“Women using a vaginal ring or skin patch for contraception are at around double the risk of a blood clot compared to those taking the Pill,” the Daily Mail has reported.
The news is based on a large Danish study that looked at contraceptive use in more than 1.5 million women. The study looked at how different hormone-based methods such as implants, the patch and the pill related to the risk of blood clots. Between 2001 and 2010 researchers recorded a total of 3,434 blood clots, also known as venous thromboembolisms or VTE. The background rate of VTE among women not using hormonal contraception was 2.1 per 10,000 woman-years (for example, 2.1 would occur if 1,000 women were followed for 10 years). The highest rate of VTE was among women who used the contraceptive patch, with 9.7 per 10,000 woman-years. Women using a common oral contraceptive pill experienced a rate of 6.2 per 10,000 woman-years.
Despite what some news coverage might suggest, hormonal contraceptives containing oestrogen ( the combined oral contraceptive pill, transdermal patch and the vaginal ring) are already recognised as increasing the risk of VTE, although the risk is very low. Instead of discovering a new danger from using hormone-based contraceptives, the research simply refines estimates of the clot risk associated with different methods.
Women should be fully informed of the potential risks and benefits of any contraceptive option that they choose. They can talk to their GP or nurse about these. Despite the small increase in risk associated with the patch or vaginal ring compared with the combined oral contraceptive pill, there may be women for whom this is still an appropriate choice.
Where did the story come from?
The study was carried out by researchers from the University of Copenhagen and did not receive external funding. It was published in the peer-reviewed British Medical Journal.
News coverage generally failed to reflect the true context of this research. It is already known that there is a clot risk associated with use of oestrogen-containing contraceptives, and this research has helped to analyse some of the finer points around the issue rather than revealing any previously unknown risk. This research provides valuable quantification of the possible risk among users of hormonal contraception but the findings are not as unexpected as the media implies.
In particular, the Daily Mail’s headline is misleading and may scare women: ‘Women using alternative contraception to the Pill are at double the risk of blood clot’. This might suggest to readers that any alternative option to the combined oral contraceptive pill doubles the risk. This is not true. The oestrogen-containing patch or vaginal ring increase risk slightly more than the oestrogen-containing pill, but the pill itself actually significantly increases risk of VTE compared with non-use, or use of progestogen-only contraceptives or barrier methods.
What kind of research was this?
This was a large, national cohort study that compared contraceptive use and VTE risk among more than 1 million Danish women. It used four national registries in Denmark to look at all non-pregnant women aged 15-49 (who were free of cancer or thrombotic disease) and collected data on their contraceptive use over the period 2001 to 2010. From these data, researchers were able to see how the rate of VTE among users of non-oral hormonal contraceptives compared with the rate in users of the oral contraceptive pill, as well as in women who didn’t use hormonal contraception.
A cohort study is a good way of evaluating whether a certain exposure increases risk of a certain outcome. The researchers of this cohort study when conducting their analyses have attempted to adjust for some of the possible confounding factors that could be affecting the results.
What did the research involve?
Data available in Danish registries allowed for 1,626,158 non-pregnant women to be followed between January 2001 and December 2010. The researchers were only interested in first-ever events of VTE, so excluded women who had had any type of thrombotic event in their veins or arteries before the study period (assessed by checking medical registries from 1977 to 2000). They also excluded those with cancer, those who had had a hysterectomy or both their ovaries removed and those who had been sterilised.
Since 1995 the registries consulted by the study have recorded all filled prescriptions, and so the researchers were able to obtain information on all hormonal contraceptives prescribed between 1995 and 2010. They recorded the products according to progestogen type, oestrogen dose, method of administration and duration of use. The registry also records all hospital admissions.
Any hospital admission for suspected VTE (a clot in a vein or blood vessel) or pulmonary embolus (a clot in the blood supply to the lungs) was confirmed by examining prescribed anticoagulation therapy recorded in the national registry of medicinal products for at least four weeks after the diagnosis. Fatal VTEs were captured by the national causes of death registry.
The researchers also obtained information on some possible confounders that could influence VTE risk, such as educational status, age and calendar year (contraceptives prescribed or healthcare in general may have changed subtly over the nine-year study period). However, they didn’t have information on other relevant confounders such as smoking.
What were the basic results?
The researchers had 9,429,128 woman-years of follow-up data (for example, 90 woman-years of follow-up could be 90 women followed for one year, or nine women followed for 10 years). Over this period there were 3,434 confirmed first-event VTEs.
The researchers then calculated the VTE rate according to use of different contraceptive types:
- not using hormone-based contraception: women not using any hormone-based contraception experienced a background rate of 2.1 events per 10,000 woman years (for example 2.1 would occur if 1,000 women were followed for 10 years)
- contraceptive patch: a rate of 9.7 per 10,000 woman years
- vaginal ring: a rate of 7.8 per 10,000 woman years
- combined oral contraceptive pill (30-40 micrograms of oestrogen in combination with levonorgestrel): a rate of 6.2 per 10,000 woman years
- combined oral contraceptive pill (30-40 micrograms of oestrogen in combination with norgestimate): a rate of 4.5 per 10,000 woman years
- progestogen implant: a rate of 1.7 per 10,000 woman years
- progestogen-releasing intrauterine system: a rate of 1.4 per 10,000 woman years
The researchers calculated that, after adjustment for confounders, the risk of confirmed VTE among users of the contraceptive patches was 7.9 times that of women not using hormonal contraception (95% confidence interval 3.54 to 17.65), and 2.3 times that of users of the combined oral contraceptive pill (95% CI 1.02 to 5.23).
The risk of confirmed VTE among users of the vaginal ring was 6.5 times that of non-users, and 1.9 times that of users of the combined oral contraceptive pill. Compared with women who did not use hormonal contraception, women who used the combined oral contraceptive pill had around a trebled risk of VTE.
Compared with women who did not use hormonal contraception, users of the progestogen implant or progestogen-releasing intrauterine system had no increased risk of VTE.
How did the researchers interpret the results?
The researchers conclude that “women who use transdermal patches or vaginal rings for contraception have [respectively] a 7.9 and 6.5 times increased risk of confirmed venous thrombosis compared with non-users of hormonal contraception of the same age”. Respectively, this equates to 9.7 and 7.8 events per 10,000 woman-years (for example, for the transdermal patch a rate of 9.7 events among 1,000 women followed for 10 years).
This large study provides valuable information on the rate of VTE that may be experienced among users of hormonal contraception.
However, the findings are not completely surprising. Oestrogen-containing hormonal contraceptives are already known to increase the risk of VTE, and medical professionals already consider this potential side effect when prescribing contraception and monitoring patients. Instead of revealing some new or major danger, this study provides a good indication of how the risks compare for a variety of different contraceptive methods.
The oestrogen-containing contraceptives currently available are the combined oral contraceptive pill, the transdermal patch (of which there is one licensed product – brand name Evra) and the vaginal ring (of which there is one licensed product – brand name NuvaRing). There are many different preparations of combined oral contraceptive pill that contain different strengths and forms of oestrogen and progestogen. Different progestogens contained in combined oral contraceptive pills are considered to have a differing effect on risk of venous thromboembolism. This study chose to look separately at VTE rate among users of combined oral contraceptive pills containing levonorgestrel or norgestimate, but there are various other types of progestogen contained in other combined pills, and this study has not examined those.
Progestogen-only contraceptives are not known to increase risk of VTE, and this study supports this. Users of implants and the progestogen-releasing intrauterine system had no higher risk than non-users of hormonal contraception. Information was not available for progestogen-only pills or injections.
There are some further points to note about the study:
- This was a cohort study looking at associations within a large population using contraception in an everyday setting rather than in the artificially controlled setting of a clinical trial. As such, the method of contraception used will have been down to the woman’s personal choice in consultation with her doctor, and there may be health and lifestyle factors that have influenced the choice of contraceptive and that could also influence risk of VTE. The researchers adjusted their results for possible confounders of age, education and calendar year, and also excluded women who may be at particularly increased risk of VTE. However, information on other relevant confounders such as smoking or body mass index was not available.
- Use of contraception was determined by looking at filled prescriptions. Although the women are likely to have used the method prescribed for them, and for the prescribed time period, this may not always have been the case.
- There were far fewer women in the study using the patch (6,178 woman years) or vaginal ring (50,334 woman years) compared with the combined oral contraceptive pill (530,241 woman years). The event rate of VTE among users of the patch or vaginal ring was correspondingly low (six events among users of the patch; 39 with the ring). Therefore, although the ring and patch were calculated to give double the risk of the combined oral contraceptive pill, the low event rates mean that the risk figures are only estimates, and may not be completely accurate. This is reflected by the wide confidence intervals. In other words, even a small spike in cases could inflate the rate seen.
Overall, the study highlights the importance of women being fully informed of the potential risks and benefits of any contraceptive option that they choose. Despite the small increase in risk associated with the patch or vaginal ring compared with the combined oral contraceptive pill, there may be women for whom this is still an appropriate choice and for whom the benefits, such as not having to take a daily pill, outweigh the small extra risk.