This Article  • Full text  • PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Contact me when this article is cited  Related Content  • Related letter  • Related article  • Similar articles in JAMA  Topic Collections  • Statistics and Research Methods  • Genetics, Other  • Alert me on articles by topic

	Since this article does not have an abstract, we have provided the first 150 words of the full text and any section headings.

To the Editor: In their assessment of genetic associations, Dr Patsopoulus and colleagues¹ analyzed claims of sex-gene interactions.¹ They seem to imply that the criterion standard for evaluating such interactions is the statistical significance level of the test for sex by gene interaction, generally stated in terms of an odds ratio or relative risk (RR). However, in those instances in which the underlying incidence rates differ by sex, such tests of interaction are generally arbitrary.

If the incidence rate (or for quantitative factors such as cholesterol, the distribution) is the same in men and women, then a difference in RR implies a difference of the gene effect on any scale (eg, multiplicative, additive); in such cases testing for interaction may be biologically meaningful. However, when the incidence (or distribution) differs significantly by sex, as it commonly does in the example studies cited in Patsopoulos et al, tests for interaction are . . . [Full Text of this Article]

Paul F. Pinsky, MD
pp4f@nih.gov
Division of Cancer Prevention
National Cancer Institute
Bethesda, Maryland

RELATED LETTER

Claims of Sex-Gene Interactions—Reply
Nikolaos A. Patsopoulos, Athina Tatsioni, and John P. A. Ioannidis
JAMA. 2007;298(23):2742.
EXTRACT | FULL TEXT  

RELATED ARTICLE

Claims of Sex Differences: An Empirical Assessment in Genetic Associations
Nikolaos A. Patsopoulos, Athina Tatsioni, and John P. A. Ioannidis
JAMA. 2007;298(8):880-893.
ABSTRACT | FULL TEXT