• emergency contraception

Congratulations to Drs McKay and Gilbert on trying to increase access to emergency contraception (EC) intrauterine device (IUD) and on achieving high rates.1 Even in Liverpool, UK where we pride ourselves on easy, often immediate, IUD access and where we have long been promoting its effectiveness we only achieve around 5%.

We have some concerns about the algorithm described in this article.1 The classification of level of risks is not based on what is known about variability of ovulation timing. The chance of pregnancy is greater than 10% from Day 6 to Day 21 of the cycle, and by the fifth week women still have a 4–6% chance of ovulating.2 Contrary to previous teaching, only 10% of women with a 28-day cycle will ovulate 14 days before their next bleed.2 Previous correspondence shows that when hormonal assessments were carried out, women confident of the date of their last menstrual period (LMP) and cycle length were not where they thought they were in their cycle,3 as Drs McKay and Gilbert hint in their article. So, calculations based only on LMP and cycle length are not accurate enough to inform an individual patient not wishing an unplanned pregnancy. The pregnancies recorded during the pilot study,1 in women not assessed as high risk who had levonorgestrel (LNG) EC, support the above.

Regarding IUD eligibility, it is only necessary to elucidate pre-insertion the possibility of an implanted pregnancy, should there have been any unprotected sexual intercourse (UPSI) more than 5 days prior to presentation. The method Drs McKay and Gilbert have used is useful when there have been many episodes of UPSI but before a pregnancy test would be positive. It is restrictive, but can reassure the IUD fitter that there will not be an implanted pregnancy. It does not determine the date of ovulation of the current cycle; it just estimates the earliest possible date.

Evidence-based practice suggests that wherever women are in their cycle, if all UPSI was within 72 hours, the IUD is the most effective, and ulipristal acetate (UPA) has the edge on LNG, because of its action during the luteinising hormone (LH) surge (closer to ovulation: the highest risk time of the cycle). For UPSI between 72 and 120 hours the IUD is the most effective, UPA is licensed and LNG has neither been shown to have an effect after 96 hours nor is it licensed. Only if there is UPSI more than 120 hours previously is it necessary to consider implantation with all the caveats above.

However, the Faculty of Sexual & Reproductive Healthcare does not rank the methods with regards to choice, it states: “Health professionals should discuss individual need to … [EC] and inform women about the different methods with regard to efficacy, adverse effects, interactions, medical eligibility and need for individual contraceptive precautions”.4

It is for the woman then to make an informed choice and other factors may come into play, as previous work has shown,5 and there is no evidence of a wholesale move towards UPA when free choice is introduced. The offer of an IUD to women seeking EC has been UK standard practice for over 30 years; however, the introduction of an algorithm seems to have been helpful within Drs McKay and Gilbert's service.

Restricting UPA to those aged over 18 years appears to imply that women aged under 18 years are less deserving of protection against unplanned pregnancy. The studies on UPA did include some 16- and 17-year-olds, and subsequent studies have shown that adolescents are no different to adults.6 If the authors were being consistent then they would not prescribe most contraceptives to teenagers as they are seldom included in most studies. We would argue that it is precisely the young, fertile woman presenting for EC who needs the most effective method, and though an IUD is possible, these individuals are often the ones most reluctant to accept it. The FSRH guidance explicitly supports the use of UPA in under-18s.