A new study has found a link between hormonal intrauterine devices (IUDs) and breast cancer.
The research is important, but media reports of a large increase in risk may be causing unnecessary worry.
Let’s put the findings in perspective for people who use IUDs.
What are IUDs?
IUDs are commonly used contraceptive devices. They sit inside the uterus (womb) to prevent pregnancy.
Older versions contain copper as their active ingredient. Newer “hormonal” IUDs slowly release a synthetic progesterone called levonorgestrel. This mimics the body’s natural progesterone hormone.
Both the copper and hormonal types of IUD are highly effective at preventing pregnancy over many years. Fertility is readily restored when they’re removed.
But the hormonal IUDs have the extra advantage of making periods lighter and less painful. Some people have one inserted for these reasons, even if they don’t need contraception.
Many women experience pain on insertion or spotting in the first few months of use. But compared to other contraceptives, women generally find IUDs very acceptable and continue to use them.
What did the new study find?
The new study, by researchers from Denmark, used data from national health registries to look for links between hormonal IUD use and breast cancer.
They tracked nearly 80,000 people who started hormonal IUDs across two decades. They compared these people to an equal number of people born at the same time who did not use hormonal IUDs.
On their raw numbers, you might think hormonal IUDs prevented breast cancer, because there were 720 cases of breast cancer in the hormonal IUD group, and nearly 900 in the other group. But that’s not the full story.
Ideally, when researchers study the effects of medicines, they do a “randomised controlled trial”, where researchers use chance to decide whether people get one treatment or another. This ensures the two groups are very similar apart from the treatment being studied. That’s not what happened here.
Read more:
Randomised control trials: what makes them the gold standard in medical research?
Instead, they simply studied people who had decided to have a hormonal IUD, and compared them to people who didn’t. This means the groups were different in many other ways.
So, the hormonal IUD group and the other group might appear to have a different risk of breast cancer – not because of the IUDs, but because of their other differences. For example, more highly educated women might be more likely to choose IUDs, and also more likely to attend breast cancer screening, where their breast cancer would be discovered.
The researchers “adjusted” their results to account for many differences between the two groups (including education, age, number of children, and some other medicines and medical conditions). After this “adjustment”, the numbers pointed in a different direction: towards a higher risk of breast cancer among people who used a hormonal IUD.
However, there are many other important risk factors for breast cancer the authors seem not to have adjusted for, such as body weight, alcohol use, smoking and physical activity. If there were differences between the two groups in these things, then the study’s results may still be biased. This makes me quite uncertain about the results.
Ultimately, we can’t say the IUDs caused the breast cancer – just that there’s an “association” or “link”.
Read more:
Clearing up confusion between correlation and causation
How big are the risks?
There are two different ways researchers express risk: “relative” and “absolute” risks. Here, the “relative” risk increase was about 30% for women using the IUDs for up to five years, 40% after 5–10 years, and 80% after 10–15 years of use.
These sound like massive risks. But though these statistics compare the risk of breast cancer in IUD users to the risk in non-users, they do not tell us the proportion of women who will get breast cancer. For that, we need to look at “absolute” risk increases.
These are much smaller. For every 10,000 women, this study suggests we might see an extra 14 cases of breast cancer after up to five years of use, 29 cases after 5–10 years use, and 71 cases after 10–15 years use. In “absolute” terms – as a proportion of all the IUD users – all of these risk increases are comfortably under 1%.
Frame Stock Footage/Shtterstock
Reporting the dramatic relative risks, and not the much smaller absolute risks, is a common flaw in stories about health risk, and goes against science reporting recommendations.
What does other research say?
There are other studies on this topic, including a much larger recent study from Sweden based on data from more than half a million users of hormonal IUDs.
This suggested only a 13% relative risk increase in breast cancer – much smaller than the risk increases in the Danish study. This would mean an additional 1.46 cases of breast cancer for every 10,000 women per year.
This is in keeping with a recent large review of studies on this topic, which also found a much smaller risk than the new Danish paper.
The Swedish study also looked at other cancers. The results suggested a decreased risk of cancers of the cervix, ovaries and endometrium (womb lining). This mixed picture of some cancer risk and some cancer protection is also seen for traditional contraceptive pills.
And of course, all contraception protects women from the risks of pregnancy.
What does it mean for me?
The link between hormonal IUDs and breast cancer is probably very small, and might be a statistical illusion rather than a real thing.
Even if it’s a real risk, it may be offset by protection against other cancers.
And it may be dwarfed by other risks for breast cancer, such as high body weight, physical inactivity, alcohol use, and smoking. Online resources can help you visualise these risks.
Hormonal IUDs aren’t the right contraceptive choice for every woman. However, they deserve to stay high up on the menu of options.
