—We cannot understand why Dr Prentice wishes only to dichotomize fat intake into below and above the median, which minimizes the contrast in intake. When the highest categories of fat intake (quintiles except in the study by Howe et al) are compared with the lowest categories, the relative risks were 0.90 in our study (1439 cases), 1.30 in the study by Howe et al (519 cases), and 1.13 in the study by Kushi et al (468 cases). Moreover, two additional large prospective studies have been published subsequent to ours; the corresponding relative risks were 1.08 in the study by van den Brandt et al1 with 471 cases and 0.99 in the study by Graham et al2 with 359 cases. Thus, the prospective studies clearly provide an overall association that is very close to the null value of 1.0 and substantially different from the pooled results from
Willett WC. Breast Cancer Risk From Diet, Tobacco, and Alcohol-Reply. JAMA. 1993;269(14):1792. doi:10.1001/jama.1993.03500140042030
Customize your JAMA Network experience by selecting one or more topics from the list below.