Birth Control & Blood Clots – Visualizing The Risks

If you’ve been using the Patch, Yaz or Yasmin, you’ve got to be wondering in the wake of this week’s news whether or not you should reconsider your choice of contraceptive. After all, the FDA has pretty much confirmed that these methods have a higher risk of blood clots than older birth control pills. By now, your mom has probably called you and told you to get off that nasty pill, your roommate may already have hidden next week’s patch from you, and you’re holding your breath waiting for a call back from your doctor, who’s probably fielding about a hundred calls today alone from worried pill and patch users

All you really want to know, though, is this –

What’s My Risk? 

It’s an important question that deserves an answer and a conversation with  your doctor. Let’s see if I can help inform that discussion. But first, you need to know a a bit about the subject at hand – blood clots.

What are Blood Clots?

Blood clots are blockages in the vein or artery that occurs when the blood coagulates in the blood vessels.

  • A DVT (deep veinous thrombosis) occurs when blood coagulates in the veins of the leg or arm, where it causes pain, swelling and inflammation.
  • A PE (pulmonary embolus) occurs when a clot forms in the lungs or breaks off from a DVT in the leg and gets lodged in the lung. PE’s causes shortness of breath and chest pain, and if large enough, lack of oxygen and even death.

DVTs and PEs are treated with blood thinners, which almost always work to dissolve the clot, although long term sequelae of the clot can occur.

  • Clots to the heart and brain are exceedingly rare in young women. In birth control pill users, that risk is confined to women over 35 who smoke (and should not use estrogen containing hormonal birth control) and women who suffer from ischemic migraine with aura or underlying medical problems such as heart arrythmias that predispose them to stoke. (Women over 35 with regular migraine may also be at increased risk).

For the otherwise normal, healthy young woman on pills or the Patch, the risk for heart attack and stroke is just too rare to even begin to compare between users and non-users of these methods. And in fact, these risks have not been reliably shown to differ between currently marketed methods. So in this discussion we’re going to confine ourselves to talking about the risk from DVT, where the data appear to be a bit clearer.

How does Birth Control cause blood clots?  

Birth control methods that contain estrogen increase the risks of blood clots by altering the delicate balance of clotting factors produced in the liver. In spite of this, however, most women taking estrogen don’t get clots. So, other factors must also be at play for a clot to form.  Here are those other factors that we know and understand –

  • Long plane flights and car rides. Prolonged immobilization can cause the blood to pool in the legs and clot. In my two decades of practice, almost every clot I’ve seen in my patients using estrogen containing birth control occurred after a long plane flight or car ride. That’s why it’s so important to get out of your seat and walk around on a long plane flight and make frequent stops on long car rides.  And, unless they’re rising in business or first class with a seat that allows them to elevate their legs, I also advise all my patients taking estrogen to wear knee high travel compression socks, since studies have shown that these can lower clot risks due to plane travel.
  • Genetic mutations / Family History. Some people are predisposed to clotting because they carry a mutation in their clotting factors, the most common of which is Factor V Leyden mutation, found in about 5% of the population. Women with such mutations should avoid estrogen. Some day we’ll have an inexpensive blood test to identify these women, but right now the best clue to a genetic mutation is a family history of blood clots. In families with such histories, genetic screening can be done in the affected individuals to identify the mutation and then screening of other family members to find out who is at risk.
  • Obesity and smoking also increase the risks of clotting, in the arteries as well as the veins. Women over 35 who smoke should not use estrogen containing birth control.
  • Surgery and hospitalization. Prolonged immobilization is the reason. Hospitals use compression stocking, pneumatic air pumping leg wraps and even low doses of blood thinners to prevent clots due to hospitalization. (God forbid they get the patients up and walking, but that takes nursing staff, and we’d rather spend the dollars on devices and drugs, don’t get me started…)
  • Varicose veins. Birth control pills don’t cause varicose veins, but women who have varicosities (not superficial tiny spider veins, but large deep veins in the leg) have a higher risk of clotting due to pooling of the blood in the veins of the extremities.
  • Advancing Age. The biggest factor associated with blood clot risks is age, with the elderly being at particularly high risk compared to younger individuals. However, pills are generally still safe to use in healthy women up to the age of menopause.
  • Pregnancy. Perhaps the highest risk women take for blood clots is during pregnany, a time when your risk for clotting increases up to 5 times the rate in non-pregnant women.  Here, the hormone of pregnancy, fluid shifts and edema in the legs are culprits. Researchers often compare the risks of hormonal birth control to the risk of pregnancy, although a better comparison is to compare it to the risk from using another method, combined with the risk of getting pregnant from that method if it is less effective.

The progestin component of pills

Scientists believe one of the factors affecting clot risk in one method vs another may be the progestin component.

  • Northindrone (1st generation progestin)
  • Levonogestrel (2nd generation progestin)
  • Norgestimate, desogestrel, gestodene (3rd generation progestins)
  • Dropeserinone (the progestin in Yaz)

Why? It’s not so clear, and some experts maintain that it makes no biologic sense that, estrogen dose being equal, progestins should impact clot risk at all. Despite this, research is mounting that lower clot risks seem to exist for first and second generation pills. Norgestimate pills may have a similarly low risk, perhaps because they are metabolized in the body to levonorgestrel. The progestin in the Patch is metabolized to norgestimate, but the higher estrgoen dose in the patch probably contributes to that method’s higher clot risk compared to norgestimate pills.

Now that you’re an expert on blood clots, let’s visualize the risks of your birth control

Because the risk of PE is so low, most studies either combined DVT and PE risk, or report on DVT risks alone, which is what I’m going to do.  In general using the patch or  yaz will increase your odds of a DVT by about 50% – that’s called relative risk of 1.5. Translating that to absolute risks is difficult, because the actual numbers of clots occurring in a given study depends on so many things, not the least of which is how they define that a clot has occurred. Each study the FDA examined used a slightly different methodology – some used pharmacy database prescriptions for blood thinners, some used claims-based diagnoses and others added confirmatory chart reviews. Each method has it’s biases, and none is perfect.

Absoute Risk DVT- Background

The background rate of clots among healthy women of reproductive age that I have seen most often quoted in the literature is about 4 per 10,000.  This is your risk if you do not take hormonal birth control.The dots represent 10,000 women over a year’s time, with brown dots unaffected women and red dots those who have a DVT.

DVT Risk – Levonorgestrel and norgestimate pills

The risk in users of second generation pills containing levonorgestrel, or 3rd generation norgestimate pills is about 6 per 10,000. This is the risk that Ortho Evra and Yaz were compared to by the FDA. Here’s what that looks like –

DVT Risk – Ortho Evra Patch

Use of the patch increases that risk to 9 per 10,000.  Here’s what that looks like –

In case you’re having trouble visualizing it, I’ll put it side-by-side.

So you can see, as reported to the FDA,there is an excess of about 3 cases of clots per 10,000 women using the patch compared with those using an older pill. If you’re on the patch, clot risk does not appear to diminish over time. So your excess risk remains about 3 per 10,000 as long as you use the patch.

DVT Risk – Yaz

The risk from using Yaz and Yasmin comes in pretty close to that of the patch, at around 10 per 10,0000. The risk for clots with Yaz decreases the longer you use it. So if you’ve been on it for over a year with no clot, your risk drops significantly for getting a clot in the future, to about 5 per 10,000.

DVT Risk -Pregnancy

Finally, what about the clot risk from pregnancy? Actually, that’s the highest risk of all – about 20 per 10,000 or 2 per 1,000.

What about the Ring? And pills containing desogestrel?

Excellent question. The FDA hasn’t addressed the clot risk with the Ring specifically outside of the FDA approval process. We do know that the ring imparts a significantly lower estrogen exposure than the patch and a 35 ug pill, but it’s unclear if that translates to a lower clot risk. The ring, after all, contains etononorgestrel,  the active metabolite of desogestrel.

Pills containing desogestrel have come in at a risk of about double that of levonorgestrel pills, so I’m going to assume in the absence of data to prove otherwise that the  risk of the ring would be about 8 per 10,000.

Why does it take so long for us to learn about these risks for methods that are already on the market?

Almost all newer contraceptives will have an undefined clot risk, since clinical trials are just too small to detect a statistically significant increase in clots, which as you know now are relatively uncommon events. It’s only when a method makes it out into the general population of millions of women that an increased clot risk becomes evident. (Of course, if the manufacturer does not report all the clots that occur in a trial, that’s a different story, and the subject of recent lawsuits related to Yaz.)

Bottom Line

Your chance of dying from a blood clot related to your contraceptive is about one in a million. The chance you’ll get a blood clot is well below one percent no matter what method of birth control you use.  In that very low risk range, your chance of a blood clot, in order of increasing risk, is –


Risk per 10,000*

Percent risk*

Non-hormonal. Not pregnant.

4 per 10,000

0.04 %

Levonorgestrel pills Noregstimate pills

6 per 10,000

0.06 %

Desogestrel Pills

8 per 10,000



? 8 per 10,000


OrthoEvra Patch

9 per 10,000

0.09 %

Drosperinone pills

10 per 10,000



20 per 10,000


* These numbers are estimates based on my best good faith interpretation of the literature and data presented to the FDA on 12/8-9/11. Better numbers may be forthcoming from the FDA or other sources in the future, but for now I need something for my patients and myself to work with. Margaret Polaneczky, MD

Bottom Line

Even if you take the pill or the patch, the odds are overwhelming that you’ll make it through your reproductive years without ever having a blood clot. 

The risk of actually dying from a blood clot due to your hormonal birth control is about one in a million.

If you want to avoid the clot risk associated with estrogen containing contraceptives, you can use something else for birth control. Depo provera, the progestin only pill, Implanon, the IUD and barriers such as condoms and diaphragm are all reasonable choices.Each of these methods come with their own risks and benefit, and none is perfect.


Additional Reading from TBTAM

24 Responses to Birth Control & Blood Clots – Visualizing The Risks

  1. Thank you so much for circulating this information on birth control and DVT. About a year ago, a good friend of mine died when a blood clot formed in her leg and traveled to her brain. She was young, healthy, and training for a marathon. She contributed the swelling and soreness in her leg to her new exercise regime, and therefore did not seek medical attention. Since her death, Anna’s friends and family have been raising awareness about DVT in hopes that women experiencing these symptoms will seek the medical attention they need. Thank you so much for doing the same.


  2. Thanks for the explanation. I really like the way the dot things help me visualize.

    And thanks for the reminder to get up a bit during plane rides. It was timely!

  3. Here via the Grumpies. I’m not a candidate for any sort of BCPs, but thanks for posting this — very clear and informative.

    In the “full information” vein I thought I’d add that it looks like a woman interested in getting a Factor V Leiden test could order and pay for one out of pocket herself for $300 via the healthcheckUSA (add dot com to get there) website (with which I have no affiliation, though I have had friends who have used it to order tests and have been pleased with the ease of use and with what they characterized as professional delivery of results). That may well be a high enough cost that it is not cost effective at the population level (I have no idea, honestly, how the math works out), and it’s certainly possible it’s not a reliable way to get the test done, but I though it might be useful for people to know what the actual cost is (especially given that “expensive” in medicine can mean anything from a few hundred dollars to tens of thousands) and that’s the best I could do to find an estimate.

    • bogart –
      This article summarizes well the arguments for and against routine screening

      Here’s the relevant section of that article –

      “Selective population screening: before OCP use
      The annual risk of venous thrombosis in an otherwise healthy woman of reproductive age is around 0.01% (table 2​2).). This is increased around 2–5‐fold with OCP use and justifies careful assessment of thrombosis risk before prescription of this form of contraception. However, the OCP is the most acceptable and effective form of contraception for many women and a clear understanding of the absolute risk of VTE is required. The risk of OCP‐related venous thrombosis in factor V Leiden (FVL) heterozygotes is multiplicative, not simply additive, when compared with the risk associated with either of these factors alone. Although case–control studies have estimated this risk at approximately 0.1–0.2% per annum7 for V Leiden heterozygotes without a personal or family history of thrombosis (up to 20 times baseline risk), the risk is substantially greater at around 0.5% per annum if FVL heterozygosity is identified through testing of an individual in whom there is a history of VTE in a first‐degree relative.13 We draw three conclusions from these observations.
      Firstly, most women with factor V Leiden will not have VTE even after using OCP for many years. Indeed, as population screening is not carried out and the family history may be negative for VTE, many factor V Leiden heterozygotes are currently using oral contraception.
      Secondly, 10000 women would have to be screened and the OCP withheld in 500 otherwise healthy FVL heterozygotes to prevent one episode of VTE. A recent review concluded that this approach, when applied to all inherited thrombophilic defects, is not cost effective (£202 402 per thrombosis prevented).31 Crenin32 also evaluated the cost of preventing death due to VTE secondary to OCP use in combination with FVL (case fatality per VTE around 1–2%) and calculated that >92000 carriers would have to be identified and the pill withheld, at a cost of >300 million USD, to prevent one death.
      Thirdly, the value of thrombophilia testing is further undermined by the current recommendation that in women with a history of VTE in a first‐degree relative, use of the OCP is relatively contraindicated.
      Testing for heritable thrombophilia may occasionally be of value in the above‐mentioned situation—that is, before prescription of the OCP in an asymptomatic woman who has a first‐degree relative with a history of VTE. The affected relative must be available and give permission for testing to be carried out. Careful counselling is required. If a heritable thrombophilia can be identified which segregates with thrombosis in the affected kindred, and the person is found to be negative for this, then it would be reasonable to prescribe the OCP on the assumption that the thrombosis risk is close to that of the general population. In fact, Lensen33 found an annual thrombosis risk of 0.56% in FVL‐positive asymptomatic relatives who were identified through family screening of patients with venous thrombosis with FVL, and a rate of 0.15% in relatives who were FVL negative. The rate in FVL‐negative people may be a little higher than background, possibly owing to the coinheritance of as yet unidentified thrombophilic genes in clinically affected families.
      Often we are referred women who are already using the OCP who find that a relative has had VTE. In advising such women it is noteworthy that OCP‐related venous thrombosis is most likely to occur in the first 6–12 months of use.34 Thus, if a woman has been using the OCP for several years without complications, discontinuation on the basis of the family history or test results may not be justified in every case.
      In conclusion, population screening before OCP use to withhold oral contraception in women who test positive is not recommended. Testing may be useful in women with a family history of thrombosis, provided a family study is possible. Pretest counselling is essential.”

      Thanks for your comments!


  4. Peggy, thanks … yes, that’s pretty detailed. I was actually willing to buy the basic argument (widespread screening not useful or cost-effective) up front without seeing the numbers :)!

  5. Little late to the party, but I just wanted to thank you for this post. I’ve been on birth control for many years for hormone replacement, and blood clots are one of my big fears. It’s very reassuring to see everything laid out so clearly and matter-of-factly. I think it’s a good idea to stay watchful and pay attention to symptoms, but this goes a long way to easing some of my paranoia. 🙂

  6. In the second chart, you state that it shows the risk for users on 2nd or 3rd generation Levonorgestrel and norgestimate pills.

    What about Northindrone ,1st generation progestin ?

  7. Ali –

    Not many studies reference norethindrone, but in this recent study it came in around the same or lower than levonorgestrel

    Remember that these differences between progestins are overall small, because the overall risk is small to start with.

    Hope this helps. Thanks for reading.


  8. Hi
    I’ve just found out I am Leiden v factor heterozygous as i had several tests to determine if i was at risk from strokes as several of my close family have had them…and was advised to have my daughter tested now that we have and her results are the same we have been advised that the pill is out of the question for her….how do you tell a 17 year old who no doubt will become sexually active soon that she cannot go on the pill what are the options. I am at a loss. I went on the pill at 17 with no problems have had 3 children surgery and have flown all over the world..I guess I have been lucky not to have any Dvts …what is the best option for my daughter now.

  9. Lisa, your daughter still has lots of options she just shouldn’t take estogen containing methods. She can use iuds, there is even a copper one with no hormones, she can use depo provera,implanon,the mini pill. I assure you she will find one that works for her. I hope that helps

  10. What about age factor? Women over 35, birth control clot risk vs. pregnancy related clots with increased age? Non-smoker.

  11. Hi, I am a young female living in Cape Town, South Africa. I started taking Yasmin at the age of 16 to help treat bad skin. I was never made aware of any of the risks involved. Last year, at the age of 26, I started seeing a strange dark or blind spot. I was sent to eye specialists and then onto a neurologist. I had suffered a minor stroke. If I could go back and I knew this information, I would NEVER take the chance of using hormone related contraceptives. You never know when YOU might just be that one in a million…

    I am now in search of alternative contraceptives as I have tried the copper loop, I am now in search of a diaphragm which seems to be like trying to find a needle in a hay stack in SA.

    I wish I had known what I know now, and am so very grateful that the stroke I suffered was not worse.

  12. You are a wonderful knowledgeable writer and I thank you. So what would you reccommend for an 18 year old who is not sexually active but has very irregular periods. The gyno rx the birth control pills. I took provera as a child. We are testing to see if she has PCoS. Is there a safe progesterone she can use to bring on her periods? I would rather her not take ANY unnecessary hormones. Thanks for your wisdom.

    • This is a late reply but the main treatment of PCOS is to reverse the effects of hyperinsulinemia. Diet, exercise, wt loss and metformin. Hormonal management hasn’t been first line treatment for a couple decades.

      • Paul –

        Thanks for your comment, but I’d disagree with you on one statement you made. Use of the birth control pill remains a very viable first line treatment for women with PCOS, used in conjunction with lifestyle approaches and sometimes metformin to decrease weight and lower insulin levels. The pill is especially helpful in women who need contraception, have dysfunctional heavy bleeding or if hirsutism is an issue. Unfortunately, Metformin is not very effective at treating hirsutism. Nothing about the pill precludes lifestyle (ie diet and exercise) approaches, which you correctly state are first line for all women with pcos and obesity – or metformin. In some women we often use these all together.

        In non-sexually active adolescents with pcos without heavy dysfunctional bleeding, lifestyle alone or metformin would absolutely be the way to go. But if there is concommittent heavy dysfunctional bleeding, then use of progestin for inducing menses or the OCP is not a treatment of the past – it is a very viable option until the other treatments kick in so to speak.

        Thanks again for weighing in.


  13. Hi, I am a 23 years old and have been on Yaz for 4 years. I always struggled with acne and Yaz made my skin perfect. I gained no weight and suffered no other side-effects while on Yaz. 3 months ago I got DVT in my upper right arm and right part of my chest. I am right handed. There are no history of DVT in our family and I went for scans and blood tests and they could not find anything wrong with me as to why I got DVT. So my GP said that I should stop using Yaz and any other oral contraceptive pills.

    Now that I am no longer on Yaz, I’ve actually gained weight and my skin looks terrible! I am depressed all the time and my studies suffer because of this. I have anxiety problems and are stressed even when I am relaxing. My normal heart rate varies between 83 and 113 beats per minute.

    My father got DVT once in his late 50’s because his cycling shirt sat too tight. Then he was so careful with his right arm that he put too much strain on his left arm and got DVT again in his left arm when he washed the car. I read that people tend to get DVT in their dominant arm, but I don’t remember whether I did something to damage my veins.

    So my question is, do you think that Yaz caused my DVT, because I can not cope without it. And will consulting with a gynecologist help or just deplete my medical aid? I am still a student with little income.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.