To the Editor.—
The article by Mustacchi and Greenspan (237:1446-1447, 1977) reanalyzes the data of Kapdi and Wolfe (236:1124-1127, 1976) to reach the conclusion that "the retrospective study of Detroit women does not contain information indicating that premedication with thyroid hormone affected their risk of breast cancer either beneficially or adversely." That conclusion is not valid.It is true that Kapdi and Wolfe did not provide the strongest analysis of their data and that their formal statistical testing was done on tables in which confounding effects (such as age) were not excluded. It is also true that Mustacchi and Greenspan show that confounding could have affected those tests.On the other hand, Mustacchi and Greenspan performed tests of a composite null hypothesis (that breast cancer incidence is independent of age and thyroid therapy) and report only that "more than half of the contribution to the χ2 value comes from
Venezian EC. Thyroid Hormone and Breast Cancer. JAMA. 1977;238(11):1147. doi:10.1001/jama.1977.03280120039004
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: