[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address Please contact the publisher to request reinstatement.
[Skip to Content Landing]
June 10, 1974

Smoking and Statistical Significance

Author Affiliations

Hahnemann Medical College Philadelphia

JAMA. 1974;228(11):1368. doi:10.1001/jama.1974.03230360016003

To the Editor.—  While commenting on coffee, tobacco, and cardiovascular disease in a letter to the editor (228:160, 1974), Hickey et al stated that "Pearl, in a prospective study of males, found more survivors among moderate smokers than among nonsmokers or heavy smokers at age 95, and that for five-year subsets increasing from age 70, the mortality rates of moderate smokers were lower than rates for nonsmokers or heavy smokers." This statement is not warranted by the data in Pearl's paper.1Pearl's study was based on 2,094 nonsmokers, 2,814 moderate smokers, and 1,905 heavy smokers. It is true that the death rates for moderate smokers aged 75 and over were slightly smaller than those for nonsmokers and heavy smokers, but the rates were undoubtedly calculated from populations of such small size that the differences could not possibly be statistically significant. Figure 1 in Pearl's paper, based on the calculated