Bisphosphonates in FAERS: 191,890 adverse-event reports by the numbers

The FDA Adverse Event Reporting System (FAERS) holds 191,890 individual safety reports that name at least one bisphosphonate — alendronate (Fosamax), zoledronic acid (Reclast, Zometa), risedronate (Actonel, Atelvia), ibandronate (Boniva), pamidronate (Aredia), or an older agent such as etidronate or clodronate — in any role, from the first structured records in the mid-1980s through the openFDA extract dated June 8, 2026. Alendronate anchors the class with 117,295 reports, reflecting two decades as the first-line generic oral bisphosphonate; zoledronic acid follows at 41,411 and risedronate at 25,737. The class is heavily serious-skewed: 158,070 of the reports — 82.4 percent — carry an FDA seriousness flag, and 22,138 (11.5 percent) record a death flag or a fatal outcome, a proportion driven by an elderly osteoporosis population, an oncology bone-metastasis subset, and the surgical burden of fractures. The most-coded reaction across the class is arthralgia (16,996), the severe musculoskeletal-pain signal the FDA flagged in 2008; the two class-signature toxicities — femur fracture (11,293 reports, including 1,724 explicitly coded atypical femur fracture) and osteonecrosis of the jaw (6,685) — sit clearly inside the list.

This article is a class-level descriptive read of bisphosphonate reporting in FAERS, written for pharmacovigilance, medical-affairs, bone/oncology market-access, and payer teams who meet these numbers in formulary reviews, prior-authorization and reauthorization criteria, drug-holiday policy, and dental-clearance pathways. 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. For the class-wide FAERS methodology and the database aggregate picture, see the companion analysis of 20 million FAERS reports and the FAERS report-reading workflow; for adjacent class cuts, see checkpoint inhibitors, JAK inhibitors, and TNF inhibitors in FAERS.

Methodology, in one paragraph

A report is counted for the class when a bisphosphonate name appears anywhere in that report's drug list — as a suspect, concomitant, or interacting drug — using a substring match across the drug_substances, drug_generics, and drug_brands fields (so "alendronic acid" and "alendronate sodium" both count). Because a single report can name several drugs, the per-molecule totals sum to more than the 191,890 class-level (union, deduplicated) reports. A "serious" report is any report the FDA coded as serious. A report counts toward death when the seriousness field carries a death flag or the outcome field carries a Fatal outcome. Reactions are coded to MedDRA Preferred Terms and case-normalized. FAERS does not link a reaction to a specific drug within a multi-drug report, so the reaction counts below are report-level for reports that name a bisphosphonate — they are not proof that the bisphosphonate caused the event, and, as the reaction table shows, they include the large rheumatoid-arthritis and autoimmune population that takes a bisphosphonate as glucocorticoid-induced-osteoporosis prophylaxis while the report is really about a concomitant biologic. This "any role" convention is intentionally inclusive and matches the other class cuts in this series; it will exceed suspect-only counts cited elsewhere.

Reporting volume, by molecule

Alendronate's dominance is a market-share and duration effect, not a risk differential: it has been the default generic oral bisphosphonate for postmenopausal and glucocorticoid-induced osteoporosis for two decades, so it accumulates the largest reporting base. Zoledronic acid's 41,411 reports span two very different populations — once-yearly Reclast for osteoporosis and monthly high-dose Zometa for hypercalcemia of malignancy and bone metastases — which is why its outcome mix runs more severe. Risedronate rounds out the oral generics. The union count of 191,890 counts each report once no matter how many bisphosphonates it names.

Seriousness, mortality, and who these patients are

The 82.4 percent serious share and 11.5 percent death-flag rate are population effects, not a bisphosphonate lethality signal. The reporting base is elderly (median 63, and far older in the oncology and fracture subsets), overwhelmingly female (75.6 percent, matching postmenopausal-osteoporosis epidemiology), and heavily weighted toward events — hip and femoral-shaft fractures, jaw surgery, and advanced malignancy — that are serious by definition and often end in hospitalization or death regardless of drug attribution. The report-level outcome field bears this out: 14,445 reports (7.5 percent) carry an explicit Fatal outcome, while 37,344 (19.5 percent) are "not recovered" and 33,188 (17.3 percent) "recovered," against a 37.7 percent "unknown outcome" share that is typical of FAERS follow-up gaps.

What the reaction list actually shows — and the any-role caveat

Because bisphosphonates are so often background therapy, the raw reaction ranking is a mix of the true class signal and the disease-and-drug profile of the patients who happen to be taking them:

This table is the honest version of a claim the older draft of this analysis got wrong. You cannot "filter to the bisphosphonate as suspect drug" in these data and recover a clean skeletal-only signal — FAERS stores drug roles as a per-report set, not aligned to a specific drug, so no reproducible per-drug suspect cohort exists. What you can say is reproducible and more useful: the any-role reaction list carries two overlapping populations. One is the true bone signal — arthralgia (16,996), femur fracture (11,293), osteonecrosis of jaw (6,685), bone pain (3,593), myalgia (5,122). The other is a large rheumatoid-arthritis and autoimmune cohort (arthropathy 8,742, alopecia 11,280, anti-CCP positive 5,464) who take a bisphosphonate as glucocorticoid-induced-osteoporosis prophylaxis while their report is really about a concomitant biologic or methotrexate. The indication field confirms it: rheumatoid arthritis is the second-largest recorded indication (30,809) after osteoporosis (46,133). Read the bone terms as the class signal; read the autoimmune terms as concomitant-therapy noise.

Osteonecrosis of the jaw: 6,685 reports

Osteonecrosis of the jaw (ONJ) — exposed maxillofacial bone that fails to heal within eight weeks in an antiresorptive-exposed patient without prior local radiation — appears in 6,685 reports that name a bisphosphonate. The pathophysiology combines profound suppression of bone remodeling, local microvascular compromise, and secondary infection, usually triggered by invasive dental work such as extractions or implants. Risk is dose- and potency-dependent: it is rare with low-dose oral alendronate for osteoporosis and materially higher with monthly high-dose intravenous zoledronic acid in oncology, where cumulative exposure can be roughly an order of magnitude greater.

The American Association of Oral and Maxillofacial Surgeons (AAOMS) stages medication-related ONJ to guide management: Stage 0 (non-specific symptoms, no exposed bone), Stage 1 (asymptomatic exposed/necrotic bone without infection), Stage 2 (exposed bone with infection and pain), and Stage 3 (exposed bone extending beyond the alveolus, with pathologic fracture or extraoral fistula). Stages 0–1 are managed conservatively with antimicrobial rinses and monitoring; Stage 2 adds oral antibiotics and limited debridement; Stage 3 requires systemic antibiotics and surgical resection. The consensus prevention rule that shapes access pathways is concrete: complete needed invasive dental work before starting high-dose antiresorptive therapy, and favor non-invasive dentistry once on treatment.

Atypical femur fracture: 11,293 femur-fracture reports, 1,724 coded atypical

Atypical femur fractures (AFF) — subtrochanteric or diaphyseal fractures of the femoral shaft after minimal or no trauma, linked to prolonged suppression of bone remodeling — surface as 11,293 femur-fracture reports, of which 1,724 are explicitly coded "atypical femur fracture." The mechanism is accumulation of unrepaired microdamage under long-term osteoclast inhibition, and absolute risk rises sharply after roughly five years of continuous use.

The American Society for Bone and Mineral Research (ASBMR) case definition is precise, and it is worth stating correctly because payer and clinical criteria cite it. Location is a prerequisite, not a feature: the fracture must lie along the femur from just distal to the lesser trochanter to just proximal to the supracondylar flare, and femoral-neck, classic intertrochanteric, periprosthetic, and pathologic (tumor-associated) fractures are specifically excluded. Within that region, at least four of five major features must be present:

1. Minimal or no trauma (fall from standing height or less).
2. A fracture line that originates at the lateral cortex and is substantially transverse.
3. Complete fractures extend through both cortices, often with a medial spike; incomplete fractures involve only the lateral cortex.
4. The fracture is non-comminuted or minimally comminuted.
5. Localized periosteal or endosteal thickening of the lateral cortex ("beaking" or flaring) at the fracture site.

Minor features — not required for the definition — include generalized cortical thickening of the diaphysis, prodromal dull or aching thigh/groin pain, bilateral fractures or symptoms, delayed healing, and comorbid conditions or drugs (glucocorticoids, proton-pump inhibitors, vitamin D deficiency). Prodromal thigh pain is therefore a supporting minor sign, not a defining major criterion — a common misstatement. When an AFF is diagnosed, bisphosphonate therapy is stopped; complete fractures are managed with intramedullary nailing, and patients are often transitioned to an anabolic agent such as teriparatide to promote cortical healing.

Oncology versus osteoporosis: two populations, one class

The indication field splits the class into its two clinical worlds. Osteoporosis (46,133 reports) and osteopenia (6,961) define the low-dose oral world — older, ~4:1 female, alendronate- and risedronate-treated, where AFF is the dominant long-duration concern and ONJ is rare. The oncology world — plasma cell myeloma (5,827), breast cancer (4,409), and bone metastases (3,541) — is high-dose intravenous zoledronic acid and pamidronate territory, where ONJ risk is materially higher because cumulative dosing, concurrent chemotherapy, anti-angiogenics, and poor dental status stack, but where AFF is less common because survival rarely spans the years of exposure that AFF requires. Rheumatoid arthritis (30,809) and psoriatic arthropathy (6,862) mark the third population — GIOP prophylaxis — whose adverse events are mostly about their biologics, as the reaction table shows.

Regulatory history, stated correctly

• October 2010: After a March 2010 review announcement and a September 2010 ASBMR task-force report, the FDA issued a Drug Safety Communication and required a class-wide labeling change adding atypical subtrochanteric and diaphyseal femur fractures to the Warnings and Precautions of all bisphosphonates approved for osteoporosis, plus a patient Medication Guide. Osteonecrosis of the jaw had already been added to bisphosphonate labeling over the preceding years (intravenous agents first, then the class).
• 2011–2012: The FDA reviewed the long-term efficacy of bisphosphonates and publicly questioned optimal treatment duration beyond 3–5 years, seeding the "drug holiday" concept.
• 2016: The ASBMR published its task-force report on managing osteoporosis in patients on long-term bisphosphonate therapy, which formalized drug-holiday guidance — reassess after 5 years of oral (or 3 years of intravenous) treatment in lower-risk patients, and continue in high-risk patients.

There was no separate, distinct "2015 class label update"; the durable label events are the 2010 atypical-fracture class labeling and the earlier ONJ additions, with drug-holiday practice anchored to the 2016 ASBMR report. Payers translated this into reauthorization criteria that require documented continued fracture risk (low T-score, prior fragility fracture, or high FRAX) to justify treatment beyond the holiday window.

The oral GI-tolerability signal and administration rules

Beyond the headline skeletal toxicities, the oral bisphosphonates carry a distinct upper-gastrointestinal signal that shapes real-world adherence: abdominal discomfort (11,138 reports), abdominal pain upper (5,226), and the smaller but mechanistically specific oesophagitis (659) and oesophageal ulcer (264) terms. Oral bisphosphonates are directly irritating to the oesophageal mucosa if they linger, which is why the labels carry strict administration rules — take on an empty stomach with a full glass of plain water, remain upright for at least 30 minutes (60 for ibandronate), and avoid other food, drink, or medication during that window — and why oesophageal stricture, achalasia, or an inability to sit upright are contraindications. For access and hub teams, this is the reproducible case for the intravenous agents (zoledronic acid, ibandronate IV) in patients who cannot tolerate or reliably follow oral dosing, and for counseling touchpoints at the specialty-pharmacy and refill level: much of the oral GI signal is administration-technique-driven and preventable, not an intrinsic drug failure.

Bisphosphonate versus denosumab

Denosumab (Prolia, Xgeva) is the RANKL-inhibitor antiresorptive that competes with bisphosphonates across the same indications, but with a very different discontinuation profile:

The access consequence is the rebound-fracture risk unique to denosumab: a prior-authorization lapse or specialty-pharmacy delivery gap can trigger multiple vertebral fractures within 6–12 months, which is why bisphosphonates — with their long skeletal reservoir and no rebound — remain the safer default for step-therapy. For the biologic side of this market, see the denosumab biosimilars and access guide.

The reporting trajectory, by year

Reporting rose through the 2010–2012 window around the atypical-fracture communication and the peak of alendronate use, then settled into a stable band of roughly 8,000–11,000 reports per year — the steady state of a mature generic class. The 2012 spike (15,294) coincides with the post-2010-labeling reporting surge. (Years omitted from the table fall within the same band; 2026 is a partial year through the extract cut-off.)

What this means for pharmacovigilance, medical-affairs, and access teams

• Read the reaction list in two layers. The bone terms (arthralgia, femur fracture, osteonecrosis of jaw, bone pain) are the class signal; the autoimmune terms (alopecia, arthropathy, anti-CCP positive) are the rheumatoid-arthritis GIOP-prophylaxis population reporting on concomitant biologics. Do not quote alopecia or anti-CCP as bisphosphonate effects.
• Build reauthorization around the drug holiday, not a fixed number. The reproducible AFF duration signal supports reassessment at 3–5 years, with continuation gated on documented fracture risk (T-score, prior fracture, FRAX) rather than an arbitrary stop.
• Gate high-dose IV zoledronic acid on dental clearance. The ONJ count (6,685) concentrates in the oncology high-dose population; a completed pre-infusion dental exam is a defensible medical-necessity step.
• Keep the generic oral bisphosphonate as first step. With no rebound-fracture risk and the lowest acquisition cost, alendronate remains the appropriate step-therapy anchor before denosumab or anabolic agents.
• State the ASBMR criteria correctly in policy language. Location is a prerequisite and prodromal pain is a minor feature; criteria that misclassify these invite appeals.

FAQ

Does FAERS prove bisphosphonates cause osteonecrosis of the jaw or atypical femur fractures?

No. FAERS is a spontaneous, denominator-free reporting system, and a report is not proof of causation — concomitant steroids, dental disease, malignancy, or other drugs may contribute. But the reproducible presence of osteonecrosis of the jaw (6,685) and femur fracture (11,293, including 1,724 coded atypical) in reports naming a bisphosphonate is consistent with the mechanism of prolonged osteoclast inhibition and with the FDA's class warnings.

Why are three-quarters of bisphosphonate reports in women?

The class is 75.6 percent female (144,998 of 191,890) because postmenopausal osteoporosis is the dominant indication. Estrogen loss accelerates bone resorption, making older women the largest treated — and reported — population.

Is zoledronic acid or alendronate more associated with ONJ?

In absolute report volume alendronate leads (117,295 vs 41,411), because it is used far more widely. But ONJ risk per patient is higher with monthly high-dose intravenous zoledronic acid in oncology, where cumulative dosing can be roughly ten times the osteoporosis exposure. Absolute FAERS counts track utilization, not per-patient risk.

How should access teams handle the autoimmune "noise" in these reports?

Recognize that any-role bisphosphonate reports include a large rheumatoid-arthritis population on GIOP prophylaxis whose events are driven by concomitant biologics. Weight the bone-specific terms (femur fracture, osteonecrosis of jaw, arthralgia, bone pain) when reasoning about the bisphosphonate itself, and treat alopecia, arthropathy, and anti-CCP positivity as concomitant-therapy markers.

Sources

• US FDA, openFDA FAERS (Adverse Event Reporting System) extract, export dated 2026-06-08 (20,328,575 total reports). Class-union, per-molecule, outcome, reaction, indication, and trajectory aggregates computed by PharmaDossier from the public FAERS extract using an any-role substring match across drug substance, generic, and brand fields. https://open.fda.gov/data/faers/
• US FDA, Drug Safety Communication: "Safety update for osteoporosis drugs, bisphosphonates, and atypical fractures" (October 2010) — class-wide Warnings and Precautions labeling change and Medication Guide for atypical subtrochanteric and diaphyseal femur fractures. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-safety-update-osteoporosis-drugs-bisphosphonates-and-atypical
• Shane E, Burr D, Abrahamsen B, et al. "Atypical Subtrochanteric and Diaphyseal Femoral Fractures: Second Report of a Task Force of the American Society for Bone and Mineral Research." J Bone Miner Res 2014;29(1):1-23 — case definition (location prerequisite, four of five major features; prodromal pain a minor feature). https://pubmed.ncbi.nlm.nih.gov/23712442/
• Adler RA, El-Hajj Fuleihan G, Bauer DC, et al. "Managing Osteoporosis in Patients on Long-Term Bisphosphonate Treatment: Report of a Task Force of the American Society for Bone and Mineral Research." J Bone Miner Res 2016;31(1):16-35 — drug-holiday framework. https://pubmed.ncbi.nlm.nih.gov/26350171/
• Ruggiero SL, Dodson TB, Aghaloo T, et al. AAOMS Position Paper on Medication-Related Osteonecrosis of the Jaw — staging and management. https://www.aaoms.org/practice-resources/aaoms-position-papers
• US FDA, postmarket drug safety information for bisphosphonates (Fosamax, Actonel, Boniva, Reclast, Zometa, Aredia). https://www.fda.gov/drugs/postmarket-drug-safety-information-patients-and-providers/information-bisphosphonates-market-name-fosamax-fosamax-plus-d-actonel-actonel-with