In Reply.—
Goldstein raises the important question of possible effects of nut consumption on other causes of morbidity and mortality aside from coronary heart disease. We have looked at this question in our database as far as is possible. Associations between nut consumption and all incident cancers, and also between nut consumption and all-cause mortality, can be evaluated. Using Mantel-Haenszel analyses, stratifying on age, sex, and vegetarian status, the relative risks of incident cancer for the three consumption categories of nuts (</wk, 1 to 4/wk, ⩾5/wk) were 1.0, 0.92, 0.90 (not significant). For all-cause mortality, the corresponding relative risks are 1.0, 0.72, 0.67 (P<.00001). Thus, there is no evidence for any increase in risk of cancer, and, clearly, a decreased risk of all-cause mortality, that we presume largely reflects the decrease in coronary heart disease mortality. We hope that Goldstein may now feel more comfortable with his "nuts on whole wheat."