Validity of dietary assessment – reply to B Watzl and G Rechkemmer
Sommaire de l'article
We are grateful for the opportunity to inform readers of an error in our publication in the Journal of the National Cancer Institute in November 2000 [(1); erratum in press]. The error, which does not affect the conclusions, pertains to the distribution of men (in the Health Professionals Follow-Up Study cohort) across categories of servings of vegetables per day [Table 1 of the article by Michels et al (1)]. It is shown in Table 1 that 89% of men in the Health Professionals Follow-Up Study cohort ate ≤ 2 servings of vegetables daily; however, this is incorrect. The median intake of total vegetables in this cohort was 3.1 servings/d; the distribution is presented in Table 1 of the article by Michaud et al (2). At baseline, almost 20% of men consumed ≥ 5 servings of vegetables daily.
We would like to raise a few additional points. In our population, similar to most populations in the United States, lycopene contributed most to the total carotenoid values (3). The main sources of lycopene are tomato-based products, such as tomato sauce and tomato soup. These items may not be commonly consumed in some other non-US populations, and are often not included in the calculations of total fruit and vegetable intake. Thus, fruit and vegetable and carotenoid intakes across populations are not directly comparable. Additionally, the median values for total carotenoid intake from Table 1 of the publication in the American Journal of Clinical Nutrition (3) were based on categories and not quintiles. As detailed in the footnote of the table and in the Methods, high intake corresponded to participants consuming high amounts (fifth quintile) of ≥ 3 of the individual carotenoids, whereas individuals consuming low amounts (first quintile) of ≥ 3 of the individual carotenoids were placed in the bottom category. The more restrictive categorization based on simultaneous high consumption of multiple carotenoids created more extreme comparisons than quintiles of total intake.
Watzl and Rechkemmer were also concerned about the low prevalence of diet-related cancers in these 2 healthy cohorts. Although the incidence of certain cancers in these 2 cohorts is lower than that in the average US population, the range of fruit and vegetable intake is quite large. Thus, the relation between fruit and vegetable intake and cancer risk can be determined over a wide range of intakes. The associations observed over comparable ranges of intakes in these populations should be similar in other, maybe less educated, populations. Although overall rates of lung cancer were lower in these populations largely because of lower smoking prevalences, we did examine a wide range of fruit and vegetable intakes over strata of current smokers, past smokers, and never smokers.
1. Michels KB, Giovannucci E, Joshipura KJ, et al. Prospective study of fruit and vegetable consumption and incidence of colon and rectal cancers. J Natl Cancer Inst 2000;92:1740–52.
2. Michaud D, Spiegelman D, Clinton S, Rimm E, Willett W, Giovannucci E. Fruit and vegetable intake and incidence of bladder cancer in a male prospective cohort. J Natl Cancer Inst 1999;91:605–13.
3. Michaud DS, Feskanich D, Rimm EB, et al. Intake of specific carotenoids and risk of lung cancer in 2 prospective US cohorts. Am J Clin Nutr 2000;72:990–7