The Effectiveness of the Yuzpe Regimen of Emergency Contraception. (ARTICLES).

Emergency contraception prevents pregnancy after unprotected sexual intercourse. If widely used, it could substantially reduce the number of unintended pregnancies that occur each year in the United States. (1) Emergency contraceptives available in the United States include regular oral contraceptive pills containing the hormones estrogen and progestin, less common birth control pills called minipills that contain progestin only, and the copper-T IUD.

The purpose of this article is to examine rigorously the efficacy of one method of emergency contraception: the Yuzpe method. The Yuzpe regimen has replaced the older postcoital therapy of high doses of the estrogen diethylstilbestrol, primarily because the Yuzpe regimen has fewer side effects. It involves taking two doses of pills containing a combination of estrogen and progestin, with each dose containing 100 mcg of ethinyl estradiol and 1.0 mg of norgestrel. (2) One dose is taken within 72 hours after unprotected coitus, and the second is taken 12 hours later. The total dosage is therefore 200 mcg of ethinyl estradiol and 2.0 mg of norgestrel (equivalent to 1.0 mg of levonorgestrel). (*) In this article, we use the terms Yuzpe regimen, Yuzpe method and emergency contraceptive pills (ECPs) interchangeably.

Investigators have often evaluated the efficacy of the Yuzpe method by calculating its failure rate (pregnancies divided by the number of women treated), which we consider an inappropriate measure. In this article, we discuss why the failure rate is a poor measure of ECP efficacy and present estimates of a better measure, the effectiveness rate (the proportionate reduction, due to treatment, in the risk of pregnancy). We then discuss four methodological issues involved in measuring effectiveness--the appropriateness of pooling data across studies, the effects of loss to follow-up and protocol violations, and the accuracy of the expected number of pregnancies reported in the clinical trials we review.

Data

This research updates and extends an earlier analysis (3) in which we challenged the conclusion of Silvestre, Bouali and Ulmann, (4) based on five clinical trials, that the Yuzpe regimen may not be effective. Here, we review in depth all published results from clinical trials of the Yuzpe method in which the data needed to calculate effectiveness rates were collected and reported. Summary data on the 10 trials meeting this criterion (5) are reported in Table (1).

The criteria by which subjects were selected differed somewhat across trials. All but one trial required women to have experienced an act of unprotected intercourse within the previous 72 hours; the sample in that trial was limited to women who had had unprotected intercourse within the previous 48 hours. (6) Trials also differed in whether they included women with more than one unprotected act within the past 72 hours, more than one unprotected act since the last menses or more than one act (protected or unprotected) since the last menses. Some trials required that women abstain or use condoms until the results of the therapy could be verified. In all trials, most subjects were young and had never been pregnant (at least in the trials reporting this characteristic).

The fraction of treated women subsequently lost to follow-up varied from 0.0% to 22.7%. Several investigators reported that they were confident that women lost to follow-up were not pregnant because they would have heard of failures from colleagues (7) or because they believed pregnant women would have returned to the same clinics for management of their pregnancy. (8)