Romy F Willemsen1,2, Jessica McNeil2,3, Emily Heer2, Steven T Johnson4, Christine M Friedenreich2,5, Darren R Brenner6,7. 1. Division of Human Nutrition and Health, Wageningen University, Wageningen, The Netherlands. 2. Department of Cancer Epidemiology and Prevention Research, CancerControl Alberta, Alberta Health Services, Calgary, AB, Canada. 3. Department of Kinesiology, Faculty of Health and Human Sciences, University of North Carolina Greensboro, Greensboro, North Carolina, USA. 4. Faculty of Health Disciplines, Athabasca University, Athabasca, AB, Canada. 5. Departments of Oncology and Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada. 6. Department of Cancer Epidemiology and Prevention Research, CancerControl Alberta, Alberta Health Services, Calgary, AB, Canada. darren.brenner@ucalgary.ca. 7. Departments of Oncology and Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada. darren.brenner@ucalgary.ca.
Abstract
BACKGROUND/ OBJECTIVES: Poor diet quality has been associated with an increased risk of cancer. Here, we examine the association between dietary patterns derived with two methods, and combined and site-specific cancer incidence in Canada. SUBJECTS/ METHODS: Dietary data were obtained from participants enrolled in Alberta's Tomorrow Project, a prospective cohort study, between 2000 and 2008. Principle component analysis (PCA) and reduced rank regression (RRR) were used to derive dietary patterns, and data linkage with the Alberta Cancer Registry was used for incident cancer cases. Cox proportional hazard regressions were used to estimate multivariable-adjusted models for the association between each dietary pattern score with combined and site-specific cancer incidence. RESULTS: PCA revealed three dietary patterns ("western", "prudent", and "sugar, fruits, and dairy") and RRR resulted in four patterns ("dietary fiber", "vitamin D", "fructose", and "discretionary fat"). Five cancer sites were included in our site-specific analysis: lung, colon, breast, prostate, and endometrial cancers. The most protective dietary patterns for combined cancer sites were the "Prudent" pattern (HR = 0.82, CI = 0.73-0.92) and the "Dietary fiber" pattern (HR = 0.82, CI = 0.69-0.97). The "Fructose" pattern was associated with increased risk of combined cancers (HR = 1.14, CI = 1.02-1.27). Three dietary patterns were protective against colon cancer ("Prudent", "Dietary fiber", and "Discretionary fats"), and other risk reductions were seen for the "sugar, fruit, and dairy" pattern (lung cancer), and the "Dietary fiber" pattern (prostate cancer). CONCLUSIONS: These results support cancer prevention strategies for a diet high in vegetables, fruits, fish, and whole grains. Further studies should explore the possible association between discretionary fats and colon cancer.
BACKGROUND/ OBJECTIVES: Poor diet quality has been associated with an increased risk of cancer. Here, we examine the association between dietary patterns derived with two methods, and combined and site-specific cancer incidence in Canada. SUBJECTS/ METHODS: Dietary data were obtained from participants enrolled in Alberta's Tomorrow Project, a prospective cohort study, between 2000 and 2008. Principle component analysis (PCA) and reduced rank regression (RRR) were used to derive dietary patterns, and data linkage with the Alberta Cancer Registry was used for incident cancer cases. Cox proportional hazard regressions were used to estimate multivariable-adjusted models for the association between each dietary pattern score with combined and site-specific cancer incidence. RESULTS: PCA revealed three dietary patterns ("western", "prudent", and "sugar, fruits, and dairy") and RRR resulted in four patterns ("dietary fiber", "vitamin D", "fructose", and "discretionary fat"). Five cancer sites were included in our site-specific analysis: lung, colon, breast, prostate, and endometrial cancers. The most protective dietary patterns for combined cancer sites were the "Prudent" pattern (HR = 0.82, CI = 0.73-0.92) and the "Dietary fiber" pattern (HR = 0.82, CI = 0.69-0.97). The "Fructose" pattern was associated with increased risk of combined cancers (HR = 1.14, CI = 1.02-1.27). Three dietary patterns were protective against colon cancer ("Prudent", "Dietary fiber", and "Discretionary fats"), and other risk reductions were seen for the "sugar, fruit, and dairy" pattern (lung cancer), and the "Dietary fiber" pattern (prostate cancer). CONCLUSIONS: These results support cancer prevention strategies for a diet high in vegetables, fruits, fish, and whole grains. Further studies should explore the possible association between discretionary fats and colon cancer.
Authors: A F Subar; D Midthune; M Kulldorff; C C Brown; F E Thompson; V Kipnis; A Schatzkin Journal: Am J Epidemiol Date: 2000-08-01 Impact factor: 4.897
Authors: A F Subar; F E Thompson; V Kipnis; D Midthune; P Hurwitz; S McNutt; A McIntosh; S Rosenfeld Journal: Am J Epidemiol Date: 2001-12-15 Impact factor: 4.897
Authors: Abbey E Poirier; Yibing Ruan; Lauren A Hebert; Xin Grevers; Stephen D Walter; Paul J Villeneuve; Darren R Brenner; Christine M Friedenreich Journal: Prev Med Date: 2019-05 Impact factor: 4.018
Authors: Abbey E Poirier; Yibing Ruan; Karena D Volesky; Will D King; Dylan E O'Sullivan; Priyanka Gogna; Stephen D Walter; Paul J Villeneuve; Christine M Friedenreich; Darren R Brenner Journal: Prev Med Date: 2019-05 Impact factor: 4.018
Authors: Nathan M Solbak; Jian-Yi Xu; Jennifer E Vena; Ilona Csizmadi; Heather K Whelan; Paula J Robson Journal: Prev Med Date: 2017-06-07 Impact factor: 4.018