Childhood energy intake and adult mortality from cancerBMJ 1998; 317 doi: https://doi.org/10.1136/bmj.317.7155.414a (Published 08 August 1998) Cite this as: BMJ 1998;317:414
Authors should have used family as unit of analysis
- Charlotte Wright, First assistant in community child health
- Department of Child Health, Community Child Health, Gateshead NE8 1EB
- Department of Social Medicine, University of Bristol, Bristol BS8 2PR
EDITOR—Frankel et al suggest that they have shown that restriction of nutrients in childhood protects against cancers not related to smoking.1 Unfortunately, while working ingeniously with the data, they seem not to appreciate their more fundamental failings. Individual dietary assessment as a research tool is notoriously unreliable,2 and when individual intake is estimated from family intake the potential for bias and error multiplies.3 The authors acknowledge this in their discussion, but they still present a misleadingly precise looking table of age specific mean intakes, which must actually have been estimated from the family data.
An association was found despite the imprecision of the survey instrument, but the authors have not adjusted for all confounders. The two main determinants of energy intake are lean body mass and activity levels. Activity levels are not likely to be associated with risk of cancer, but lean body mass is …