June 7, 1985

Lowering Cholesterol and the Incidence of Coronary Heart Disease

Author Affiliations

University of Otago Medical School Dunedin, New Zealand

JAMA. 1985;253(21):3090. doi:10.1001/jama.1985.03350450062006

To the Editor.—  Being interested in medicine and having a pocket calculator, I have done some sums on the data from the Lipid Research Clinics Coronary Primary Prevention Trial results (LRC-CPPT).1 A group of 1,906 men received cholestyramine resin for seven years in comparison with 1,900 men who took a placebo. Two such groups are unlikely to have exactly the same incidence of coronary heart disease (CHD). One group will have more than the other. The point of interest is whether the difference is significant. The Table shows that it is not, if the proper two-sided statistical test is used. In any case, there is nothing magic about the one in 20 chance. On average, it crops up once in every 20 tests of difference between samples from the same population, and usually disappears on a second test. The point to be faced is that the trial data do