Which Blood-Thinning Medication Is Right for You?

Atrial fibrillation (AF) is a medical condition where the two upper chambers of the heart contract too quickly and irregularly. The irregular heartbeat causes blood to pool, which makes it more likely that a person will develop blood clots. Due to this possibility of blood clots, people with AF are 4-5 times more likely to have a stroke than a person without AF.1 In order to reduce the chance that a person with AF will have a stroke, their doctor may prescribe a blood-thinning medication (also called an anticoagulant). This article will explain the risks and benefits of some commonly prescribed blood-thinners. 

Benefits and Risks of Blood Thinning Medications

Over the years, the FDA has approved several blood-thinning medications. Warfarin (brand name Coumadin) was first approved in 1954, and it is still used today. It works by affecting how the body activates vitamin K, which is important for blood clotting.2 In more recent years, some newer medications have been approved that work without affecting vitamin K; they are called non-vitamin K oral anticoagulants (NOAC). Some examples of NOACs include dabigatran (brand name Pradaxa), approved in 2010, and rivaroxaban (Xarelto), approved in 2011. 

All these drugs share some of the same risks and benefits. The greatest benefit is reducing the likelihood that a person will have a stroke. The greatest risk is the risk of serious, potentially fatal, bleeding. If someone taking a blood thinner also takes a nonsteroidal anti-inflammatory drug (NSAID), such as ibuprofen, naproxen sodium, or aspirin, there is an increased chance that they will experience bleeding. For example, one meta-analysis combining the results of seven other studies found that patients taking warfarin were twice as likely to have gastrointestinal bleeding if they also took NSAIDs.3 In fact, some studies included in the analysis found a difference after only 7 days of taking both warfarin and NSAIDs. One of the studies analyzed compared over 8,000 patients taking warfarin to patients taking warfarin plus an NSAID, and the researchers found that there were only 7 gastrointestinal bleeds per 1,000 patient years for patients taking warfarin alone, compared with 28 gastrointestinal bleeds for patients taking both warfarin and an NSAID.4 This is equivalent to 0.7 percent bleeds per year compared to 2.8% per year. “Patient years” is a statistic that adds the combined number of years that all patients in the study took a particular drug. For example, if a study had 500 patients and followed all of them for 2 years, the researchers would say they had 1,000 “patient years” of data. The researchers then count how many of a specific type of health problem occurred in that total sample of patients, and divide it by the number of patients times the number of years studied. Due to this increased chance for bleeding, patients taking blood thinners are encouraged to take acetaminophen for pain instead.

Patients wondering if a medication they are considering taking contains an NSAID can check the chart on this webpage. 

Some of the benefits of the drugs are shared as well; all of the drugs discussed in the article help reduce the likelihood that a person will have a stroke. However, there are some differences between the drugs, which are discussed in more detail below. 

Warfarin (Coumadin)

After nearly 70 years of use, warfarin has stood the test of time. It has been shown to help prevent blood clots, and is commonly prescribed for those with AF. Like all blood thinning medications, warfarin can increase the chance of bleeding, such as intracranial bleeding (inside the skull) bleeding or gastrointestinal bleeding.5 Warfarin can interact with other drugs besides NSAIDs as well. If you are considering taking warfarin, talk with your healthcare provider about all medications you take, including supplements. 

Patients taking warfarin need regular blood tests called the International Normalized Ratio (INR) test. Patients can either go to the doctor for the tests or self-monitor their blood clotting time at home using a small drop of blood from a simple finger stick. The results of an INR test help healthcare providers know if their patients are on the right dose, or if there is a risk for either blood clots or bleeding. 

Patients taking warfarin also have to be careful about what they eat and drink.. Too much vitamin K can affect how warfarin works, so patients need to be careful about how much they eat of vitamin K-rich foods, such as dark leafy greens.6 Patients should also try to limit how much alcohol they drink because alcohol can make it more likely that a patient will experience bleeding.6

Dabigatran (Pradaxa)

Pradaxa is one of the NOAC drugs described above, so that a patient taking the drug does not have to watch what they eat the way someone taking warfarin does. Also, patients taking dabigatran do not need to have regular blood tests the way patients taking warfarin do, and it has fewer interactions with other drugs. 

Observational research has found that patients taking dabigatran were less likely to have a stroke than patients taking warfarin, but they were also more likely to experience gastrointestinal bleeding.7,8 An observational study evaluates patients who are already taking particular drugs and compares them. Observational studies can be useful for gathering large amounts of “real world” data, although the studies have limitations. For example, if medical records indicate that patients were prescribed a medication, we don’t know and can’t ask if they took the medication correctly. Those are questions that can be asked in a clinical trial. For example, one observational study compared over 60,000 AF patients taking dabigatran to 60,000 similar patients taking warfarin over the course of 2 years. For the patients taking Pradaxa, there were 11.3 strokes per 1,000 patient years, compared with 13.9 strokes per 1,000 patient years for those taking warfarin, (the equivalent of 1.1% of patients per year compared to 1.4% per year). That same study found that gastrointestinal bleeds happened 34.2 times per 1,000 patient years (3.4% per year) for those taking dabigatran, compared with 26.5 (2.7% per year) for patients taking warfarin.8 Another study comparing almost 8,000 AF patients taking Pradaxa to the same number of similar patients taking warfarin found that there was not a difference between the 2 drugs in the rates of gastrointestinal bleeds, although patients over 75 taking Pradaxa were the most likely to experience gastrointestinal bleeding.9 The company that makes Pradaxa has faced lawsuits from families of patients who died from gastrointestinal bleeding based on lack of warnings about bleeding.10 However, while patients taking Pradaxa are more likely to have gastrointestinal bleeding than those taking warfarin, they are less likely to experience intracranial bleeding (1% per year for patients taking warfarin vs. 0.3% per year for patients taking Pradaxa in the study of 120,000 patients).8

If someone taking a blood thinning medication needs to get surgery or has an injury that can cause bleeding, there is a danger of excess bleeding because the medication affects how their blood clots. When Pradaxa was first sold, there were no drugs approved to reverse its effect, which meant that patients had a risk of bleeding to death. However, in 2015, the FDA approved Praxbind (idarucizumab) for reversing the effect of dabigatran. A study done by the drug company was small (only 123 patients), 85% of the patients were White,11 and 85% of patients experienced adverse effects, including 2% experiencing the serious adverse effects of pulmonary embolism and another 2% experiencing deep vein thrombosis.12 More recent research looked at patients taking Pradaxa who were taking Praxbind to reverse its effect in order to stop life-threatening bleeding or prepare for surgery.13 The study found that Praxbind was effective at reversing the effects of dabigatran and helping the blood to clot, although almost 7% of patients experienced dangerous thrombotic events (such as pulmonary embolism) within 90 days of receiving Praxbind, and about 2% experienced delirium. This study was not a placebo controlled trial, which the researchers justify because it would be unethical to randomly assign some patients to not receive medication that could prevent severe bleeding or bleeding to death.14 

Rivaroxaban (Xarelto)

Xarelto is another type of NOAC drug. Like Pradaxa, it has fewer interactions with other medications than warfarin does, and patients do not need to watch their diet the same way or receive regular blood tests. Several studies have compared Xarelto to warfarin. A clinical trial of more than 14,000 AF patients randomly assigned to take either Xarelto or warfarin found that 1.7% per year of Xarelto patients had strokes and systemic embolism compared to 2.2% patients per year taking warfarin.15 That study combined the results for strokes and embolism. What about the chances for strokes alone? A study of more than 5,000 AF patients taking Xarelto with demographically similar patients taking warfarin found that there was no statistically significant difference in the rates of stroke between the drugs, with less than 1% of patients taking either drug having a stroke each year.9

About 25% of patients with non-valvular atrial fibrillation have diabetes as well. A study of over 21,000 AF patients with diabetes compared kidney health outcomes between patients taking Xarelto to warfarin and found that patients taking warfarin were twice as likely to go to the ER or be admitted to the hospital for acute kidney injury; almost 8% of Pradaxa patients developed acute kidney injury, compared to 13.5% of patients taking warfarin. Warfarin patients were also more likely to go on to be diagnosed with stage 5 chronic kidney disease or undergo dialysis (3.7% per year for Pradaxa compared to 6% for warfarin patients).16 

Like with all blood thinners, patients taking Xarelto have a risk for major bleeding. When it was first sold, there was no medication to specifically stop bleeding, which meant taking the drug could make something like a minor fall into a dangerous situation because the patient would bleed uncontrollably. However, since 2018, patients taking Pradaxa can take a medication called andexanet alfa to reverse the blood-thinning effects. Andexanet alfa is not without its risks, such as blood clots and heart attacks, and it carries a boxed warning from the FDA about these risks.17 For example, one study found that 11% of patients experienced thrombosis within 30 days of being given the drug.18 

An observational study comparing over 3,500 AF patients who were obese found that there was no statistically significant difference in major bleeding between the two groups (2.2% for Xarelto and 2.7% for warfarin).[19] However, a study of over 14,000 patients found that those taking Xarelto are slightly less likely to have intracranial bleeding (0.5%) than patients taking warfarin (0.7%).15 The observational study comparing over 5,000 matched pairs of patients taking Xarelto and warfarin did not find a statistically significant difference in gastrointestinal bleeds between the patient groups (2.8% per year vs 3.1% per year), but it did find more gastrointestinal bleeding for Xarelto patients who were over age 75 compared to warfarin patients over 75 (5.5% vs. 4.1%).9 The study of over 14,000 patients found that Xarelto patients were less likely to die from major bleeding (0.2%) than patients taking warfarin (0.5%).15 However, other research has found that the patients taking either drug are equally likely to die from bleeding (less than 1 fatal bleed per 100 patient years for either drug).20 These differences in results may be due to differences in the ages of patients studied as well as the challenges of drawing statistical conclusions comparing the small number of patients who die from either type of bleeding.

Which Type of Drug Is Best?

In 2019, the American College of Cardiology, the American Heart Association, and the Heart Rhythm Society released a joint statement that recommended that NOAC drugs (such as dabigatran and rivaroxaban) are recommended over warfarin, due to some evidence that they are better at preventing strokes and lead to a lower chance of serious bleeding. However, patients who have moderate-to-severe mitral stenosis (narrow mitral valve) or have a mechanical heart valve are recommended to take warfarin instead.21 It is important to note that the American Heart Association has received financial support from the drug companies that make rivaroxaban and dabigatran, taking a total of almost 2 million dollars from the 2 companies.22

All of the drugs discussed above have shown to be effective blood-thinners that reduce the risk of stroke for AF patients. Observational studies have found that overall, patients taking NOACs (like dabigatran and rivaroxaban) have a similar chance of having a stroke than patients taking warfarin (less than 1 stroke per 100 patient years), though as described above, they have a lower chance of developing intracranial bleeding and a higher chance of developing gastrointestinal bleeding.9,23 All of the drugs carry risks for bleeding, and warfarin’s blood thinning effects could possibly be easier to reverse. As noted above, drugs exist to reverse the effects of the NOACs, although those drugs carry side effects as well.

A generic version of warfarin is available, while generics of Pradaxa and Xarelto are not yet on the market. This means that it is much less expensive to fill a warfarin prescription. However, all of these drugs have additional expenses, especially for obese patients. For example, a study of more than 7,000 obese patients half of whom took warfarin and half of whom took Xarelto found that over approximately 10 months, 50% of those taking Xarelto were hospitalized, compared with 54% taking warfarin, and those taking Xarelto stayed an average of 7.5 days in the hospital, compared with the average of 9.1 days that those taking warfarin stayed.19

Overall, no drug is without risks and side effects. All of the drugs discussed in this article are effective at reducing the chance that an AF patient will have a stroke, but they have different risks. Warfarin requires regular blood monitoring, and patients have to be careful what they eat or drink, or what other medications or supplements they take. The NOACs do not have as many drug interactions or a need for monitoring blood or diet, but they have risks when it comes to needing to reverse their effects if a patient is bleeding or needs surgery. 

The Bottom Line

There are several blood-thinning medications used to reduce the chance that a person with atrial fibrillation will have a stroke, and this article focused on three of them. None of these drugs is without risks and side effects. All of the drugs discussed in this article are effective at reducing the chance that an AF patient will have a stroke, but they have different risks. Warfarin requires regular blood monitoring, and patients have to be careful what they eat or drink, or what other medications or supplements they take. The NOACs do not have as many drug interactions or a need for monitoring blood or diet, but they have risks when it comes to needing to reverse their effects if a patient is bleeding or needs surgery. Warfarin is an older medication, and while effective, it has some drawbacks such as the need to watch one’s diet and the need to get regular blood testing. Newer medications such as Pradaxa and Xarelto do not require the same level of patient monitoring, but they have their own risks for bleeding. Patients with atrial fibrillation should speak with their healthcare providers about the risks and benefits of the blood-thinning medications they are considering.

All articles are reviewed and approved by Dr. Diana Zuckerman and other senior staff.

The National Center for Health Research is a nonprofit, nonpartisan research, education and advocacy organization that analyzes and explains the latest medical research and speaks out on policies and programs. We do not accept funding from pharmaceutical companies or medical device manufacturers. Find out how you can support us here.


  1. Virani SS, Alonso A, Aparicio HJ, Benjamin EJ, Bittencourt MS, Callaway CW, Carson AP, Chamberlain AM, Cheng S, Delling FN, Elkind MS. Heart disease and stroke statistics—2021 update: a report from the American Heart Association. Circulation. 2021 Feb 23;143(8):e254-743.
  2. Ageno W, Gallus AS, Wittkowsky A, Crowther M, Hylek EM, Palareti G. Oral anticoagulant therapy: antithrombotic therapy and prevention of thrombosis: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2012 Feb 1;141(2):e44S-88S.
  3. Zapata LV, Hansten PD, Panic J, Horn JR, Boyce RD, Gephart S, Subbian V, Romero A, Malone DC. Risk of bleeding with exposure to warfarin and nonsteroidal anti-inflammatory drugs: a systematic review and meta-analysis. Thrombosis and haemostasis. 2020;120(07):1066-74.
  4. Cheetham TC, Levy G, Niu F, Bixler F. Gastrointestinal safety of nonsteroidal antiinflammatory drugs and selective cyclooxygenase-2 inhibitors in patients on warfarin. Annals of Pharmacotherapy. 2009;43(11):1765-73.
  5. American Society of Health-System Pharmacists. Warfarin. Drugs.com. https://www.drugs.com/monograph/warfarin.html. Updated October 2021. 
  6. Sheps S G. Warfarin diet: What foods should I avoid?. Mayo Clinic. https://www.mayoclinic.org/diseases-conditions/thrombophlebitis/expert-answers/warfarin/faq-20058443. February 2021. 
  7. Lauffenburger JC, Farley JF, Gehi AK, Rhoney DH, Brookhart MA, Fang G. Effectiveness and safety of dabigatran and warfarin in real‐world US patients with non‐valvular atrial fibrillation: a retrospective cohort study. Journal of the American Heart Association. 2015;4(4):e001798. 
  8. Graham DJ, Reichman ME, Wernecke M, Zhang R, Southworth MR, Levenson M, Sheu TC, Mott K, Goulding MR, Houstoun M, MaCurdy TE. Cardiovascular, bleeding, and mortality risks in elderly Medicare patients treated with dabigatran or warfarin for nonvalvular atrial fibrillation. Circulation. 2015;131(2):157-64.
  9. Abraham NS, Singh S, Alexander GC, Heien H, Haas LR, Crown W, Shah ND. Comparative risk of gastrointestinal bleeding with dabigatran, rivaroxaban, and warfarin: population based cohort study. BMJ. 2015;350.
  10. Silvestrini E. Pradaxa Lawsuits. Drug Watch. https://www.drugwatch.com/pradaxa/lawsuits/
  11. U.S. Food and Drug Administration. Drug Trials Snapshots: PRAXBIND. https://www.fda.gov/drugs/drug-approvals-and-databases/drug-trials-snapshots-praxbind. May 2017. 
  12. U.S. Food and Drug Administration. Summary Review for Regulatory Action: 761025Orig1s000.  https://www.accessdata.fda.gov/drugsatfda_docs/nda/2015/761025Orig1s000SumR.pdf. 2015. 
  13. Pollack Jr CV, Reilly PA, Van Ryn J, Eikelboom JW, Glund S, Bernstein RA, Dubiel R, Huisman MV, Hylek EM, Kam CW, Kamphuisen PW. Idarucizumab for dabigatran reversal—full cohort analysis. New England Journal of Medicine. 2017;377(5):431-41.
  14. Pollack Jr CV, Reilly PA, Eikelboom J, Glund S, Verhamme P, Bernstein RA, Dubiel R, Huisman MV, Hylek EM, Kamphuisen PW, Kreuzer J. Idarucizumab for dabigatran reversal. New England Journal of Medicine. 2015;373(6):511-20.
  15. Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, Breithardt G, Halperin JL, Hankey GJ, Piccini JP, Becker RC. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. New England Journal of Medicine. 2011;365(10):883-91.
  16. Hernandez AV, Bradley G, Khan M, Fratoni A, Gasparini A, Roman YM, Bunz TJ, Eriksson D, Meinecke AK, Coleman CI. Rivaroxaban vs. warfarin and renal outcomes in non-valvular atrial fibrillation patients with diabetes. European Heart Journal-Quality of Care and Clinical Outcomes. 2020;6(4):301-7.
  17. U.S. Food and Drug Administration. Package insert ANDEXXA. https://www.fda.gov/media/113279/download. 2018.  
  18. Reed M, Tadi P, Nicolas D. Andexanet Alfa. [Updated 2021 Sep 29]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2022. Available from: https://www.ncbi.nlm.nih.gov/books/NBK519499/
  19. Peterson ED, Ashton V, Chen YW, Wu B, Spyropoulos AC. Comparative effectiveness, safety, and costs of rivaroxaban and warfarin among morbidly obese patients with atrial fibrillation. American Heart Journal. 2019;212:113-9.
  20. Sherwood MW, Nessel CC, Hellkamp AS, Mahaffey KW, Piccini JP, Suh EY, Becker RC, Singer DE, Halperin JL, Hankey GJ, Berkowitz SD. Gastrointestinal bleeding in patients with atrial fibrillation treated with rivaroxaban or warfarin: ROCKET AF trial. Journal of the American College of Cardiology. 2015;66(21):2271-81.
  21. January CT, Wann LS, Calkins H, Chen LY, Cigarroa JE, Cleveland JC, Ellinor PT, Ezekowitz MD, Field ME, Furie KL, Heidenreich PA. 2019 AHA/ACC/HRS focused update of the 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. Journal of the American College of Cardiology. 2019;74(1):104-32.
  22. American Heart Association. 2017-2018 American Heart Association Support from Pharmaceutical and Biotech Companies, Device Manufacturers and Health Insurance Providers. https://www.heart.org/-/media/files/finance/pharma-device-insurance-corporate-funding-fiscal-20172018.pdf
  23. Bengtson LG, Lutsey PL, Chen LY, MacLehose RF, Alonso A. Comparative effectiveness of dabigatran and rivaroxaban versus warfarin for the treatment of non-valvular atrial fibrillation. Journal of Cardiology. 2017;69(6):868-76.