Analysis · Health
When Dementia Drug Safety Reports Reveal a Signal, Not a Risk Rate
By AWEI · AI-compiled · Published · Analysis prepared · 1 source · doi.org
Among 5,918 US adverse event reports involving rivastigmine, 995 listed death. That striking count does not tell us how often people taking the dementia drug died: the database contained reports of suspected harm, not a count of everyone treated. This distinction is central to interpreting a 2015 study by Thibault B. Ali and colleagues.
The researchers retrospectively analyzed 9,877 reports from the US Food and Drug Administration database and 2,247 from Canada Vigilance. They compared three cholinesterase inhibitors used for dementia: rivastigmine, donepezil and galantamine. This was an analysis of existing safety reports, not an experiment assigning patients to treatments.
To make the databases more comparable, the team combined several US categories of fatal outcomes into one death category. They then calculated reporting odds ratios to identify events appearing disproportionately often in reports involving particular drugs. Rivastigmine showed a higher relative representation of death reports in both countries.
Agreement across two databases made the signal worth investigating, but did not establish that rivastigmine caused the deaths. Reporting was voluntary, adverse events could go unreported, and the researchers lacked prescription totals. Consequently, neither the counts nor the reporting odds ratios measured an individual user's probability of dying.
The authors also raised possible explanations that their analysis could not resolve. Rivastigmine might have been prescribed more often to people with advanced Alzheimer's disease. Errors involving its skin patches were another hypothesis, supported by previously published reports but not demonstrated as the explanation for this database pattern.
The practical value is a focused research question: what explains the excess representation of fatal outcomes in rivastigmine reports? Safety databases can help identify priorities for investigation. Here, they could neither establish a causal drug effect nor determine which treatment was safest for a particular patient.
Sources used for this article (1)
Direct links to the publisher reports used to prepare this article.
- Source 1
- Adverse Effects of Cholinesterase Inhibitors in Dementia, According to the Pharmacovigilance Databases of the United-States and Canada Thibault B. Ali, Thomas R. Schleret, Brian M. Reilly …
