Newer types of hormonal contraceptive may still carry small breast cancer risk

Thursday December 7 2017

"All forms of hormonal contraception carry breast cancer risk, study finds," The Guardian reports.

The link between older versions of the oral contraceptive ("the pill") and breast cancer has long been recognised, as this type of oral contraceptive relies on the hormone oestrogen, and prolonged exposure to oestrogen is a known risk factor for breast cancer.

Researchers wanted to see if newer forms of contraception, which tend to make use of an alternative hormone called progestogen, carried a similar risk. Examples of these newer contraceptives are the IUD and the contraceptive injection.

The researchers looked at data on 1.8 million Danish women aged 15 to 49 years to investigate the link between hormonal contraceptive use and breast cancer. This study found that women who either currently or recently used hormonal contraceptives were 20% more likely to develop breast cancer compared to those who did not use hormonal contraception. This risk decreased gradually over a few years once women stopped using contraception.

While this may sound alarming, the number of cases of breast cancer was small, occurring in less than 1% of the women. This means that if 7,690 women took the pill for a year, this could lead to one extra case of breast cancer.

Because of the millions of women who use hormonal contraception, the implications of this study will need to be considered by policymakers and in future breast cancer screening guidelines.

But it remains the case that the actual risk of breast cancer to an individual woman is tiny.

Where did the story come from?

The study was carried out by researchers from the University of Copenhagen and Aberdeen University. It was funded by the Novo Nordisk Foundation, which is a large Danish Multinational pharmaceutical company. The researchers state Novo Nordisk had no input in the study analysis or interpretation of results. The study was published in the peer-reviewed New England Journal of Medicine.

The Guardian generally reported the results of the study accurately, explaining the importance of further research to investigate contraceptive options that do not increase the risk of breast cancer. It also considered the benefits of hormonal contraception for preventing unwanted pregnancy and reducing the risk of ovarian and womb cancers.

The Guardian did miss some details about the limitations of the study. It reported "women who were older were more at risk than younger women," but did not mention that researchers had less information about confounding factors for women who were older, which could have also biased the results.

What kind of research was this?

The researchers used data from a large, ongoing nationwide cohort study, which attempts to include all women in Denmark between the ages of 15 and 49. The study collects individually updated information about the use of hormonal contraception, breast cancer diagnoses, and potential confounders.

The study aimed to see if women of reproductive age, who were using currently available hormonal contraception, were more at risk of breast cancer compared to women not taking hormonal contraception.

A cohort study following women over the reproductive age is a good starting point to test the link between contraceptives and breast cancer, but it’s important to remember this kind of study cannot prove cause and effect.

What did the research involve?

All women living in Denmark aged between 15 and 49 on January 1 1995, and those who were 15 years of age before December 31 2012, were eligible for this study. Women with cancer, deep vein thrombosis, and those who had received infertility treatment, were excluded. A total of 1,797,932 women were included in the study.

Women were followed until:

  • the first diagnosis of breast cancer
  • death
  • registry recorded emigration
  • reaching age 50
  • the end of the follow-up period, which was on average around 11 years

Information on the women's use of hormonal contraception was taken from the National Register of Medical Product Statistics, and was regularly updated throughout the follow-up period. The Danish Cancer Registry was used to identify invasive breast cancers. The use of hormonal contraceptives was categorised as:

  • current or recent use (stopped within the last six months)
  • previous use (stopped at least six months ago)

Start of use was the date the prescription was purchased. If women had the contraceptive coil fitted, it was assumed this was used for four years, unless the woman became pregnant or another contraceptive was prescribed before the end of the four-year period.

The following confounders were considered in the analysis:

  • education
  • number of previous pregnancies
  • diagnosis of polycystic ovary syndrome
  • endometriosis (a condition where tissue that acts like the lining of the womb is found in other parts of the body)
  • family history of breast and ovarian cancer
  • body mass index (BMI)
  • smoking status
  • age

For the analysis, the researchers calculated risks for developing breast cancer for women using hormonal contraceptives, compared to those who did not. The researchers also noted the number of new cases in the population at risk over the study period.

What were the basic results?

Among the 1.8 million women who were followed for around 11 years, 11,517 cases of breast cancer occurred, which was less than 1% of the study population.

The main results were as follows:

  • Compared to women who had never used hormonal contraception, women who did had a 20% increased risk of breast cancer (relative risk (RR) 1.20, 95% confidence interval (CI) 1.14 to 1.26).
  • Using hormonal contraception for less than one year reduced this increased risk of breast cancer to 10% (RR 1.09, 95% CI 0.96 to 1.23).
  • Using hormonal contraception for more than 10 years had around a 40% increased risk (RR 1.38, 95% CI 1.26 to 1.51).
  • Women using progestogen-only intrauterine system (a coil with progesterone) had a 21% higher risk of cancer than women who had never used hormonal contraceptives (RR 1.21, 95% CI 1.11 to 1.33).
  • The overall absolute increased risk of a breast cancer diagnosis among current and recent users of any hormonal contraceptive was 13 cases (95% CI, 10 to 16) per 100,000 person years.
  • They estimated this means one extra breast cancer case for every 7,690 women using hormonal contraception for one year.
  • The risks among women who used hormonal contraception for less than five years rapidly decreased after discontinuation of hormonal contraception.

How did the researchers interpret the results?

The researchers state: "the risk of breast cancer was higher among women who currently or recently used contemporary hormonal contraceptives than among women who had never used hormonal contraceptives." They also said "this risk increased with longer durations of use; however absolute increases in risk were small."


This study has some definite strengths, as it uses a large, nationwide cohort of women in Denmark, with access to accurate exposure and outcome data for contraceptive use and cancer diagnosis using two registries (The National Register of Medicinal Product Statistics, and the Danish Cancer Registry). The study is able to produce valuable information on the small increased risk of breast cancer for women who use hormonal contraceptives.

There are, however, limitations that may have introduced bias into the results:

  • The study was not able to adjust for some confounders also known to be associated with developing breast cancer, such as the start date of each women's period, whether they breast fed, how much alcohol they consumed, and physical activity.
  • Information on women's BMI was only available for 538,979 women in the study (around 30%).
  • Information on women who had polycystic ovary disease was only available for those who had been hospitalised, so the rate was likely to be higher.
  • Adjusting for a family history of breast cancer may have introduced an underestimation of the results, as these women are perhaps less likely to use hormonal contraceptives.
  • Using prescription data does not prove the women are taking the contraception. They may have a prescription and stop taking it before the end of the prescription, or not be taking the contraceptive correctly, which is fairly common.
  • This study was conducted in Denmark, which has a publically funded healthcare system, so while these results may be applicable to the UK, global generalisability would require data from different populations.
  • This is a cohort study, therefore the results are only able to show an association, not cause and effect.

The results of this study are significant at population level as millions of women across the globe use hormonal contraception. But the risk at an individual level remains tiny.

If you are concerned you can discuss alternative methods of contraception with your GP.

Analysis by Bazian
Edited by NHS Website