To the Editor.—
Weiss (228:1368, 1974) objected to our passing reference (228:160, 1974) that Pearl's data on smoking and longevity (Science 87:216, 1938) appear to be compatible with other material we discussed on behavioral regulation of physiological homeostasis. He asserted, with pejoratives, that our comment is "not warranted" by Pearl's data. We must reply, with all due respect, that our statement, which Weiss quoted in his first paragraph, is an exact and true statement taken directly from Pearl's data. We said nothing about differences between the groups being statistically significant; we merely mentioned that differences existed—as anyone who reads Pearl's paper will see.Weiss acknowledged that "differences between the groups in specific mortality rates disappear... after age 70." However, if there were no significant differences in Pearl's mortality rates for five-year subsets of moderate smokers vs nonsmokers increasing from age 70, even for larger samples this absence in itself requires
Hickey RJ, Clelland RC, Boyce DE, Bowers EJ. Smoking, Health, and Confusion. JAMA. 1974;230(2):209–210. doi:10.1001/jama.1974.03240020017008
Customize your JAMA Network experience by selecting one or more topics from the list below.