Atrial fibrillation affects roughly 6 million people in the United States, and for decades warfarin has been the mainstay of stroke prevention. But warfarin requires regular blood tests to measure the international normalized ratio (INR), a marker of how thin the blood is. Without those checks, the drug can become ineffective or dangerous. A growing body of evidence suggests that a substantial fraction of patients on warfarin are not getting monitored at all — and the consequences are severe.

According to a 2023 analysis of Medicare claims, nearly half of atrial fibrillation patients prescribed warfarin went six months or longer without an INR test. That finding, published in the Journal of the American Heart Association, echoes earlier work from the Veterans Health Administration and large integrated health systems. The gap is not a niche problem; it is a systemic failure that affects hundreds of thousands of patients.

This article examines why INR monitoring slips through the cracks, what the evidence says about the risks, and what could be done to close the gap. The focus is on the US context, where fragmented care and variable access to monitoring create a situation where multiple barriers — transportation difficulties, forgetfulness, limited clinic hours, and lack of automated reminders — combine to cause missed tests.

The Gap Between Guidelines and Reality

Clinical practice guidelines from the American Heart Association and the American College of Cardiology recommend that patients on warfarin have their INR checked at least once a month. For patients with unstable INRs or those starting therapy, weekly or even more frequent checks are advised. The rationale is straightforward: warfarin has a narrow therapeutic window. An INR below 2.0 increases stroke risk; an INR above 3.0 raises bleeding risk. Regular monitoring allows dose adjustments to keep the INR in the 2.0–3.0 range.

Yet the real-world data tell a different story. The Medicare analysis found that 47% of warfarin users had no INR test recorded in a six-month window. Among patients with a prior stroke or transient ischemic attack — those at highest risk — the rate was only slightly better, at 41%. These numbers align with a 2021 study from the University of Michigan that used claims data from a large commercial insurer and found that 38% of patients had gaps of at least three months.

The consequences are not abstract. A retrospective cohort study from Kaiser Permanente Northern California, published in 2019, reported that patients with gaps in INR monitoring had a 1.8-fold higher risk of ischemic stroke compared with those who were monitored regularly. The risk was even higher — roughly 2.5-fold — among patients whose INR was subtherapeutic at the time of the gap. Missed monitoring does not just mean a missed test; it means a missed opportunity to prevent a stroke.

One example of the lack of tracking comes from a 2022 report by the Office of the Inspector General for the US Department of Health and Human Services, which found that Medicare does not routinely assess anticoagulation monitoring as a quality metric. The report noted that fewer than 10% of Medicare Advantage plans had any program in place to identify patients with gaps in INR testing. Without systematic tracking, the problem remains invisible to policymakers and the public.

Why INR Monitoring Slips Through the Cracks

The reasons patients miss INR checks are multiple and overlapping. For many, the sheer burden of frequent lab visits is a barrier. Warfarin users typically need a test every two to four weeks, which means traveling to a clinic or laboratory, waiting for a venipuncture, and then waiting for results. For older adults — who make up the majority of atrial fibrillation patients — mobility limitations, lack of transportation, or dependence on family members for rides can make each visit a logistical challenge.

Patient forgetfulness is another factor. Unlike daily medications, which can be tied to a routine, INR tests are scheduled at variable intervals. Without a strong reminder system, appointments slip. A 2022 survey of patients in an urban anticoagulation clinic found that 28% reported forgetting to schedule their next test, and 19% said they did not understand why monthly testing was necessary. Health literacy plays a role: patients with limited English proficiency or lower educational attainment were more likely to have gaps.

Clinic-level factors also contribute. Anticoagulation clinics, which specialize in managing warfarin, are not universally available. In rural areas, patients may have to drive an hour or more to the nearest lab. Even when clinics exist, staffing shortages can limit hours. A 2020 report from the American College of Chest Physicians noted that many anticoagulation clinics operate with a single pharmacist or nurse, making it difficult to accommodate walk-ins or call patients with reminders.

Electronic health records (EHRs) could help by generating automated alerts when a test is overdue, but most systems do not have this functionality. A 2021 study of EHRs across 12 US health systems found that fewer than 20% had any automated reminder for INR monitoring. Even when alerts existed, they were often overridden or ignored because clinicians received too many alerts for other tasks. The technology exists, but the implementation lags.

The Evidence: What the Trials Actually Show

The landmark trials that established warfarin's efficacy for stroke prevention in atrial fibrillation — such as the BAATAF and SPAF studies from the 1990s — were conducted in tightly controlled settings where INR monitoring was frequent and rigorous. Patients in those trials had their INR checked weekly or biweekly, and time in therapeutic range (TTR) exceeded 70%. The real-world TTR is often below 60%, and the gap between trial conditions and everyday practice is stark.

The RE-LY trial, published in 2009, compared dabigatran with warfarin and found that dabigatran was noninferior for stroke prevention with lower rates of intracranial hemorrhage. But the trial also highlighted the monitoring burden: warfarin patients in RE-LY had a median TTR of 64%, which is higher than what many real-world populations achieve. Critics have argued that dabigatran's advantage was partly due to the difficulty of maintaining good INR control in practice, not just the drug's pharmacology.

The ARISTOTLE trial, which compared apixaban with warfarin, reported similar findings. Apixaban reduced stroke and systemic embolism by 21% and major bleeding by 31% compared with warfarin. Importantly, apixaban required no routine blood monitoring. The trial's results were so favorable that apixaban and other direct oral anticoagulants (DOACs) have largely replaced warfarin in new prescriptions. But many patients remain on warfarin, and the monitoring gap persists.

Real-world analyses confirm that subtherapeutic INR is common. A 2020 study from Kaiser Permanente found that 30% of warfarin users had at least one INR below 2.0 during a one-year period, and 12% had an INR below 1.5 — a level at which stroke risk rises sharply. The study also showed that patients with a single subtherapeutic INR had a 1.6-fold higher risk of stroke within the next 30 days. These numbers are a reminder that the problem is not just about missed tests; it is about the quality of monitoring when it does occur.

Who Is Most at Risk of Falling Off Monitoring

Older adults living alone are particularly vulnerable. Atrial fibrillation prevalence rises steeply with age, affecting roughly 10% of people over 80. Many of these patients live alone, have multiple comorbidities, and take several medications. A 2022 analysis of Medicare data found that patients aged 85 and older had the highest rate of monitoring gaps, with 52% going six months without an INR test. Social isolation likely contributes: without a family member to remind or transport them, appointments are easily missed.

Patients with lower health literacy or limited English proficiency also face disproportionate barriers. A 2021 study in the Journal of General Internal Medicine examined a cohort of warfarin users in a safety-net hospital and found that patients with limited English proficiency had 40% lower odds of having an INR test within the recommended interval. Written instructions about testing schedules and dose adjustments are often provided only in English, and interpreter services may not be available at lab visits.

Rural populations are another high-risk group. The closure of rural hospitals and clinics over the past decade has reduced access to laboratory services. A 2023 report from the University of Minnesota Rural Health Research Center estimated that 15% of rural residents live more than 30 minutes from the nearest lab that can perform INR testing. For patients who rely on public transportation or cannot drive, that distance can be prohibitive.

Finally, patients with cognitive impairment — including dementia, which affects up to 30% of atrial fibrillation patients over 80 — are at extreme risk. They may forget appointments, misunderstand dosing instructions, or be unable to communicate symptoms of bleeding or stroke. Yet these patients are often excluded from clinical trials, so the evidence base for managing their anticoagulation is thin.

Technology That Could Close the Gap

Point-of-care INR devices, such as the CoaguChek and INRatio, allow patients to test their INR at home with a fingerstick sample. These devices have been available for years and are widely used in some European countries. In the US, however, adoption has been slow. Medicare covers home INR testing only for patients who meet specific criteria, such as those on long-term warfarin who have demonstrated ability to self-test. A 2021 analysis found that fewer than 5% of eligible Medicare beneficiaries used home testing.

Telemonitoring programs that combine home testing with remote oversight by a pharmacist or nurse have shown promise. A 2022 randomized trial from the Mayo Clinic assigned patients to either standard clinic-based monitoring or a telemonitoring program that included a home INR device and weekly phone check-ins. The telemonitoring group achieved a TTR of 72% compared with 62% in the clinic group, and had 15% fewer hospitalizations for bleeding or stroke. The effect was largest among patients who lived more than 20 miles from the clinic.

Smartphone apps for self-testing reminders are another low-cost intervention. Apps like MyINR and Warfarin Manager allow patients to log their INR values, set reminders for next tests, and receive dose adjustment suggestions. But adoption remains low. A 2023 survey of anticoagulation clinic patients found that only 12% used any app to manage their warfarin. Barriers include lack of awareness, difficulty navigating the app, and concerns about data privacy.

Direct oral anticoagulants (DOACs) — including apixaban, rivaroxaban, edoxaban, and dabigatran — eliminate the need for routine monitoring altogether. For patients who are good candidates, switching to a DOAC is the simplest way to close the monitoring gap. Yet many patients remain on warfarin. The reasons are complex and are explored in the next section.

Why Many Clinicians Still Prescribe Warfarin

Cost is a major factor. Warfarin costs pennies per day — typically $4 to $10 per month without insurance — while DOACs can cost $300 to $500 per month. For patients on high-deductible insurance plans or without prescription coverage, the price difference is prohibitive. Even with Medicare Part D, copays for DOACs can be $50 to $100 per month, compared with $5 for warfarin. Some patient assistance programs exist, but they are not universally accessible.

Reversibility is another consideration. Warfarin's anticoagulant effect can be reversed with vitamin K, and in emergencies, with fresh frozen plasma or prothrombin complex concentrate. DOACs have reversal agents — idarucizumab for dabigatran and andexanet alfa for factor Xa inhibitors — but these are expensive, not always available in smaller hospitals, and may not work as quickly as vitamin K. For patients at high risk of bleeding, such as those with a history of gastrointestinal bleeding or falls, warfarin's reversibility can be reassuring.

Renal clearance concerns also drive warfarin use. DOACs are partially cleared by the kidneys, and their dosing must be adjusted in patients with chronic kidney disease. For patients with creatinine clearance below 15 mL/min, DOACs are generally not recommended, and warfarin remains the only option. Given that many atrial fibrillation patients are elderly and have some degree of renal impairment, this is not a small population.

Finally, patient preference matters. Some patients have been on warfarin for years and feel comfortable with the routine. They may distrust newer drugs or prefer the perceived control of seeing their INR numbers. A 2022 qualitative study of warfarin users found that some viewed the INR test as a "safety check" and felt anxious about the idea of taking a medication without monitoring. Shared decision-making requires respecting these preferences, even when the evidence favors a switch.

Trade-Offs and Counter-Arguments: The Limitations of DOACs

While DOACs offer the clear advantage of eliminating routine monitoring, they are not without their own challenges. Adherence to DOACs is a concern because there is no regular blood test to remind patients to take their medication. A 2021 analysis of pharmacy claims found that 25% of patients prescribed DOACs had lapses in refills of 30 days or more within the first year. Without monitoring, a missed dose goes undetected until a stroke or embolism occurs. In contrast, a missed warfarin dose is often caught at the next INR check, allowing for dose adjustment.

Cost is a significant barrier for DOACs, as discussed, but even when cost is not an issue, some patients experience side effects that are less common with warfarin. For example, dabigatran is associated with higher rates of dyspepsia (affecting up to 10% of users), and rivaroxaban taken once daily may have a higher risk of gastrointestinal bleeding compared with apixaban. Warfarin, while requiring monitoring, has a longer track record and a well-understood side effect profile.

Another counter-argument is that the monitoring gap for warfarin could be addressed through systemic improvements, rather than abandoning the drug. For some patients, especially those with stable INRs and good access to care, warfarin can be as safe and effective as DOACs. The key is to identify patients who are likely to adhere to monitoring and support them, while switching those who are not. A one-size-fits-all approach — either universal warfarin or universal DOACs — would not serve all patients equally.

Finally, there are clinical scenarios where warfarin is preferred despite the monitoring burden. Patients with mechanical heart valves, for instance, require warfarin because DOACs are contraindicated. Similarly, patients with antiphospholipid syndrome, especially those with a history of thrombosis, may have better outcomes with warfarin. The monitoring gap in these populations is particularly concerning because there is no alternative therapy.

What a Safer System Would Look Like

A safer system would start with a national registry that tracks INR monitoring adherence and sends alerts when patients are overdue. The Veterans Health Administration has a centralized anticoagulation tracking system that has been associated with higher TTR and fewer adverse events. Expanding a similar system to Medicare and private insurers would require investment in data infrastructure, but the potential savings from prevented strokes could offset the costs.

Pharmacist-led anticoagulation clinics have been shown to improve monitoring adherence. A 2021 meta-analysis of 18 studies found that pharmacist-managed warfarin therapy increased TTR by an average of 12 percentage points and reduced major bleeding by 30%. Yet such clinics are not standard in many health systems, particularly in rural areas. Expanding pharmacist scope of practice and reimbursement for anticoagulation management could help.

Shared decision-making tools that explicitly discuss the trade-offs between warfarin and DOACs — including the monitoring burden — could help patients make informed choices. A 2023 pilot program at the University of Colorado used a decision aid that included a pictogram of the number of clinic visits required per year for warfarin versus a DOAC. Patients who used the aid were more likely to choose a DOAC, but those who chose warfarin reported higher satisfaction with their decision.

Medicare pilot programs for mailed self-test kits have shown encouraging results. A 2020 demonstration project in rural Montana provided home INR devices and test strips to 200 warfarin users, along with telephonic support from a pharmacist. Over 12 months, the proportion of patients with at least one monitoring gap longer than 60 days fell from 45% to 12%. The program was cost-neutral after accounting for reduced hospitalizations. Scaling such programs nationally would require changes to Medicare coverage rules and investment in device distribution.

Ultimately, the goal should be to ensure that every patient on warfarin has a reliable path to regular INR monitoring — whether through home testing, a well-staffed clinic, or a switch to a DOAC. The current reality, in which half of patients go unsupervised for six months, is a failure of the system, not of individual clinicians or patients. Closing the gap will require changes at every level: policy, payment, technology, and clinical practice.