A claims-based study found an association between benzodiazepine exposure and epilepsy diagnosis. The authors also recognized reverse causality and missing confounders, which limit individual causal conclusions.
The observed association
Harnod and colleagues compared 1,065 adults newly diagnosed with epilepsy with 4,260 controls matched on age, sex, and index year. They used Taiwan insurance data to reconstruct prescribing exposure and adjusted for selected recorded comorbidities.
The adjusted odds ratio for the association with benzodiazepine exposure was 2.02, with a 95% confidence interval of 1.68–2.42. Higher exposure measures were also associated with higher odds. These are odds from a case-control analysis, not a randomized estimate of what taking a prescription would cause.
Alternative explanations remain
Symptoms preceding an epilepsy diagnosis may themselves prompt prescribing. The authors expressly discussed this reverse-causality problem. Records also lacked information on alcohol, smoking, family history, and other potentially important characteristics.
The diagnostic category combined different epilepsy types, and prescriptions could not be confirmed through direct participant contact. The discussion’s proposed withdrawal and receptor-adaptation mechanisms were interpretations, not mechanisms measured by this database study.
A disciplined use of the result
The paper can support further investigation of a medication and seizure chronology. It does not prove that prescribed benzodiazepines caused epilepsy in an individual, distinguish every withdrawal seizure from epilepsy, or establish a safe or unsafe concentration threshold.
Source
Original analysis of the supplied paper: Association Between Benzodiazepine Use and Epilepsy Occurrence A Nationwide Population Based Case Co. Paper identifier: PMC4635822.
Leave a Reply