“A new morning-after pill can be used up to five days after sex, longer than any protection against pregnancy developed so far,” the Daily Mail reported. It said a trial of the new drug found it prevents up to two-thirds of unwanted pregnancies if taken within 72 hours, and 50% if taken within five days.
This trial compared the effectiveness of two emergency contraceptives, EllaOne (ulipristal acetate) and Levonelle (levonorgestrel). Levonelle is the main oral emergency contraceptive in the UK and is available from pharmacists without a prescription. It is approved for use up to 72 hours (three days) after unprotected sex. EllaOne is approved for use up to 120 hours (five days), but is a prescription-only medicine, available through GPs.
The drugs appeared to be equally effective at preventing pregnancy if taken within 72 hours of having sex. The researchers also combined data from an earlier study to look at the effectiveness of this medication after 72 hours. Their research confirms the licensed indication for this drug, but because few women became pregnant in the studies, larger post-marketing studies will be needed to assess the new drug’s effectiveness at preventing pregnancy after 72 hours.
Protected sex is the best option, but if emergency contraception is needed, the sooner it is taken the better. EllaOne is at least as good as Levonelle in the first 72 hours after unprotected sex, but may also be used up to 120 hours (five days) after unprotected sex. The copper intrauterine device can also be used for up to five days.
Where did the story come from?
This research was carried out by Professor Anna F. Glasier from Edinburgh University, and other researchers from universities in the US. The research was carried out for HRA pharma, which makes a ulipristal acetate product called EllaOne. The study was published in the peer-reviewed medical journal The Lancet.
The newspapers focused on the potential social impact of an emergency contraceptive pill that can be taken five days after unprotected sex. The research supports neither the Daily Mail’s suggestion the pill is still 50% effective five days after sex, nor the Sun’s statement that it prevents 98% of pregnancies five days after sex.
What kind of research was this?
This is a randomised non-inferiority trial, a type of randomised controlled trial that assesses whether a new drug is at least as good as an existing treatment. This study tested whether ulipristal acetate, a contraceptive pill, was as effective as levonorgestrel as emergency contraception when taken within 72 hours of unprotected sex. Levonorgestrel is the main oral emergency contraceptive licensed in the UK, but is not effective when taken more than 72 hours after unprotected sex.
The researchers also carried out a meta-analysis comparing the effectiveness of ulipristal acetate with levonorgestrel, using this data and data from a previous study.
What did the research involve?
In the non-inferiority trial, the researchers recruited 2,221 women from 35 family planning centres in the UK, Ireland and the US. The women had to be older than 16 (UK) or 18 (US) with regular menstrual cycles, and seeking emergency contraception up to five days after having had unprotected sex. The study excluded women who were pregnant, breastfeeding, sterilised, taking the contraceptive pill, fitted with an intrauterine contraceptive device, or whose partner had been sterilised.
Where suitable, women who presented to the clinic 72 or more hours after sex were initially offered an intrauterine device, which can be used as emergency contraception within five days after having sex.
All women had a pregnancy test and a blood sample taken when they arrived at the clinic. They were then grouped into those who had sex less than 72 hours beforehand, and those who had sex between 72 and 120 hours before. Within these two groups, the women were randomly assigned to receive either ulipristal acetate or levonorgestral.
After receiving either pill, the women were asked to keep a diary of their sexual activity, use of contraception, vaginal bleeding, whether they were on any other medication, and any side effects they experienced.
The women were followed up until five to seven days after their next expected period. If their period was delayed, they were followed and given routine pregnancy tests.
The study was designed to examine differences in pregnancy rates in women who took the two treatments within 72 hours of unprotected sex, and this was the primary outcome of the study. A secondary outcome was pregnancy rates in the much smaller number of women who took the emergency contraceptives after the recommended time of 72 hours.
What were the basic results?
There were 1,696 women who received emergency contraception within 72 hours of having had sex. There were 15 pregnancies out of the 844 women in the ulpristal acetate group, and 22 out of 852 in the levonorgestrel group. There was no difference between the two types of pill at preventing pregnancy when given within 72 hours (odds ratio 0.68, 95% confidence interval 0.35 to 1.31).
There were 203 women who used emergency contraceptive pills between 72 and 120 hours. Three women out of 106 in the levonorgestrel group became pregnant. There were no pregnancies among the 97 women who received ulipristal acetate.
Similar numbers of women reported side effects with both drugs. In both cases, 94% of these side effects were mild or moderate.
The researchers also compared and combined this data with data from a 2006 randomised controlled trial, which had compared ulipristal acetate to levonorgestrel in 1,546 women. There were some differences between the studies, including the women in the non-inferiority trial being on average younger, with a higher body mass index and more likely to have waited longer before taking emergency contraception (39.7 hours vs 35.3 hours). These differences were adjusted for in the analyses.
When the two trials were combined, there were more pregnancies in the levonorgestrel group than in the ulipristal acetate group when the data was analysed in groups that took the pill within 24 hours, within 72 hours and within 120 hours. The difference was of borderline significance for the within-72 hour group. Odds ratios for the within-24 hours group was 0.35 OR, 95% CI 0.11 to 0.93, for the within-72 hours group: 0.58 OR, 95% CI 0.33 to 0.99. For the within-120 hours group it was 0.55 OR, 0.32 to 0.93).
How did the researchers interpret the results?
The researchers concluded that “ulipristal acetate provides women and healthcare providers with an alternative choice for emergency contraception that can be used up to five days after unprotected sexual intercourse”.
They suggest that neither study alone had enough participants to assess a statistical difference between levonorgestrel and ulipristal acetate. However, combining the two indicated that ulipristal acetate prevented more pregnancies than levonorgestrel when taken within 24 hours or 120 hours after unprotected sexual intercourse.
They also commented that levonorgestrel prevented less pregnancies in their study than they were expecting based on the World Health Organisation’s (WHO) estimations of effectiveness when taken at increasing lengths of time after unprotected sex.
This study shows that ulipristal acetate is at least as effective as levonorgestrel as an emergency contraception pill when taken within 72 hours of unprotected intercourse. The researchers combined their data with data from a previous study to test whether ulipristal acetate was more effective than levonorgestrel when taken up to 120 hours after unprotected sex and found there were fewer pregnancies in the ulipristal acetate group.
The study has some limitations, some of which the researchers highlight:
- This study was conducted at family planning centres. However, emergency contraception is available at pharmacists without prescription. The study also excluded women taking the contraceptive pill, but emergency contraception is frequently used by women who have missed taking a pill. As such, the women in this study may not be fully representative of these other populations.
- The WHO states that the effectiveness of levonorgestrel decreases with time. As such, it recommends that women take the pill at the earliest opportunity. The researchers in this study did not examine how fast the effectiveness of ulipristal acetate reduces.
- The drugs’ effectiveness at preventing pregnancy when taken more than 72 hours after sex was only a secondary outcome of this study. Also, as it involved only three unwanted pregnancies, it will be important to test how well the drug works after 72 hours in trials designed to look at its use up to 120 hours.
- Although the researchers combined data with an earlier study to look at the effectiveness of the emergency contraceptives on a larger population, further primary research would be needed to confirm these findings.
Levonorgestrel is currently the main oral medication licensed in the UK for use as an emergency contraceptive. Women should be assured that it is highly effective if taken correctly and within 72 hours (three days) of unprotected intercourse. The earlier the pill is taken, the more effective it is. The optimal time for taking the pill is within 12 hours.
Ulipristal acetate has recently been licensed for use in the UK, and will be marketed for women within five days of unprotected intercourse. The current alternative is a copper intrauterine device, which can be used up to five days after unprotected intercourse. More research of ulipristal acetate compared to these devices can be expected.
Links to the headlines
New morning after pill ‘can be taken up to five days after sex’. The Daily Telegraph, January 29 2010
Morning-after pill that works five days later. Daily Mail, January 29 2010
New-morning after drug is abortion pill, campaigners say. The Times, January 29 2010
Pill that can halve birth risk. Daily Express, January 29 2010
Links to the science
Glasier AF, Cameron ST, Fine PM et al. Ulipristal acetate versus levonorgestrel for emergency contraception: a randomised non-inferiority trial and meta-analysis.The Lancet, Early Online Publication, 29 January 2010