DOACs in FAERS: 330,426 adverse-event reports by the numbers

The FDA Adverse Event Reporting System (FAERS) holds 330,426 individual safety reports that name at least one of the four marketed direct oral anticoagulants (DOACs) — rivaroxaban (Xarelto), dabigatran (Pradaxa), apixaban (Eliquis), or edoxaban (Savaysa) — in any role, filed between 2010 and early 2026. Rivaroxaban alone accounts for 185,369 of them — 56% of the class — followed by dabigatran (76,925) and apixaban (64,288). Eighty-five percent of every class-level report — 280,614, or 84.9% — carries an FDA seriousness flag, the highest serious-case share of any class cut we have analyzed, and 43,243 reports (13.1%) record a fatal outcome. The reaction list is a hemorrhage registry: gastrointestinal haemorrhage is the single most coded term at 29,890 reports, ahead of cerebrovascular accident (12,789) and general haemorrhage (11,969).

This article is a class-level descriptive read of DOAC reporting in FAERS, written for pharmacovigilance, medical-affairs, pharmacy-and-therapeutics, and market-access teams who encounter DOAC bleeding numbers in formulary positioning, prior-authorization rationale, and reversal-agent purchasing decisions. Every figure here is computed from the public openFDA FAERS extract (20,328,575 total reports, export dated 2026-06-08). It is not a disproportionality signal analysis, it is not a safety claim, and it is not incidence data. FAERS is a spontaneous-reporting system: it captures what was reported, never what was caused, and it has no exposure denominator. Those caveats shape every number below, and the reporting-era distortions they produce — particularly the concentration of rivaroxaban reports around its 2011–2016 post-launch scrutiny window — are spelled out in the methodology and per-drug sections. For the class-wide FAERS methodology and the database aggregate picture, see the companion analysis of 20 million FAERS reports; for the DOAC access landscape, see the DOAC access landscape. This article stays on the four marketed oral anticoagulants.

Methodology, in one paragraph

A report is counted for a substance when the substance's generic name appears anywhere in that report's drug list — as a suspect, concomitant, or interacting drug — using a substring match so that "rivaroxaban," "apixaban," and their combination or salt variants all count. Because a single report can name several drugs, the per-substance totals sum to more than the 330,426 class-level (union) reports. A "serious" report is any report the FDA coded as serious. A report counts toward death when either the seriousness field carries a death flag or the outcome field carries a Fatal outcome. Reactions are coded to MedDRA Preferred Terms; the openFDA extract carries mixed casing for some terms, so reaction counts are case-normalized. The "any role" counts are intentionally inclusive and will exceed suspect-only or differently normalized counts cited elsewhere; betrixaban (Bevyxxa), approved 2017 and commercially withdrawn, is excluded because it carries no reports in this extract.

The reporting-volume picture, by drug

Two patterns matter for access teams. First, rivaroxaban's commanding report lead is a reporting-era artifact as much as a safety one. Rivaroxaban launched in late 2011 and became the highest-volume DOAC prescription of the early decade, but its FAERS reports peaked in 2015–2016 amid concentrated post-launch bleeding scrutiny and early litigation — a window that inflated spontaneous reporting beyond what pure exposure would predict. Published FAERS analyses of the 2014–2019 period similarly counted 53,085 hemorrhage reports for rivaroxaban against 14,100 for dabigatran and 13,151 for apixaban, the same ordering this longer extract reproduces.

Second, apixaban's report count (64,288) sits below both rivaroxaban and dabigatran despite apixaban being, by the mid-2020s, the most-prescribed DOAC by dispensing volume. That inversion is meaningful for formulary committees: across head-to-head observational studies and the FDA Sentinel system, apixaban has consistently shown the lowest gastrointestinal bleeding risk of the class — with Sentinel reporting rivaroxaban versus apixaban major-GI-bleed hazard ratios near 2.3 in patients 65 and older — so a lower FAERS reporting rate for apixaban is consistent with, not contradictory to, its real-world uptake. The 95.6% serious-case share for apixaban reports is a reporting-channel effect (apixaban reports skew heavily to hospitalized HCP-filed bleeding events), not an indication that apixaban is the most dangerous agent.

The bleeding footprint, by category

Bleeding is the class's entire safety story, and aggregating the dispersed MedDRA terms makes the scale visible:

The intracranial-bleeding category — cerebral haemorrhage (5,478), haemorrhage intracranial (3,680), subdural haematoma (2,926), and subarachnoid haemorrhage (1,513) — is the headline advantage DOACs hold over warfarin in the pivotal trials, and it appears here at meaningful but lower volume than gastrointestinal bleeding, consistent with the trial finding that intracranial hemorrhage is lower with all DOACs than with warfarin while gastrointestinal bleeding is not uniformly lower. The descriptive ordering here is corroborated by disproportionality work: a 2012–March 2024 FAERS analysis found rivaroxaban carried a significantly higher reporting odds ratio than apixaban for cerebral hemorrhage (ROR 1.73, 95% CI 1.60–1.86) and for gastrointestinal, rectal, and renal hemorrhage. The same ordering appears outside FAERS — the ARISTOPHANES study of more than 280,000 nonvalvular-atrial-fibrillation patients found apixaban and dabigatran had lower major-bleeding rates than warfarin with the inverse for rivaroxaban, and a 2026 head-to-head trial in acute VTE reported major bleeding in 0.4% of apixaban patients versus 2.4% of rivaroxaban patients. For P&T committees, the practical implication is that DOAC selection within the class is driven less by overall bleeding than by the gastrointestinal-versus-intracranial tradeoff, renal function, and patient age — which is exactly why apixaban's lower GI-bleed profile drives its formulary preference.

The reporting trajectory, by year

The class reporting curve rises sharply with the 2010 dabigatran launch and the 2011 rivaroxaban launch, peaks in 2016 at 35,508 reports, and then declines steadily even as DOAC prescribing continued to outpace warfarin through the 2020s. That decline is a reporting-dynamics signal, not a safety improvement: spontaneous FAERS reporting is heaviest in the early post-launch window when a new drug class attracts both heightened prescriber attention and active litigation-driven reporting, then normalizes downward. The decline should not be read as evidence of falling real-world bleeding.

The reversal-agent layer that shapes formulary and PA design

DOAC bleeding reports cannot be read in isolation from the reversal-agent infrastructure, because that infrastructure is what makes a life-threatening DOAC bleed manageable — and it is unevenly distributed across the class.

Idarucizumab, a monoclonal antibody fragment with roughly 350-fold higher affinity for dabigatran than thrombin itself, gave dabigatran the first specific antidote in the class in 2015. Andexanet alfa, a recombinant modified Factor Xa decoy that sequesters apixaban and rivaroxaban, followed in 2018 under accelerated approval based on anti-Factor Xa activity reduction and ANNEXA-4 hemostasis results — though with a noted thrombotic-rebound signal that has kept many hospitals using lower-cost 4-factor prothrombin complex concentrate off-label instead. Edoxaban has no approved specific reversal agent. The access consequence is concrete: institutional DOAC choice is shaped not only by bleeding incidence but by whether a hospital stocks the corresponding reversal agent, the cost and preparation time of andexanet alfa, and the thrombotic-risk tradeoff — all of which enter P&T committee and pharmacy-policy decisions alongside the FAERS bleeding footprint.

How to read these numbers (and how not to)

• Rivaroxaban's report volume is reporting-era inflated. The 185,369 figure reflects a 2011 launch, early-decade market leadership, and concentrated 2015–2016 post-launch scrutiny plus litigation-driven reporting. It is not evidence that rivaroxaban is the most dangerous DOAC at equal exposure.
• Apixaban's lower report count is consistent with its bleeding profile. Observational data and FDA Sentinel show apixaban has the lowest gastrointestinal bleeding risk in the class, so a lower FAERS reporting rate alongside the highest dispensing volume is the expected pattern — not an anomaly.
• The 84.9% serious-case share reflects a reporting channel, not incidence. DOAC reports skew heavily to hospitalized, HCP-filed bleeding events, so the serious-share denominator is biased upward; it cannot be read as the probability that a DOAC adverse event is serious.
• FAERS has no denominator. None of these numbers is an incidence rate, a relative risk, or a causal estimate. A 330,426-report class total says the surveillance system has watched this class closely for 15 years; it does not say the class caused 330,426 events.

Sources

• US FDA, FDA Adverse Event Reporting System (FAERS) / openFDA drug adverse event data, public extract (20,328,575 reports, export dated 2026-06-08). Aggregates computed by PharmaDossier from the openFDA FAERS full extract. https://open.fda.gov/data/faers/
• US FDA, accelerated approval of Andexxa (coagulation factor Xa (recombinant), inactivated-zhzo) for reversal of rivaroxaban and apixaban, May 3, 2018; idarucizumab (Praxbind) approval for dabigatran reversal, October 2015. https://www.fda.gov/drugs/drug-approvals-and-databases
• Pollack CV Jr, Reilly PA, van Ryn J, et al. "Idarucizumab for Dabigatran Reversal — Full Cohort Analysis." New England Journal of Medicine, 2017;377:431–441 (idarucizumab 350-fold binding affinity; dabigatran reversal). https://www.nejm.org/doi/full/10.1056/NEJMoa1707278
• Connolly SJ, Milling TJ Jr, Eikelboom JW, et al., for the ANNEXA-4 Investigators. "Andexanet Alfa for Acute Major Bleeding Associated with Factor Xa Inhibitors." New England Journal of Medicine, 2016;375:1131–1141 (andexanet alfa reversal of apixaban/rivaroxaban; hemostasis and thrombotic-event data). https://www.nejm.org/doi/full/10.1056/NEJMoa1607887
• Graham DJ, Reichman ME, Wernecke M, et al. "Cardiovascular, Bleeding, and Mortality Risks in Elderly Medicare Patients Treated with Dabigatran or Warfarin for Nonvalvular Atrial Fibrillation." Circulation, 2015;131:157–164 (FDA Sentinel DOAC bleeding methodology). https://www.ahajournals.org/doi/10.1161/CIRCULATIONAHA.114.012061
• Lau WCY, et al. "Apixaban vs Doacs and the Risk of Gastrointestinal Bleeding." Annals of Internal Medicine (apixaban 19–28% lower GI-bleed risk vs other DOACs across 527,226 new users; standard- and reduced-dose, elderly, and CKD subgroups). https://www.acpjournals.org
• US FDA Sentinel Initiative, modular program report on thromboembolic stroke, intracranial hemorrhage, gastrointestinal bleeding, and major extracranial bleeding following apixaban and rivaroxaban use in atrial fibrillation (rivaroxaban vs apixaban major GI-bleed HR ~2.3, age ≥65). https://www.sentinelinitiative.org
• "Apixaban- and Rivaroxaban-Associated Bleeding: A Retrospective Analysis Using the FDA Adverse Event Reporting System," 2024–2025 (FAERS Jan 2012–March 2024; rivaroxaban cerebral-hemorrhage ROR 1.73, 95% CI 1.60–1.86 vs apixaban; higher GI, rectal, renal hemorrhage, epistaxis, hemoptysis). https://pmc.ncbi.nlm.nih.gov/articles/PMC12232983
• Lip GYH, et al. ARISTOPHANES retrospective observational study of major bleeding across DOACs vs warfarin in nonvalvular atrial fibrillation (>280,000 patients; apixaban and dabigatran lower major bleeding vs warfarin, rivaroxaban higher GI bleeding). https://www.acc.org
• Castellucci LA, et al. "Bleeding Risk with Apixaban vs. Rivaroxaban in Acute Venous Thromboembolism," New England Journal of Medicine, 2026;394:1051–1060 (major bleeding 0.4% apixaban vs 2.4% rivaroxaban; composite bleeding 3.3% vs 7.1%). https://www.nejm.org
• US FDA, MedWatch adverse-event reporting program. https://www.fda.gov/safety/medwatch-fda-safety-information-and-adverse-event-reporting-program