The conventional wisdom about drug safety is not just outdated—it is economically dangerous. Three persistent myths continue to shape how pharmaceutical companies, biotechs, and even some regulators approach pharmacovigilance (PV), and each one carries a multi-million-dollar price tag.
The myths—that silence signals safety, that a single database suffices for signal detection, and that PV is merely a cost center—persist not because they are true, but because they are comforting. This article dismantles each one with data, regulatory evidence, and case studies from across the globe.
Myth 1: “No News Means a Good Safety Profile”
This is the most seductive myth in pharmacovigilance: the belief that a low volume of adverse event reports indicates a safe product. In reality, it often signals a broken reporting system.
The Underreporting Reality
Adverse drug reaction (ADR) underreporting is not a marginal problem—it is the dominant reality. Global historical data suggest that only 2–4% of non-serious cases and approximately 10% of serious cases are ever reported by healthcare professionals. A study in general practice estimated that GPs might report only 1 out of every 24,433 adverse drug reactions they observe. In Australia, a recent study found that only 36% of patients who experienced an ADR actually reported it, despite 58% knowing they could.
The WHO has established a benchmark: a healthy national pharmacovigilance system should receive at least 200 reports of medication-related harm per million inhabitants annually. Many countries still fall far short of this goal.
Regional Disparities in Reporting
The problem is globally distributed but regionally concentrated. A 2025 joinpoint regression analysis across eight countries found that healthcare professional-to-consumer reporting ratios were significantly decreasing in most nations studied, with average annual percent changes ranging from −43.7 to −24.9—meaning the burden of reporting is increasingly falling on patients rather than trained clinicians.
VigiBase, the WHO global database of individual case safety reports, reveals stark regional disparities: the United Kingdom alone accounted for over 90% of European clozapine-related fatal outcomes reported to VigiBase, a finding that almost certainly reflects reporting practices rather than actual clinical reality.
What “No News” Actually Means
When companies mistake underreporting for safety, three dangerous consequences follow.
First, genuine safety signals remain buried because the raw numbers never trigger statistical thresholds—this is especially acute for orphan drugs, where the combination of underreporting and low patient exposure makes false negatives highly probable.
Second, companies become complacent, reducing PV investment precisely when it is most needed, as postmarketing trials funded by industry have been shown to be ineffective at identifying safety issues.
Third, the absence of robust safety data creates regulatory vulnerability: when a signal finally emerges, the company lacks the longitudinal data to contextualize it.
The Remediation Framework
Companies serious about addressing underreporting should implement patient-centric reporting channels, including mobile applications and QR-code-based systems directly linked to national regulatory databases; deploy targeted HCP education programs with real-time feedback mechanisms to close the awareness gap (studies show only 12–60.8% of HCPs have ever reported an ADR); and adopt active surveillance methodologies—such as sentinel site monitoring and electronic health record mining—rather than relying solely on passive spontaneous reporting, aligning with the FDA’s Sentinel Initiative model.
Myth 2: “Our Database Is Enough for Signal Detection”
The second myth reflects a technological overconfidence: the belief that a single database—whether FAERS, EudraVigilance, or an internal safety database—is sufficient for robust signal detection.
EMA’s Position on Signal Detection Limitations
The European Medicines Agency has been explicit on this point. The EMA’s Guideline on Good Pharmacovigilance Practices (GVP) Module IX Addendum I details the methodological aspects of signal detection from spontaneous reports and emphasizes that statistical summaries alone are insufficient. The EMA’s Pharmacovigilance Risk Assessment Committee (PRAC) is responsible for prioritizing and assessing signals, and its work is built on the premise that no single data source provides a complete picture.
A critical finding published in the evaluation of historical safety signals from the EMA showed that none of the positive controls could be detected in the THIN database at an early stage, whereas detection was possible in VigiBase for some signals—underscoring that different databases capture different safety information at different times.
The Multi-Database Imperative
Single-source database dependency fundamentally limits signal context. An association that appears weak in isolation often becomes statistically meaningful when evaluated across FAERS, EudraVigilance, and VigiBase simultaneously. Furthermore, performing signal detection within therapeutic area strata—rather than across the full database—limits inappropriate cross-area comparisons that can generate spurious signals. The EMA has also acknowledged that quantitative methods cannot detect all signals, and the concept of a statistical threshold implies that not all reports will be reviewed.
The practical implication is that companies relying on a single database are operating with a systematically incomplete view of their product’s safety profile. The EMA’s GVP Module IX requires marketing authorization holders to monitor multiple data sources, including the scientific literature, and to integrate findings across databases.
Why One Database Fails
The limitations of single-database approaches stem from four structural factors. Population coverage varies by region, disease prevalence, and prescribing patterns. Reporting culture differs dramatically: the UK’s Medicines and Healthcare products Regulatory Agency (MHRA) Yellow Card scheme has historically generated far higher reporting rates than many continental European systems. Coding practices diverge: MedDRA coding conventions and versioning affect how identical clinical events are captured. And temporal sensitivity differs: an emerging safety signal may manifest in EudraVigilance months before appearing in FAERS, or vice versa.
An Integrated Signal Detection Architecture
A robust signal detection system should integrate three layers: quantitative analysis using disproportionality methods (PRR, ROR, IC) across multiple databases simultaneously; qualitative review by safety physicians who assess clinical plausibility, temporality, and biological gradient; and contextual data from the published literature, regulatory intelligence, and real-world evidence sources.
Companies should establish a signal detection governance board with defined escalation pathways to ensure that weak signals are tracked and re-evaluated as new data accumulates, rather than being dismissed prematurely.
Myth 3: “Pharmacovigilance Is Just a Regulatory Cost Center”
The third myth is perhaps the most commercially damaging. It frames PV purely as a compliance overhead—a necessary evil to satisfy regulators rather than a strategic asset. The evidence tells a radically different story.
The True Cost of Pharmacovigilance Failure
The financial consequences of PV failure are measured in billions, not millions. When Merck withdrew Vioxx (rofecoxib) from the market in 2004, the drug had been generating more than two and a half billion dollars in annual revenue, representing approximately ten percent of Merck’s worldwide sales. Merck ultimately agreed to pay four point eight five billion dollars to settle approximately twenty seven thousand lawsuits, with some estimates projecting total legal exposure as high as fifty billion dollars. The company also paid a nine hundred and fifty million dollar criminal fine.
GlaxoSmithKline’s experience with paroxetine (Paxil/Seroxat) tells a similar story. GSK paid approximately one billion dollars to settle lawsuits related to the antidepressant, including roughly two hundred million dollars for addiction and birth defect claims and another four hundred million dollars for antitrust, fraud, and design claims. In a separate action, the company recorded a legal charge of one point five seven billion pounds (two point three six billion dollars) in a single quarter to resolve litigation. Even when GSK won a legal case over paroxetine withdrawal effects, the thirteen-year litigation still cost nine point three three million pounds in legal fees alone.
More recently, GSK agreed to pay two point two billion dollars to settle approximately eighty thousand lawsuits related to allegations that Zantac (ranitidine) caused cancer, resolving ninety three percent of the state court litigation it faced in the United States.
WHO Data: The Macroeconomic Scale of Medication Harm
The World Health Organization frames the issue in macroeconomic terms. Unsafe medication practices and medication errors cost an estimated US$42 billion annually worldwide. In low- and middle-income countries, unsafe care contributes to an estimated 134 million adverse events annually, resulting in around 2.6 million deaths.
The WHO launched its third Global Patient Safety Challenge, “Medication Without Harm,” in 2017 with the explicit goal of reducing severe, avoidable medication-related harm by 50% globally within five years. The very existence of this initiative underscores that the status quo—in which PV is treated as a compliance checkbox—is failing patients and health systems alike.
The Strategic Value of Pharmacovigilance
The industry conversation is shifting. As ArisGlobal articulated in a 2023 analysis, the key to transforming safety “from a reactive cost center to a proactive profit driver” lies in providing safety teams with proper tools and resources to focus on actionable insights. EVERSANA’s 2025 industry commentary noted that PV leaders are being pulled in two directions: the relentless demands of regulatory compliance on one side, and the opportunity to transform aggregate reports into strategic safety intelligence on the other.
The strategic value of PV manifests in at least six concrete domains: labeling optimization, where robust safety data support favorable label language and differentiated risk-benefit positioning; lifecycle extension, where proactive signal management identifies new indications or patient subpopulations; payer negotiation, where comprehensive safety evidence strengthens reimbursement submissions; due diligence, where PV capability assessment becomes a critical factor in M&A valuation; brand protection, where early signal detection prevents the reputational damage that accompanies high-profile safety controversies; and regulatory efficiency, where inspection-ready PV systems reduce the cost and disruption of regulatory actions.
A 2025 Everest Group analysis confirmed that PV operations have become a strategic priority for life sciences enterprises facing heightened regulatory scrutiny, evolving safety standards, and growing complexity across global markets. The WHO’s global smart pharmacovigilance strategy, published in 2025, envisions PV systems that are “strong, adaptive, and trusted”—not merely compliant.
From Cost Center to Value Driver: An Implementation Roadmap
The transformation from reactive compliance to strategic asset requires concrete operational changes. Companies should invest in advanced analytics capabilities, including natural language processing for literature screening and machine learning for signal prioritization, to free human experts for high-value interpretative work. They should integrate PV with real-world evidence (RWE) teams to enable continuous benefit-risk evaluation across the product lifecycle.
They should establish PV representation in cross-functional governance bodies, ensuring safety insights inform commercial, regulatory, and clinical development decisions. And they should measure and communicate PV’s value through metrics such as signals detected before regulatory identification, adverse event prevention rates, and avoided litigation exposure—not merely case processing volumes.
Conclusion: The Price of Believing Myths
The three myths of pharmacovigilance—that silence means safety, that one database suffices, and that PV is only a cost—are not merely intellectual errors. They are operational failures with quantifiable financial consequences.
The Vioxx withdrawal destroyed approximately twenty-five to fifty billion dollars in market capitalization and settlement costs. The World Health Organization estimates that medication errors cost the global economy forty-two billion dollars annually. These figures dwarf the incremental investment required to build robust, multi-source, strategically integrated pharmacovigilance systems.
For pharmacovigilance professionals, the opportunity lies in reframing the conversation: not “How much does PV cost?” but “How much does PV failure cost—and how much value does PV excellence create?” The data is clear.
The question is whether the industry will act on it. Companies that continue to treat PV as a compliance checkbox rather than a strategic imperative are not saving money—they are borrowing against a future liability that compounds with every unreported adverse event, every undetected signal, and every missed opportunity to generate safety intelligence that could shape a product’s commercial trajectory.
