Results Summary
What was the research about?
Each year, more than 500,000 people in the United States are hospitalized for blood clots that form in their veins. Many people who have a clot will have another within 10 years. Patients can take medicines called blood thinners to treat and prevent new clots. But blood thinners may cause problems, like severe bleeding. Questions remain about how long people should take these medicines to prevent future clots.
In this study, the research team compared ways to prevent blood clots in people who have already had one. The team compared patients who took blood thinners after a clot for 3 months with patients who took them for more than 3 months and up to 12 months. The team also compared the safety of taking three common blood thinners—warfarin, apixaban, and rivaroxaban—for more than 3 and up to 12 months.
The research team looked at risk of a hospital stay for new blood clots or bleeding.
What were the results?
Compared with patients who stopped taking blood thinners after three months, patients who took blood thinners for more than three months had:
- A lower risk of a hospital stay to treat new blood clots within six months
- A higher risk of a hospital stay to treat severe bleeding within six months
Statistical tests found that, among patients who took blood thinners for more than three months, patients taking:
- Apixaban had a lower risk of a hospital stay to treat new blood clots than patients taking warfarin
- Apixaban and patients taking rivaroxaban had a similar risk of a hospital stay to treat new blood clots
- Rivaroxaban and patients taking warfarin had a similar risk of a hospital stay to treat new blood clots
The risk of a hospital stay to treat severe bleeding didn’t differ among patients who took warfarin, apixaban, or rivaroxaban.
Who was in the study?
The study included health records from three nationwide health insurance claims databases for 90,921 adult patients. All patients had been in the hospital for a blood clot and took blood thinners after leaving the hospital for at least 90 days. The average age was 70, and 57 percent were women.
What did the research team do?
The research team looked at health records from 2010 to 2021 to find patients who had been hospitalized for a blood clot and who hadn’t used blood thinners before the clot. Then the team tracked hospital stays for patients who filled prescriptions for a blood thinner and took it for at least 90 days.
Patients with prior blood clots and doctors helped design the study.
What were the limits of the study?
The study didn’t assign patients by chance to length of treatment. Differences between patients who extended treatment and who stopped treatment may have affected study results.
Future research could determine the best length of time for patients to take blood thinners to reduce the risk of future blood clots and bleeding.
How can people use the results?
Patients and doctors can use these results when considering treatments for blood clots.
| PCORI identified long-term use of blood thinner medicines as an important research topic. Blood thinners help prevent blood clots but may cause health problems if taken for a long time. Patients, clinicians, and others wanted to learn: How do different long-term blood thinner treatments compare for patients who have had a blood clot? To help answer this question, PCORI launched an initiative in 2015 on New Oral Anticoagulants (NOACs) in the Extended Treatment of Venous Thromboembolic Disease. The initiative funded this research project and others. |
Professional Abstract
Objective
(1) To compare the safety and effectiveness of extended treatment with oral anticoagulants for blood clots versus stopping treatment at 90 days in patients with a recent hospitalization for venous thromboembolic disease; (2) To compare the safety and effectiveness of extended treatment with three common oral anticoagulants
Study Design
| Design Element | Description |
|---|---|
| Design | Observational: cohort study |
| Population | Health records for 90,921 patients from 3 nationwide healthcare claims databases, with an initial hospitalization due to pulmonary embolism or deep vein thrombosis who were prescribed an oral anticoagulant after discharge and continued the treatment for at least 90 days |
| Interventions/ Comparators | Objective 1:
Objective 2: Extended treatment with one of the following:
|
| Outcomes | Hospitalization due to VTE diagnosis; hospitalization due to major bleeding |
| Timeframe | 1-year follow-up for study outcomes |
This observational cohort study compared the safety and effectiveness of extended treatment with oral anticoagulants for more than 90 days versus stopping oral anticoagulants at 90 days in patients with venous thromboembolism (VTE). The study also compared extended treatment with one of three oral anticoagulants.
Researchers examined health records from three nationwide healthcare claims databases between 2010 to 2021 to identify patients who were hospitalized due to a VTE diagnosis and who had no prior anticoagulant use. Using database records, researchers tracked safety and effectiveness outcomes.
The study included health records for 90,921 patients who were hospitalized for VTE and who completed anticoagulant treatment for at least 90 days and up to 360 days. The average age of patients was 70, and 57% were female.
Patients with prior VTE and clinicians helped design the study.
Results
Compared with patients who stopped anticoagulant treatment after 90 days, patients who extended treatment had lower rates of subsequent hospitalization due to VTE (hazard ratio [HR]=0.30; 95% confidence interval [CI]: 0.25, 0.35) but higher rates of hospitalization due to major bleeding (HR=1.16; 95% CI: 0.99, 1.37) within a median follow-up of 3.5 months.
Statistical tests found that, among the 67,720 patients who extended treatment beyond 90 days:
- Patients taking apixaban experienced lower rates of hospitalization due to VTE than did patients taking warfarin (9.8 versus 13.5 per 1,000 person years; HR=0.69; 95% CI: 0.49, 0.99).
- Patients taking apixaban and patients taking rivaroxaban had similar rates of hospitalization due to VTE (HR=0.80; 95% CI: 0.53, 1.19).
- Patients taking rivaroxaban and patients taking warfarin had similar rates of hospitalization due to VTE (HR=0.87; 95% CI: 0.65, 1.16).
Rates of hospitalizations due to major bleeding did not differ significantly among patients taking warfarin, apixaban, or rivaroxaban.
Limitations
Because this study did not use a randomized design, potential differences between patients who extended treatment and those who stopped treatment may have affected the results.
Conclusions and Relevance
In this study, patients who extended anticoagulant treatment beyond the initial 90 days had lower rates of hospitalization due to recurrent VTE and higher rates of hospitalization due to bleeding compared with those who stopped treatment after 90 days. Among patients who extended treatment beyond 90 days, patients taking apixaban had lower rates of hospitalization due to VTE compared with patients taking warfarin. Otherwise, patients taking warfarin, apixaban, or rivaroxaban did not differ significantly in rates of hospitalization due to VTE or major bleeding.
Future Research Needs
Future research could identify the optimal treatment duration for anticoagulants in patients with different risk factors for VTE recurrence and bleeding.
| PCORI identified long-term use of blood thinner medicines as an important research topic. Blood thinners help prevent blood clots but may cause health problems if taken for a long time. Patients, clinicians, and others wanted to learn: How do different long-term blood thinner treatments compare for patients who have had a blood clot? To help answer this question, PCORI launched an initiative in 2015 on New Oral Anticoagulants (NOACs) in the Extended Treatment of Venous Thromboembolic Disease. The initiative funded this research project and others. |
Final Research Report
View this project's final research report.
Journal Citations
Results of This Project
Related Journal Citations
Peer-Review Summary
Peer review of PCORI-funded research helps make sure the report presents complete, balanced, and useful information about the research. It also assesses how the project addressed PCORI’s Methodology Standards. During peer review, experts read a draft report of the research and provide comments about the report. These experts may include a scientist focused on the research topic, a specialist in research methods, a patient or caregiver, and a healthcare professional. These reviewers cannot have conflicts of interest with the study.
The peer reviewers point out where the draft report may need revision. For example, they may suggest ways to improve descriptions of the conduct of the study or to clarify the connection between results and conclusions. Sometimes, awardees revise their draft reports twice or more to address all of the reviewers’ comments.
Peer reviewers commented and the researchers made changes or provided responses. Those comments and responses included the following:
- The reviewers noted that the researchers used a composite outcome consisting of recurrent venous thromboembolism, major bleeding, and all-cause mortality for all of the data sets they reviewed even though the mortality component is only available in one of the data sets. The reviewers asked the researchers to explain how this was handled for the other two data sets. The researchers explained that mortality in the two other data sets was captured for in-patient deaths and for outpatient could be inferred from insurance disenrollment. The researchers noted in their study limitations section that the results for this composite outcome should be interpreted with caution given the limitations in capturing mortality.
- One reviewer questioned whether the study’s comparisons of treatment continuation versus treatment discontinuation both included only data for those patients who were still alive at each time point, noting lack of that data restriction could introduce bias to the results. The researchers responded by including additional information in their methods specifying that all patients were followed unless they disenrolled from insurance, there were no further data on them, or they died.
- The reviewers requested that the researchers expand their discussion to consider potential confounding, lack of focus on race and ethnicity, and lack of discussion regarding clinical implications of the study results. The researchers expanded their discussion section as suggested, including some reorganization with subheadings so that readers could follow these points more clearly.