High IVF success rates rely on young eggs
A summary of my first ever published paper!
Today, my first ever peer-reviewed paper is coming out at Population Studies! I’m really excited about this milestone in my academic career and wanted to share a quick summary here. The full paper is open access and you can find it here. Huge thanks go to my wonderful co-author, Ester Lazzari! I learned so much from her and it was really fun working together.
As more people delay having children, there is a widespread belief that in-vitro fertilisation (IVF) can serve as a reliable backup plan that easily allows you to compensate for age-related infertility. Unfortunately, this is a misconception: IVF success rates drop sharply with age, and especially for women over 43, IVF success rates are very low. As a result, very few IVF patients manage to have a baby while using their own eggs at these advanced ages.
Our paper highlights this fact by looking at the contribution of IVF1 to fertility rates in the UK. In particular, we break down the contribution to fertility by whether patients are using their own eggs (autologous eggs) or donor eggs. This distinction makes it clear that the contribution to fertility rates in the oldest age groups comes predominantly from donor eggs and that treatments with autologous eggs are usually unsuccessful in these age groups.
In this post, I will briefly present our main findings which are based on data from the UK but are relevant to other countries as well:
Treatment rates are going up in all age groups
Success rates have almost doubled between 1991 and 2018
When using a woman’s own eggs, success rates fall sharply with age and the oldest age groups have seen almost no improvement in success rates over time
When using donor eggs, success rates are similar across age groups
This means that IVF births in older women are mostly due to donor eggs: More than half of IVF births in the age group 43-44 are the result of donor eggs and more than 90% of IVF births in women aged 45-50 are due to donor eggs
The main takeaway from our paper is that IVF using a woman’s own eggs is very unlikely to be able to support fertility at older ages. Egg donation and egg freezing are more promising but also suffer from important limitations. The public should understand better what the consequences of fertility postponement are and that they shouldn’t just rely on IVF.
We use data2 from the Human Fertilisation and Embryology Authority (HFEA) in the UK which has all IVF treatments conducted in the UK between 1991 and 2018. The data covers more than 1.2 million treatment cycles and more than half a million unique patients.
Treatment rates are rising
First of all, we look at how treatment rates have evolved over time. Our data begins in August 1991 so 1992 is the first full year of data that we have. In that year, 11,000 patients began their first IVF cycle, while in 2018, at the end of our data set, almost 25,000 patients started IVF. Using female population data from the Human Fertility Database (HFD), we can look at treatment rates per 1,000 women in the different age groups:
Source: Figure 1, Bruckamp and Lazzari (2025)
Note that the y-axes are different because otherwise it would be hard to see what’s going on with the donor egg treatments! We can see that treatment rates are going up in all age groups during our study period. Women between 35 and 39 have the highest treatment rates. Based on these treatment rates, women under 40 use donor eggs in around 2% of treatments. This rises to 5.1% in the 40-42 age group, 15.6% in the 43-44 age group and 53.0% in the 45-50 age group.
Success rates are rising and depend almost exclusively on the age of the eggs
Many people don’t realise how IVF success rates depend almost exclusively on the age of the egg used. We break down success rates for first treatment cycles3 over time and distinguish between women using their own eggs and women using donor eggs:
Source: Figure 3, Bruckamp and Lazzari (2025)
There are several notable things in this figure. First of all, we can see that IVF has become much more effective over time! If we look at general success rates for all patients, then 14.7% of first treatments were successful in 1991 and 28.3% were successful in 2018, meaning that success rates have almost doubled. However, if we look at the oldest age groups in pink and red at the bottom of the left panel, there has barely been any improvement in success rates for these groups. Success rates remain under 5% for the first cycle for women aged 43 and above.
The most important thing in this graph is how different the left and the right panels look: When using women’s own eggs, there are massive differences in success rates between the different age groups and success rates fall very rapidly with age. When using donor eggs however, success rates are basically indistinguishable between age groups! This really underscores the fact that the most important determinant of success is the age of the egg: Older women can have high success rates if they use donor eggs collected from younger women.
The contribution of IVF to fertility rates is rising and is driven by donor eggs at older ages
Taken together, rising treatment rates and rising success rates have led to IVF contributing increasingly to fertility rates in the UK. In 1992, IVF births were responsible for 0.3% of the total fertility rate (TFR). By the end of our study period, in 2018, IVF births were responsible for 3.0% of the TFR.
We decompose this by looking at the contribution of IVF to age-specific fertility rates (ASFR) in the different age groups and further distinguishing between autologous and donor eggs:
Source: Figure 5, Bruckamp and Lazzari (2025)
As expected, the contribution is much higher in the older age groups. I was a bit surprised when I first made this graph that even in the oldest age group more than 80% of births are non-IVF births throughout the whole time period! At the end of our study period, 7.8% of births in the 40-42 age group were due to IVF, 8.4% in the 43-44 age group, and 14.9% in the 45-50 age group.
The darker shaded areas of the bars show the contribution that comes from donor eggs. In the 43-44 age group, 53.7% of IVF births in 2018 were due to donor eggs and in the 45-50 age group 92.3% were due to donor eggs. At older ages, IVF births thus come predominantly from donor egg treatments and especially in the oldest age group, there would be almost no contribution to fertility rates without donor eggs.
Conclusion
As I said above, the main takeaway from our paper is really that IVF doesn’t work very well at older ages when using a woman’s own eggs. I think a lot of people aren’t sufficiently aware of this fact and might make decisions about postponing childbearing thinking that IVF is a good backup option if they end up wanting to have kids later in life and can’t conceive naturally.
The technology of assisted reproduction has improved over time but it still very much relies on young eggs for high success rates. This means that egg donation but also egg freezing are much more promising when thinking about supporting fertility at older ages: While donor eggs come from young women, egg freezing stops the ageing of the eggs and therefore leads to higher success rates when using those eggs later in life. Doing this research contributed to me deciding to freeze my eggs and I have previously written about egg freezing in much more detail here.
However, egg donation and egg freezing aren’t magical solutions either. Most people have a preference for having biological children, as evidenced by the fact that in most age groups, the majority of patients try with their own eggs. Even in the 45-50 age group, where success rates using autologous eggs are so low, 47% of treatments are done with a patient’s own eggs. The supply of donor eggs might also not be sufficient for the rising demand and some people have ethical objections to the use of donor eggs.
Egg freezing doesn’t come with these complications but it is still a very expensive and slightly invasive procedure. When freezing enough eggs at a young age, it gives a very good chance of a live birth later in life. However, you need to be prepared to go through several rounds of egg freezing to collect enough eggs and to go through several embryo transfers later to achieve those births. Given that egg freezing works better when the woman is younger, it also usually means going through the procedure without knowing whether those eggs will actually be needed in the future.
When making decisions about when to try to have children, people should have a clear sense of how fecundity declines with age and that IVF itself doesn’t offer a safe backup option.
In our paper we talk about assisted reproductive technologies (ART) which is technically the correct umbrella term for different types of treatments like IVF or IVF with intracytoplasmatic sperm injection (ICSI). In the blog post, I will just stick with using IVF since that is the term that people colloquially use to refer to these different types of treatments that are covered by our data.
There is a replication package here with detailed instructions for how to access the data and the code with our analyses.
The reason we only look at first cycles is that this deals with selection bias. Patients going for further treatment might be negatively selected since those who are most fecund are more likely to already have been successful with the first treatment. There might also be positive selection since those who are more likely to be successful (for example because they responded to the stimulation well or because their test results indicate a good ovarian reserve) might be more willing to undergo further treatment. We explain this in more detail in the paper and also show that our results hold when including all treatment cycles.