Highlight
– Continued oral anticoagulation (OAC) after at least 90 days for unprovoked VTE was associated with a large reduction in recurrent VTE (adjusted HR 0.19) but increased major bleeding (adjusted HR 1.75), with a net clinical benefit favoring continuation.
– The absolute event rate differences translated to approximately 25 fewer recurrent VTE events but ~5 more major bleeds per 1000 person‑years among continuers; mortality was also lower with continuation.
– The net benefit was consistent across OAC types (warfarin and direct‑acting oral anticoagulants) and persisted among patients treated for up to and beyond three years.
Background
Patients with unprovoked venous thromboembolism (VTE) — deep vein thrombosis (DVT) or pulmonary embolism (PE) occurring without an identifiable transient provoking factor — face a substantial risk of recurrent thromboembolism after stopping anticoagulation. Randomized trials of extended anticoagulation using direct oral anticoagulants (DOACs) and older agents have shown reductions in recurrence but variable bleeding trade‑offs. Contemporary guideline recommendations generally endorse extended or indefinite anticoagulation for many patients with unprovoked VTE after careful individual bleeding‑risk assessment, but real‑world applicability, long‑term outcomes, and comparative effects across patient subgroups remain areas of active clinical uncertainty.
Study design
This study by Lin and colleagues (BMJ 2025) emulated a target trial using two large US claims databases: Optum Clinformatics Data Mart (2009–2025) and Medicare fee‑for‑service (2009–2022). The investigators identified adults with a first hospitalization for VTE that lacked transient reversible provoking factors who initiated oral anticoagulation (warfarin or DOAC) within 30 days and continued treatment for at least 90 days. Continued treatment was defined by filling refills; discontinuation was defined as absence of a refill within 30 days of the expected supply end.
Using propensity score one‑to‑one matching, the authors compared patients who continued versus discontinued OAC after the initial ≥90 days. Primary effectiveness and safety outcomes were hospital admission for recurrent VTE and major bleeding, respectively. Secondary outcomes included a composite net clinical benefit (recurrent VTE plus major bleeding) and all‑cause mortality. Analyses were stratified by duration of initial therapy (90–179, 180–359, 360–719, 720–1079, and ≥1080 days) and by OAC type.
Key findings
Population: The matched cohort included 30,554 pairs (mean age 73.9 years; 57.0% women). The analysis pooled data across the two databases to provide large event counts and robust subgroup estimates.
Primary outcomes
Effectiveness — recurrent VTE: Continuing anticoagulation after ≥90 days was associated with a markedly lower risk of recurrent VTE. The adjusted hazard ratio (HR) was 0.19 (95% CI 0.13 to 0.29), corresponding to an adjusted rate difference of −25.50 events per 1000 person‑years (95% CI −39.38 to −11.63). In absolute terms, the data suggest that continuing OAC prevents approximately 25 recurrent VTE hospitalizations per 1000 patient‑years compared with stopping.
Safety — major bleeding: Continued therapy increased the risk of major bleeding with an adjusted HR of 1.75 (95% CI 1.52 to 2.02) and an absolute rate difference of +4.78 per 1000 person‑years (95% CI 1.95 to 7.61). Thus the excess bleeding events were substantially fewer in absolute number than prevented recurrent VTE events.
Secondary outcomes
Net clinical benefit (composite of recurrent VTE and major bleeding) favored continuation: adjusted HR 0.39 (95% CI 0.36 to 0.42) with an adjusted rate difference of −21.01 per 1000 person‑years (95% CI −32.31 to −9.71).
All‑cause mortality was lower among continuers (adjusted HR 0.74, 95% CI 0.69 to 0.79; absolute rate difference −14.31 per 1000 person‑years, 95% CI −22.02 to −6.59).
Consistency across subgroups
The greater net clinical benefit of continued anticoagulation was consistent across OAC types (warfarin and DOACs) and across strata of initial treatment duration, including patients who had been anticoagulated for three years or longer (≥1080 days). Findings were robust in sensitivity analyses reported in the primary paper.
Clinical interpretation
From a clinician’s perspective, this large observational emulation suggests that, for patients with unprovoked VTE who have completed at least 90 days of therapy and remain at acceptably low bleeding risk, continuing oral anticoagulation yields substantial reductions in recurrent VTE and is associated with an overall net clinical benefit despite an increased risk of major bleeding. The absolute reductions in recurrence outstrip the absolute increases in major bleeding in this elderly, real‑world cohort, consistent with the principle that in many patients at moderate to high recurrence risk, extended anticoagulation is advantageous.
Strengths of the study
– Large, nationally representative claims databases provided substantial statistical power to detect differences in both thrombotic and bleeding endpoints and to examine long durations of therapy.
– Target trial emulation methods and propensity score matching were used to reduce confounding and to mimic randomized comparisons where feasible.
– Stratified analyses by initial treatment duration and OAC type helped assess the durability of benefit across clinically relevant subgroups.
Limitations and caveats
– Residual confounding: Claims data lack granular clinical detail (e.g., body mass index, D‑dimer, extent of proximal DVT or residual thrombus, frailty measures) and reasons for discontinuation; propensity matching cannot eliminate unmeasured confounding.
– Misclassification: Defining continued versus discontinued therapy by prescription fills can misclassify adherence or drug exposure; inpatient administration or sample medications are not captured.
– Outcome ascertainment: Recurrent VTE and major bleeding were identified by diagnostic codes associated with hospitalization; outpatient recurrences or bleeding events not requiring admission may be missed. Cause‑specific mortality was not available.
– Generalizability: Although the databases are large and complementary, Medicare patients were ≥65 years whereas Optum CDM included adults ≥18 years; the matched cohort had mean age ≈74 years, so findings may be most applicable to older populations.
– Observational nature: Despite emulation techniques, this is not a randomized trial; the magnitude of benefit reported may differ from that observed in randomized settings where enrollment criteria are more restrictive.
How this fits with existing evidence and guidelines
Randomized trials of extended anticoagulation (placebo‑controlled extension trials of DOACs and prior trials of warfarin) have demonstrated that prolonged therapy reduces recurrent VTE, often with variable impacts on major bleeding depending on agent and dose. Contemporary guidelines (for example, the CHEST guideline and the 2019 ESC guideline on acute pulmonary embolism) recommend considering extended or indefinite anticoagulation for patients with unprovoked VTE who have low or moderate bleeding risk, with individualized discussion of risks and benefits. The present study extends those trial findings into routine practice and supports guideline recommendations by demonstrating that, in a large real‑world elderly cohort, the balance favored ongoing anticoagulation.
Practice implications and decision making
Key clinical takeaways:
- For patients with unprovoked VTE who have completed three months of therapy, continued OAC markedly reduces recurrence risk and may reduce mortality, but clinicians must weigh this against an increased bleeding risk.
- Shared decision‑making is essential. Discuss absolute risk reductions and increases (e.g., ~25 fewer recurrent VTE vs ~5 more major bleeds per 1000 patient‑years in this cohort) and consider patient values regarding recurrence prevention vs bleeding risk.
- Assess bleeding risk using validated clinical features (prior bleeding, renal failure, uncontrolled hypertension, concomitant antiplatelet therapy, hepatic disease, thrombocytopenia) and consider modifiable factors before stopping or continuing therapy.
- Agent selection and dosing matter: lower‑dose DOAC regimens used in some trials for extended prophylaxis have shown favorable efficacy–safety tradeoffs; the study found benefit across OAC types, but individualized drug choice remains important.
Research gaps and future directions
Important unresolved questions include how to personalize duration more precisely using biomarkers (eg, D‑dimer after stopping), imaging (residual venous thrombus), and validated bleeding‑risk tools tailored to VTE populations. Randomized pragmatic trials of duration and dose de‑escalation strategies in real‑world older populations would further clarify benefit–harm tradeoffs. Finally, studies that capture patient‑reported outcomes and quality of life will help align anticoagulation decisions with patient preferences.
Conclusion
This large target‑trial emulation using US claims data indicates that, among patients with unprovoked VTE who completed at least 90 days of anticoagulation, continued OAC therapy substantially lowers recurrent VTE and mortality, at the cost of increased major bleeding, and yields an overall net clinical benefit. These real‑world results reinforce guideline‑based, individualized consideration of extended anticoagulation for patients with unprovoked VTE, particularly those at low to moderate bleeding risk.
Funding and clinicaltrials.gov
For funding sources and trial registration details, see the original publication: Lin KJ et al. Continued versus discontinued oral anticoagulant treatment for unprovoked venous thromboembolism: target trial emulation. BMJ. 2025;391:e084380. This analysis used administrative claims databases and was not a registered interventional trial.
References
1. Lin KJ, Kim DH, Singer DE, Zhang Y, Cervone A, Kehoe AR, Bykov K. Continued versus discontinued oral anticoagulant treatment for unprovoked venous thromboembolism: target trial emulation. BMJ. 2025 Nov 12;391:e084380. doi: 10.1136/bmj-2025-084380.
2. Kearon C, Akl EA, Ornelas J, et al. Antithrombotic Therapy for VTE Disease: CHEST Guideline and Expert Panel Report. Chest. 2016 Feb;149(2):315–352. doi:10.1016/j.chest.2015.11.026.
3. Konstantinides SV, Meyer G, Becattini C, et al. 2019 ESC Guidelines for the diagnosis and management of acute pulmonary embolism developed in collaboration with the European Respiratory Society (ERS). Eur Heart J. 2020;41(4):543–603. doi:10.1093/eurheartj/ehz405.
