IVF success rates vary greatly between fertility clinics. In the UK, it is a statutory duty to report all IVF cycle outcomes. In the US, it’s a regulatory obligation. Clinics in the UK inform the Human Fertilisation & Embryology Authority (HFEA). Similarly, in the US, clinics report to the Centres for Disease Control (CDC). Both the HFEA and the CDC publish patient information on their websites, allowing patients to compare clinics.
In 2017 there were over 75,000 IVF cycles in the UK. However, despite the existence of the NHS, fertility treatment in the UK is not routinely free at the point of use. The National Institute for Health & Care Excellence (NICE) produces guidelines regarding IVF provision, but unfortunately they are not followed by the majority of Clinical Commissioning Groups (CCGs). This has led to a postcode lottery and created a specialised niche service industry which responds to the demands of self-funding patients.
There is intense competition between providers in the fertility industry. Therefore, it is important for patients to understand both how success rates are created and how to interpret the individual data.
How is the data collected?
- Fertility clinics in the UK register everyone having treatment as a patient or a partner. In addition, they include relevent personal information such as name, age, gender, diagnosis, etc. To do this, they use the HFEA’s Electronic Data Exchange (EDI), which they access through their individual portal.
2. When a patient starts IVF treatment, the clinic submits an Intention To Treat Form, to the HFEA within 3 days of starting the ovarian stimulation, which indicates an intention to collect eggs.
3. At the end of the cycle, the clinic submits a treatment form, which informs the HFEA about the date of the egg collection. It confirms the number of eggs collected and whether the eggs are to be frozen or used to create embryos. If the latter, extra information is submitted about IVF, ICSI, donor sperm and the date and number of embryos transferred. Sometimes all the embryos are frozen at this stage, which is noted as ‘freeze-all’.
4. If a patient has an embryo transfer (ET) or insemination with donor sperm, the clinic must later submit an Early Outcome Form. This documents whether the cycle has been positive or negative. It’s important to note that for the cycle to be positive, a heart beat must be seen on ultrasound scan.
5. Lastly, once the outcome of the pregnancy is known, a Pregnancy Outcome Form is submitted.
Inevitably, there are sometimes errors with the data input, and so the final figures must be audited by the HFEA and validated by the clinic before they are published by the HFEA.
Why is the data collected?
The HFEA is a government body, made up of predominantly lay people, which was set up in 1991. Their remit is to regulate the use of eggs, sperm and embryos. In the UK, IVF clinics are obliged by law to provide information about all treatment cycles. This includes Donor Insemination, IVF, ICSI, Natural IVF (no stimulation), Preimplantation Genetic Diagnosis and Preimplantation Genetic Screening.
The history of IVF success rates
The HFEA’s statutory duty is to both publish and publicise the data that they collect. Their first ‘Guide to Fertility Treatments’ was released in October 1995. It presented the live birth rates for all IVF treatments, for all patients (irrespective of their age), and ranged from 0 (yes, zero!) to 19% overall. The average UK success rate at this time was 14%. Thankfully, improvements in IVF techniques meant that the success rates began to rise. However, this quickly resulted in a league table of clinics, leading to speculation that some were ‘cherry picking’ the easiest and youngest patients to boost their percentages.
To counteract this, the HFEA expanded the patient guide. Consequently, in 1998 the patient age groups were broken down to <38 years and all ages. Later, in 2002, they were broken down further into four distinct bands: <35, 35-37, 38-39 and 40-42. Finally, in 2005, they became the seven age groups that we have now.
Until 2008, the statistics had always been presented by reference to treatment cycle started. However, some IVF clinicians were beginning to request that the success rates be shown as ‘live birth Per Embryo Transferred (PET)’. This is because it removes the poor prognosis patients from the overall statistics.
How patients want success rates to be published
A consultation was launched by the HFEA, following which it produced two reports. The first in March 2009 concluded that patients preferred ‘Live birth per cycle started’. This was because they wanted to easily see the outcome for their individual age, diagnosis and treatment. The second in May 2009 noted increasing conflict between what the majority of patients and some clinicians wanted.
In 2014, as part of their ‘Information for Quality programme’, the HFEA embarked upon another consultation process, very similar to the Strategy Consultation that they are promoting from 20 May to 8 August 2019. This public consultation asked the question ‘Should we use births per embryo transferred as the headline figure for the clinic success rate?’. Of the respondents, 82% of patients, donors, donor conceived people or their parents rejected this, as they found it too confusing. They wanted to know what their individual chance of success was from starting a treatment cycle at each clinic.
How are IVF success rates presented now?
The HFEA presents its headline success rates for each individual clinic as ‘Births Per Embryo Transferred (PET). This shows the number of births (counted as a single birth event, i.e. twin births count as one live birth event) divided by the total number of embryos transferred. They present the data in this way as it ‘promotes embryo transfer practices that minimise the chance of twins, which may carry health risks for the babies and the mother’ (hfea.gov.uk).
‘However, this measure does not incorporate the outcomes of patients who do not reach embryo transfer stage‘ (hfea.gov.uk).
What is wrong with IVF success rates as ‘Live Birth Per Embryo Transferred’?
There are a number of high-profile fertility specialists who have been vocal about this way of presenting results. Robert Winston, Mohamed Taranissi and Norbert Gleicher all state that success rates ‘Per Embryo Transferred’ are misleading to patients. This is because;
1. It can falsely improve poor results
The HFEA website breaks the IVF process down into six stages, each of which have a bearing on the overall success or failure of the cycle. One of the first stages is stimulation of the ovaries. In this case, using the correct medication dosage and subsequent monitoring of response , with ultrasound and blood tests, are essential to produce the best quality and quantity of eggs. Over- or under-stimulation will produce too many or too few eggs, potentially resulting in a ‘freeze all’ due to Ovarian Hyperstimulation Syndrome (OHSS) or a cancelled cycle. However, both of these scenarios, which are bad for the patients, improve the success rates of the clinic by excluding those patients from their statistics.
In addition, effective egg collection and good laboratory techniques are important. Maximum fertilisation and successful embryo culture should be standard procedures. However, in reality, there is a wide variation between the performance of various clinics. Reduced egg numbers from an unsuccessful egg collection or poor-quality embryos will increase the number of cancelled cycles. Thereby removing these patients from the clinic statistics and falsely improving the overall success rates.
2. It penalises clinics that treat poorer prognosis patients
In recent years the focus has been to encourage patients to replace one embryo only. However, the number of embryos replaced in an IVF cycle is determined by professional body guidelines e.g. NICE and ASRM. They state that two, and sometimes three, embryos should be replaced in patients over a certain age, or in younger patients whose embryos are not of good quality or who have had previous failed treatments.
This means that a fertility clinic which has poorer prognosis patients i.e. a greater proportion of older patients, or those that have had multiple failed cycles, should routinely replace more than one embryo according to professional guidelines. This will increase the chance of the patient becoming pregnant. However, it will also significantly reduce (by up to 50%, or more sometimes) the actual success rate of those clinics.
How to interpret IVF success rates
Producing statistics in multiple ways is confusing. For example, if 100 patients start an IVF cycle, 80 have an embryo transfer, 60 became pregnant and of those 45 have a baby, the live birth rate would always be 45% per cycle started, whereas it would be anything from 56.2% to 28.1% per embryo transferred (for the same patients) depending on the number of embryos transferred, see below (Figure 2).
This is because whilst 80 patients had an embryo transfer, if all of them had to have 2 embryos transferred as per the professional guidelines, therefore, to calculate the live birth
- per cycle started would be: 45 live birth events ÷ 100 (cycles started) x 100% = 45%
- per embryo transferred would be: 45 live birth events ÷ 160 embryos transferred (80 x 2 embryos) x 100% = 28.1 %. However, if all patients have a single embryo transferred, the live birth per embryo transferred would then be: 45 live birth events ÷ 80 embryos transferred (80 x 1 embryo) x 100% = 56.2%
- Therefore, clinics accepting poorer prognosis patients, i.e. the ones that must have more than one embryo transferred, will appear to have a misleadingly lower success rate, when expressed as per embryo transferred.
Conclusion
IVF live birth rates fluctuate greatly. In the UK, for the under 35 age group (per cycle started), they vary from 14% to 63%. Both the HFEA and CDC publish success rates on their websites, which allow patients to compare clinics. In the UK, they have been published since 1995 as per treatment cycle started. However, from 2008 onward, there has been a gradual shift by the HFEA towards presenting them as per embryo transferred, which the majority of patients have opposed, and leading fertility experts have labelled confusing and misleading.
The HFEA currently presents the UK’s top headline IVF success rates as per embryo transferred. This is despite the fact that a recent public consultation (in December 2014) have shown that 82% of patients and their families requested that the live birth statistics be presented as per treatment cycle started because this allows them to understand what their actual chances of success are throughout the entirety of the IVF cycle.
There is sparse provision of funded fertility treatment worldwide, which forces patients to self-fund and creates intense competition between clinics. One of the main ways of attracting patients is by publicising IVF success rates. This applies to both private and NHS centres (as almost all of them provide self paying options). Presenting statistics as per embryo transferred falsely reduces the overall success rates of good clinics which are treating poor prognosis patients. This is because professional guidelines recommend transferring multiple embryos to those patients, which will significantly lower the actual success rates of those clinics.
In recent months, the HFEA have berated fertility treatment, with media interviews, asking clinics to be more transparent in their reporting. However, by primarily showing success rates as ‘Per Embryo Transferred’ (PET), they themselves are not helping prospective patients assess the realistic chance of having a baby from each IVF cycle they start.