This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
None of the differences in cause-specific mortality that Dr Gimlett cites, nor the difference between the two LRC-CPPT treatment groups in all-cause mortality, was statistically significant. Note, however, that from its inception, the LRC-CPPT was testing the hypothesis that the combined incidence of definite CHD death and nonfatal myocardial infarction, which was the "primary end point" specified in the protocol, would be lower in the cholestyramine group than in the placebo-treated controls.A study at least five times larger than the LRC-CPPT would have been required to test the hypothesis that cholesterol reduction lowers either all-cause or cardiovascular mortality. It is the statistically significant difference in incidence of the prespecified primary end point rather than the nonsignificant difference in CHD mortality that enables one to conclude that treatment was beneficial. While the trend toward more accidental deaths, suicides, and homicides in the cholestyramine group has to be noted,
Rifkind BM. Lipid Research Clinics Program-Reply. JAMA. 1984;252(18):2546. doi:10.1001/jama.1984.03350180015012