Women who use oral contraceptives just prior to conception may be more likely than nonusers to experience adverse birth outcomes, according to a study of health records from the Canadian province of Saskatchewan.1 Those who used the pill within 30 days of their last menstrual period had elevated odds of preterm birth and low birth weight, conditions that are associated with infant morbidity and mortality.
The researchers examined records from Saskatchewan Health Databases, which contain information on 99% of the province's residents. They obtained data on physician services, hospital stays and prescription drug use in the year prior to giving birth for a random sample of 50% of women who had a pregnancy between 1997 and 2000. Pill users, who totaled 1,540 women, were grouped according to the time interval between when contraceptive pills were dispensed to them and the estimated date of their last menstrual period (0–30, 31–60 or 61–90 days). For comparison, four women who had not used the pill in the year prior to giving birth were matched to each pill user by age, parity, infant's year of birth and whether they had a singleton or a multiple birth. The researchers made note of all women's chronic disease status (determined by their use of certain prescription medications) and socioeconomic status (determined by their participation in an income assistance program in the year in which their baby was born).
Births were evenly distributed over the four study years, and there were no significant differences in characteristics among subgroups of pill users or between users and nonusers. Most women in each subgroup were aged 20–29 (68–69%), had more than one child (57–59%), had a singleton birth (94–96%), had no chronic disease (86–90%) and were not receiving public assistance (85–89%).
Women's records were examined for the following adverse birth outcomes: very low birth weight (less than 1,500 g), low birth weight (less than 2,500 g), high birth weight (at least 4,000 g), very preterm delivery (less than 32 weeks' gestation), preterm delivery (less than 37 weeks' gestation) and postterm delivery (at least 42 weeks' gestation). The overall proportion of women experiencing an adverse birth outcome was higher among pill users than among women in control groups, but rates were highest among those who took the pill in the month prior to their last menstrual period. Among infants born to women who had used the pill in the month before their last menstrual period, 2% had a very low birth weight, 8% a low birth weight and 16% a high birth weight; 2% were born very early, while 10% were preterm and 3% were born late. Infants in the corresponding control group experienced lower rates of each of these outcomes.
In an analysis that adjusted for women's socioeconomic and chronic disease status, pill use within 30 days of the last menstrual period was positively associated with the occurrence of very low and low birth weight (odds ratios, 3.2 and 1.9, respectively) and preterm delivery (1.6). Use 2–3 months prior to the last menstrual period was not associated with adverse birth outcomes.
The researchers note that their data set is subject to certain limitations. The health records do not contain data on prescription drug compliance, so it is impossible to know whether women actually took the pills they obtained from the pharmacy. Furthermore, the records do not include information on body mass index, or on smoking or alcohol consumption during pregnancy—all factors that could contribute to adverse birth outcomes.
The investigators point out several possible explanations for the association between hormonal contraceptive use and adverse birth outcomes, including the presence of ingredients in the drug that might inhibit fetal growth and potential links between the birth outcomes and unplanned pregnancy or poor contraceptive compliance. While advocating continued research on the mechanisms by which pill use might lead to adverse birth outcomes, as well as on factors that might confound that association, the researchers encourage women to avoid the possible negative effects of the pill on their infants by considering "the use of barrier methods for [one] month after discontinuing oral contraceptive use." Clinicians, in turn, should practice "increased surveillance of pregnancies in women taking oral contraceptives near the time of conception."—H. Ball
1. Chen X-K et al., Recent oral contraceptive use and adverse birth outcomes, European Journal of Obstetrics & Gynecology and Reproductive Biology, 2009 (forthcoming).