Letters Section Editor: Stephen J. Lurie,
MD, PhD, Senior Editor.
In Reply: Our data offer no evidence regarding
the interesting suggestion of Dr Smith and colleagues that gonadotropin levels
are a mediating factor in the relationship between HRT and AD.
The remaining letters urge that our data not be overinterpreted. Indeed,
our data cannot prove a causal relationship between HRT use and AD incidence.
Users of HRT do differ from nonusers in health habits and socioeconomic status.
The potential for confounding with these factors is certainly as great in
AD as in other disorders. Experience suggests, however, that some control
on such confounding is possible. The epidemiologic literature on HRT includes
older studies without such control and newer studies that adjust for health
habits and socioeconomic distinctions. Most of the inverse association of
HRT with cardiovascular risks disappears when such confounders are controlled.1 This finding suggests that the results of the Women's
Health Initiative trial2 were not unanticipated.
Zandi PP, Breitner JCS. Estrogen Replacement and Risk of Alzheimer Disease—ReplyEstrogen Replacement and Risk of Alzheimer Disease—Reply. JAMA. 2003;289(9):1100-1102. doi:10.1001/jama.289.9.1100-a