We appreciate the comments of Dr Meuleman in response to our study. His concerns center on establishing a causal relationship between diphenhydramine use and adverse cognitive outcomes. We would therefore like to review the 5 well-established criteria for causality in clinical epidemiologic studies1: consistency of association, strength of association, specificity of association, temporal plausibility, and biologic plausibility.
Agostini JV, Inouye SK. Association of Diphenhydramine Use With Adverse Effects in Hospitalized Older Patients: Possible Confounders—Reply. Arch Intern Med. 2002;162(6):720-721. doi: