A larger number of adverse-event reports for dextroamphetamine does not establish a higher probability of harm among people taking it. Wang and colleagues’ 2025 comparison with viloxazine illustrates the problem: spontaneous-report counts can help identify questions for investigation, but they lack the treated-population denominator needed to estimate incidence.
What was actually collected?
This was a retrospective descriptive pharmacovigilance study, not a randomized drug comparison. The researchers queried WHO-VigiAccess on August 31, 2024, and extracted FAERS data on November 4, 2024. Their reported VigiAccess totals were 959 reports for viloxazine and 4,435 for dextroamphetamine, totaling 5,394.
The paper framed the comparison around newer branded products, but acknowledged that the active ingredients’ reporting histories extended to 1976 and 1969, respectively. Those histories predate the modern viloxazine extended-release capsule and dextroamphetamine patch. The accumulated active-ingredient totals therefore cannot automatically be described as adverse effects caused by those recent formulations alone.
A report is not the same unit as a reaction term
The article’s tables used different denominators. Its demographic table used the 959 and 4,435 report totals. Its system-organ-class table used 1,633 and 10,810 entries, while the table of common preferred reaction terms used 2,199 and 15,572. These denominators should not be merged into a single patient count or treated as exposed treatment populations.
For example, a percentage within a reaction-term table describes the share of entries in that table. It does not tell the reader the percentage of all dextroamphetamine users who developed that reaction. Multiple reported problems can also relate to one report; adding terms does not create additional independent people.
What does disproportionality mean?
The study used reporting odds ratios and proportional reporting ratios to examine imbalances in reported events. These statistics compare patterns within a reporting database. They do not supply the risk ratio from following comparable groups of treated and untreated people.
An excess reporting signal can justify closer examination of case narratives, competing causes and independent clinical evidence. It does not by itself demonstrate that the drug caused each event or that one medicine is safer for an individual patient. Statistical signal thresholds cannot repair missing exposure information or unmeasured differences between reporting populations.
Even “drug ineffective” needs interpretation
Dextroamphetamine’s frequently reported terms included drug ineffectiveness, unexpected therapeutic response and product-substitution issues. Those are useful real-world concerns, but they are not all manifestations of toxic tissue injury. A database category that combines general disorders with administration-site conditions is also broader than a specific diagnosis.
The separate FAERS analysis covered the first quarter of 2004 through the third quarter of 2024. The authors reported 561 patients with 1,584 adverse events for viloxazine and 1,927 patients with 6,427 events for dextroamphetamine. These are that analysis’s reported units, not additional exposed cohorts to add to VigiAccess. Potential overlap between systems prevents assuming that their totals represent wholly independent people.
Which conclusions remain unsupported?
The study cannot establish incidence, a causal head-to-head safety ranking, or patch-specific risk from the cumulative active-ingredient data. Voluntary reporting, incomplete records, different marketing histories and reporting attention all limit inference. The authors acknowledged reporting bias and the absence of causality assessment.
The appropriate use is signal generation tied to a defined database and extraction date. For an individual adverse-event opinion, chronology, dose, route, formulation, concurrent exposures, medical history and alternative explanations would need case-specific examination. These aggregate tables do not provide those facts, and the absence of a report is not evidence that an event cannot occur.
Source
Wang L, Kong Q, Li Y. Real world pharmacovigilance comparison of viloxazine and dextroamphetamine adverse reaction profiles. Scientific Reports. 2025;15:28919. doi:10.1038/s41598-025-14385-2. Historical database snapshot, not current comparative prescribing guidance.
Leave a Reply