This Article  • References  • Full text PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Citing articles on HighWire  • Contact me when this article is cited  Related Content  • Similar articles in this journal  Social Bookmarking   What's this?

Arthur L. Klatsky, MD; Gary D. Friedman, MD; Mary A. Armstrong, MA
Oakland, Calif

Arch Intern Med. 1993;153(7):902.

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

We believe that Myers and Basinski¹ erred in omitting our recent large cohort study of coffee and coronary heart disease (CHD)² from their meta-analysis. One reason given was that they regarded it as a case-control study, in which there was "matching of 740 participants in the Kaiser Permanente Program hospitalized with approximately 10 000 individuals in the program without history of myocardial infarction." Actually, we started with a cohort of 101 774 participants who had indicated their coffee consumption on a questionnaire between January 1978 and December 1985 and followed them up through December 1986 for hospitalization for CHD. This was clearly a cohort study. Merely to reduce computer costs, we randomly selected without matching, 10% of the original cohort, which provided the 9986 noncases included in our analysis along with all of the cases. The second reason given by Myers and Basinski was that cases were limited . . . [Full Text PDF of this Article]

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter     What's this?