[Skip to Content]
[Skip to Content Landing]
April 12, 1993

Coffee and Coronary Heart Disease

Author Affiliations

Oakland, Calif

Arch Intern Med. 1993;153(7):902. doi:10.1001/archinte.1993.00410070080015

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

First Page Preview View Large
First page PDF preview
First page PDF preview