[Skip to Navigation]
Sign In
Editor's Correspondence
March 12, 2001

Observational Data Do Not Establish Cause and Effect

Arch Intern Med. 2001;161(5):771-778. doi:

Dr Mosca,1 in her comprehensive review of the literature on the role of hormone replacement therapy (HRT) in the prevention of postmenopausal heart disease, falls into a trap that is all too common: she attributes causality to associations found in observational studies. Consider the following quotes from her article that appear in the "Observational Studies" subsection of the "ERT [estrogen replacement therapy]/HRT and CHD [coronary heart disease] Events" section:

Add or change institution