Mifepristone beats levonorgestrel on the outcome that counts

Design

Cochrane systematic review and meta-analysis of randomised trials, GRADE assessed

Population

87 trials, about 36,000 women seeking emergency contraception within five days of unprotected intercourse

Primary outcome

Number of pregnancies

Effect

Low-dose mifepristone vs levonorgestrel RR 0.73 (95% CI 0.59 to 0.90), high certainty; mid-dose RR 0.67 (0.49 to 0.91), moderate certainty

Eighty-seven randomised trials and about 36,000 women were pooled to compare mifepristone with the other oral emergency contraceptives. Low-dose mifepristone, under 25 mg, reduced pregnancies against levonorgestrel with high-certainty evidence, and mid-dose 25 to 50 mg did the same with moderate certainty. Side effects were consistently fewer with mifepristone. Against the old Yuzpe regimen the gap was wider still.

The trade-off is menstrual. Mifepristone makes early menses less likely and delayed menses more likely, and the delay scales with dose. Women who take it and then wait for a period that does not arrive on time will assume the method failed, and the higher the dose the more often that happens.

Seventy-nine of the 87 trials came from China. The effectiveness estimates are robust, but the side-effect and satisfaction findings were gathered in one setting and one set of expectations. In India mifepristone is licensed and widely available for medical abortion; the emergency contraception market is levonorgestrel, and the over-the-counter pathway is built around it, so this evidence does not by itself change what a woman can buy.

• Ask the date of unprotected intercourse; the trials enrolled within five days.
• Warn explicitly that the next period may be late, and that late is expected rather than a sign of failure.
• Give a pregnancy-test instruction with a date on it, not 'if your period is late'.
• Where a copper IUD is acceptable and available, it remains the most effective option; the two trials comparing it with mifepristone were too small to judge.
• Record what was supplied and when, so a repeat request in the same cycle is handled on facts.

Why it matters

The drug most clinicians associate only with abortion care is the better emergency contraceptive.

Don't overread it

Effectiveness is well supported, but nearly all the evidence on side effects and satisfaction comes from Chinese trials.

The statistics, in plain English

Low-dose mifepristone gave RR 0.73 (95% CI 0.59 to 0.90) for pregnancy against levonorgestrel, with I squared 0 per cent across 14 studies: the trials agreed with each other, and the interval sits entirely below 1.0, so a real reduction is the honest reading. Mid-dose gave RR 0.67 (0.49 to 0.91), wider because fewer women contributed. For delayed menses the direction reverses, RR 1.52 (1.11 to 2.08) low-dose and 1.29 (1.01 to 1.64) mid-dose; the mid-dose interval nearly touches 1.0, so that estimate is the softer of the two.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.