The role of body weight in oral contraceptive failure.

摘要

Each year, three million of the six million pregnancies in the United States are classified as unintended (i.e. unwanted or wanted, but at a later time). It is estimated that nearly half of these unintended pregnancies occur in the 90% of women who use some type of contraceptive. Though researchers attribute these contraceptive failures to noncompliance and ineffective use, few studies have investigated whether biologic factors, such as body weight, are involved. Previous literature on the role of body weight in contraceptive failure is sparse and mainly limited to secondary analyses of efficacy trials without proper control of confounding.; Information from several large, national datasets was used to investigate whether heavier body weight is associated with an increased risk of oral contraceptive (OC) failure. The datasets used and types of studies conducted include: 1995 National Survey of Family Growth (NSFG) for a retrospective cohort study, 1999 multistate Pregnancy Risk Assessment Monitoring System (PRAMS) for a case-control study, and 1999 Behavioral Risk Factor Surveillance System (BRFSS) from South Carolina in conjunction with 2000 PRAMS from South Carolina for a case-cohort study. However, these datasets do not collect routine information on adherence or frequency of intercourse---two variables that may be important confounders of the body weight-OC failure association. To fill in these gaps, the Contraceptive History, Initiation, and Choice (CHIC) Study, a prospective cohort study of contracepting women, was conducted at a family medicine practice in Atlanta.; In these studies, obesity was associated with a statistically significant increased risk of OC failure. However, this risk was attenuated and no longer statistically significant after adjustment for confounders. Results from the CHIC Study did not indicate any clear association between body weight and adherence to an OC regimen or frequency of sexual intercourse. When body weight-OC failure associations were externally adjusted for these two potential confounders, the associations remained largely unchanged. Larger, prospective studies of the body weight-OC failure association are needed to determine if heavier women should be advised to use a method other than OCs to prevent pregnancy.