplements: prenatal supplement vitamins, multivitamin, vitamin A, vitamin C, vitamin E, iron/ferrous sulphate, folic acid/folate, calcium, or other; specify.” If a respondent answered yes, she was specifically asked how often each item had been used (not at all, once a month, 2-3 times a month, twice a week, 3-4 times a week, 5-6 times a week, once a day, or 2 or more times a day). This information was collected for the month before conception through the third month of pregnancy.

Folic acid exposure

We collected information on folic acid content (micrograms) in each of the self-reported vitamin supplements. Using the detailed frequency information from the pregnancy questionnaire, we could calculate mean daily folic acid intake. Prenatal vitamins were estimated to contain 800 μg folic acid per tablet, whereas vitamin supplements were estimated to contain 400 μg folic acid per tablet.

A dichotomous variable was created to characterize users from nonusers. Mean daily intake was calculated for each month from the month before pregnancy through the third month, and mean folic acid intake in the first trimester was defined as the average daily intake over these 4 months. For each month and for the total first trimester, daily folic acid intake was also divided into 4 categories (0, <400 μg, 400-800 μg, >800 μg). The majority of women (n = 1457) were enrolled in the study after starting their third month of pregnancy and reported all 3 months of folic acid intake directly. Because 85% of the women with complete information reported the same intake in months 2 and 3, women who were enrolled in the study and interviewed in their second month of pregnancy (n = 42) were assumed to have the same folic acid intake in the third month as in the second.

Follow-up of the children at 6 years of age (PRAM study)

Asthma in the 6 year old children was assessed by asking the mother the following questions: “Has the child ever been diagnosed by a doctor or health professional as having asthma?” and “Has your child had wheezing or whistling in the chest in the last 12 months?” A positive answer to both these questions was considered a positive definition of current asthma.

Study ethics

The Human Investigation Committee of Yale University Medical School (New Haven, CT) approved the study, and all respondents provided written informed consent prior to their participation.

Statistical analysis

Logistic regression was used to assess the association between folic acid supplementation in pregnancy and asthma in the children and expressed the effect estimates as odds ratios with 95% confidence intervals (CIs). Information on potential confounding variables was obtained from the interviews conducted during early pregnancy and at 6 years (± 3 months) of age. They included maternal parity, ethnicity, marital status, household income, maternal asthma, smoking during pregnancy, use of other vitamins (C, D, and E), iron use, and calcium use in the first trimester.

Group differences in folic acid supplementation were initially identified using of unpaired Student t tests if the variables were dichotomized and by analysis of variance with Bonferroni and Scheffes post hoc tests when they were categorized in more than 2 groups. Covariables were identified as potential confounders and included in the analyses if they were associated with both the exposure of daily folic acid supplementation and outcome of asthma at 6 years of age, at a level of P < .10 Adjusted models used a backward elimination procedure, which retained only covariates that resulted in 10% or greater change in the effect estimate. The final model included maternal marital status, family income, and maternal asthma.

The effect of folic acid supplementation was first assessed for the continuous variable per 100 μg increase in daily intake and successively for the 4 categories of intake (no use, <400 μg, 400-800 μg, and >800 μg daily use). All analyses were done separately for 4 different time periods (the month before pregnancy and during the first, second, and third months). In these categorical analyses, no use of folic acid was used as the reference, and 2-sided P values from linear trend tests were calculated by treating the folic acid dose categories as ordinal variables in the regression model.

P < .05 and CIs that excluded the null value of 1 were considered statistically significant. Statistical analyses were performed with STATA/SE10 (STATA Corp, College Station, TX).