Although clinical trials report fewer than 0.5 pregnancies per 100 women-years of use, pregnancy rates among patients who use oral contraceptives can be as high as seven per 100 women-years. An estimated 400,000 pregnancies occur in the United States annually among oral contraceptive users. Preliminary studies suggest that body mass index (BMI) could contribute to oral contraceptive failure through an impact on steroid hormone metabolism. To clarify the role of body weight Holt and colleagues studied women who became pregnant while using oral contraceptives.

The authors used clinical data from enrollees of a large health maintenance organization to identify women who tested positive for pregnancy within three months of filling an oral contraceptive prescription. The 546 women who agreed to participate in the study were interviewed. Those who had discontinued oral contraceptive use before becoming pregnant and those who reported missing five or more pills in the month of conception were excluded. The 248 women who conceived while using oral contraceptives were matched by age with the control group of 533 sexually active women who also filled prescriptions for oral contraceptives during the study. All participants were interviewed to obtain demographic, reproductive, medical, and lifestyle information. For the treatment group, prepregnancy weight and a detailed pattern of oral contraceptive use during the month of conception were confirmed during the interview.

The pregnancy rate for oral contraceptive users was calculated as 2.8 per 1,000 women-years. Women who became pregnant were more likely to be black, married, smokers, and to have had previous pregnancies. They were five times more likely to have previously conceived while using oral contraceptives. Women who became pregnant also had lower education levels and lower family income than women in the control group. The mean BMI for pregnant oral contraceptive users was 26.3, compared with 24.9 for women in the control group. In logistic analysis controlling for other significant factors, the risk of oral contraceptive failure increased dramatically in very obese women. For women with BMI greater than 27.3, the pregnancy risk was 60 percent greater than in women with BMI of 21.3 or less. The risk associated with BMI greater than 27.3 was even stronger when the analysis was restricted to women who consistently used oral contraceptives correctly. The authors conclude that the risk of oral contraceptive failure was significantly elevated in women with BMI of 27.3 or greater. In these obese women, the risk of pregnancy, even when using oral contraceptives consistently, was doubled. Biologic explanations for this observation include higher basal metabolic rate, induction of hepatic enzymes, and increased sequestration of lipophilic hormones in adipose tissue. The authors calculate that obesity could add two to four pregnancies per 100 woman-years of oral contraceptive use.