Oral contraceptive use was not associated with mortality
ACP J Club. 1994 Sept-Oct;121:54. doi:10.7326/ACPJC-1994-121-2-054
Colditz GA for the Nurses' Health Study Research Group. Oral contraceptive use and mortality during 12 years of follow-up: the Nurses' Health Study. Ann Intern Med. 1994 May 15;120:821-6.
To examine the risk for mortality from the use of oral contraceptives.
12-year cohort analysis study of women in the Nurses' Health Study who responded to a mailed survey.
A cohort of 116 755 women who were registered nurses between 30 and 55 years of age completed questionnaires in 1967 and every 2 years thereafter until 1988. Participants were categorized according to reported use of oral contraceptives and grouped as ever-users or never-users. Women who reported angina, myocardial infarction, stroke, or cancer at baseline were excluded.
Assessment of risk factors
Ever-users were classified as current or past users. Age, smoking status, body mass index (BMI), cancer, cardiovascular disease, and duration of contraceptive use were assessed.
Main outcome measures
All-cause mortality between 1976 and 1988. Deaths were subgrouped by cardiovascular disease, cancer, accidental death and suicide, and other deaths. Classification of cause of death was determined by physician review of death certificates and pertinent medical records.
No difference was observed between ever-users and never-users for total mortality. The age-adjusted relative risk (RR) for ever-users was 0.99 (95% CI 0.91 to 1.07). When smoking status and BMI were controlled for, the strength of association between mortality and contraceptive use was further reduced (RR 0.93, CI 0.85 to 1.01). No difference was observed between ever-users and never-users for death from cardiovascular disease or total cancer mortality (RR 0.92, CI 0.81 to 1.03). A slight decrease in the risk for death from endometrial uterine cancer was observed among ever-users (RR 0.33, CI 0.10 to 1.11). Compared with never-users, women currently using oral contraceptives in 1976 had an increased risk for breast cancer mortality (RR 1.63, CI 1.07 to 2.49) but were not at increased risk for total cancer mortality. Women who had used oral contraceptives for ≥ 10 years were not at increased or decreased risk for mortality.
Oral contraceptive use was not associated with total mortality. An association existed between users of oral contraceptives in 1976 and breast cancer mortality during the ensuing 12 years.
Source of funding: National Institutes of Health.
For article reprint: Dr. G.A. Colditz, Channing Laboratory, 180 Longwood Avenue, Boston, MA 02115-5899, USA. FAX 617-432-0335.
Oral contraceptives, known to millions simply as "the pill," are one of the most frequently prescribed medications. Despite the many data on the safety and efficacy of the pill, some concerns remain about the long-term safety of oral contraceptives; the Harvard Nurses' Health Study looked broadly at mortality as the end point, and the Royal College of General Practitioners' Study looked more narrowly at stroke as the end point.
The Harvard Nurses' Health Study is an important contribution to the literature on oral contraceptive safety because of the large cohort of women followed and the 12-year duration of the study. Based on analysis of over 2800 deaths, the study concludes that the use of the pill does not increase the risk for overall mortality. Women who use the pill were not at increased risk for total cancer mortality, even though an increased risk for breast cancer mortality existed (caused, in part, by the inverse relation of ovarian and endometrial cancer with oral contraceptive use).
The increased risk for breast cancer (primarily in current users) persisted even when adjustments were made for the usual risk factors. Length of lactation, however, was not reported to be assessed as a risk. Evidence that an adequate length of lactation reduces the risk for premenopausal breast cancer is emerging (1). The results of a study in the United Kingdom suggest that the increased risk for breast cancer in women using the pill may be related to the dose of estrogen (2).
The major limitation to applying the results of this study to current practice is that the cohort of women assembled for the study in 1976 used the oral contraceptives on the market at the time. Therefore, the mortality associations largely reflect the use of pills at a much higher dose. It is only speculation, but the mortality predictions based on the newer low-dose oral contraceptives may actually be favorable because of neutral effects on breast cancer risk and improvement of cardiovascular risk.
Since the 1970s, a significant decline in the incidence of stroke and cardiovascular mortality has occurred (3). The study by Hannaford and colleagues found that social class, smoking status, and hypertension were important risk factors for stroke in women. Unfortunately, other recognized risk factors for stroke, such as hyperlipidemia, atrial fibrillation, alcohol use, and diabetes mellitus, were not controlled for in this case-control analysis. Persons who were nonsmoking former users of the pill did not have an elevated risk for stroke. Most women who had strokes while using oral contraceptives were using the older, higher-dose pills.
Women who are over 35 years of age and smoke should not use oral contraceptives. Otherwise, women should be reassured about the overall safety of the pill and should also be counseled on the potential noncontraceptive benefits of the pill, such as the decrease in the incidence of ovarian cancer.
Holly L. Thacker, MD
Cleveland Clinic FoundationCleveland, Ohio, USA