Dietary patterns and the risk of female breast cancer among participants of the Canadian national enhanced cancer surveillance system

OBJECTIVES: The role of diet in the etiology of breast cancer is not well understood despite extensive research. In the majority of this work, a single nutrient-based approach has been used which does not take into account combinations of food that are consumed. An alternative to the single nutrient approach is to identify patterns in the dietary intake information and relate these patterns to disease incidence. This investigation characterized dietary patterns among participants of a Canadian case-control study and related these dietary patterns to the incidence of breast cancer. METHODS: Dietary and other risk factor data from cases and controls of the Canadian National Enhanced Cancer Surveillance System were collected using self-administered questionnaires. Principal component factor analysis was used to classify individuals based on their dietary patterns. The relationship between these dietary patterns and breast cancer was evaluated using logistic regression. The derived odds ratios and their 95% confidence limits were adjusted for several factors, such as smoking, alcohol intake, environmental tobacco smoke exposure and obesity. RESULTS: In the 2,009 cases and 2,086 controls, three dietary patterns were identified: western, healthy and vitamin. The highest quartile of the 'healthy' dietary pattern was related to a 22% decrease in breast cancer risk (95% CI: 0.61-1.00), relative to the lowest quartile. The fourth quartile of the 'vitamin' dietary pattern was associated with a 14% decrease in breast cancer risk (95% CI: 0.70-1.04) relative to the first. No statistically significant associations between the 'western' dietary pattern and breast cancer were found. These associations were neither confounded nor modified by menopausal status. CONCLUSION: Our analyses reveal that individual dietary items tend to cluster together in such a way that there are three distinct dietary patterns in this sample of Canadian women. Some of these patterns, in turn, were associated with the risk of breast cancer. KEY WORDS: Breast cancer; case-control study; principal factor component analysis; dietary patterns OBJECTIFS: On connait mal le role du regime dans l'etiologie du cancer du sein malgre le grand nombre d'etudes sur le sujet. Dans la majorite de ces etudes, on utilise une demarche axee sur un nutriment particulier, ce qui ne tient pas compte des combinaisons d'aliments qui sont consommees. Une option de rechange a la demarche axee sur un seul nutriment consiste a reperer les grandes tendances de l'information sur les apports alimentaires et a les lier aux structures d'incidence des maladies. Notre enquete a caracterise les schemas alimentaires des participantes d'une etude castemoins canadienne et les ont lies a l'incidence du cancer du sein. METHODE: Des donnees sur le regime alimentaire et d'autres facteurs de risque provenant de cas et de temoins du Systeme national de surveillance accrue du cancer ont ete recueillies a l'aide de questionnaires a remplir soimeme. Nous avons utilise l'analyse factorielle en composantes principales pour classifier les repondantes d'apres leurs habitudes alimentaires. Nous avons evalue la relation entre ces schemas alimentaires et le cancer du sein au moyen d'analyses de regression logistique. Les rapports de cotes derives et leurs intervalles de confiance de 95% ont ete ajustes selon plusieurs facteurs, dont le tabagisme, la consommation d'alcool, l'exposition a la fumee secondaire du tabac et l'obesite. RESULTATS: Chez les 2 009 cas et les 2 086 temoins, nous avons defini trois schemas alimentaires : occidental, sain et vitamine. Le quartile superieur du schema alimentaire << sain >> etait lie a une reduction de 22% du risque de cancer du sein (IC de 95% : 0,61-1,00), comparativement au quartile inferieur. Le quatrieme quartile du schema alimentaire << vitamine >> etait associe a une reduction de 14% du risque de cancer du sein (IC de 95% : 0,70-1,04) comparativement au premier. Nous n'avons trouve aucune association significative entre le schema alimentaire << occidental >> et le cancer du sein. Ces associations n'etaient ni faussees, ni modifiees par le statut menopausique. CONCLUSION: Nos analyses revelent que les choix alimentaires individuels ont tendance a former des grappes qui dessinent trois schemas alimentaires distincts dans cet echantillon de Canadiennes. Certains de ces schemas etaient a leur tour associes au risque de cancer du sein. MOTS CLES: tumeurs du sein; etudes cas-temoins; analyse factorielle en composantes principales; habitudes alimentaires La traduction du resume se trouve a la fin de l'article. Can J Public Health 2016;107(1):e49-e55 doi: 10.17269/CJPH.107.5230
Breast cancer is a global health concern, with incidence rates steady or climbing in high-income countries and both incidence and mortality rates increasing in other countries. (1) In Canada, the [...]