Intake of dietary fats and colorectal cancer risk: Prospective findings from the UK Dietary Cohort Consortium

• Published in: Cancer epidemiology 2010-10, Vol.34 (5), p.562-567
• Main Authors: Dahm, Christina C, Keogh, Ruth H, Lentjes, Marleen A.H, Spencer, Elizabeth A, Key, Tim J, Greenwood, Darren C, Cade, Janet E, Burley, Victoria J, Shipley, Martin J, Brunner, Eric J, Stephen, Alison M, Mishra, Gita, Kuh, Diana, Fentiman, Ian S, White, Ian R, Luben, Robert, Khaw, Kay Tee, Rodwell (Bingham), Sheila A
• Format: Article
• Language: English
• Subjects: Cancer; Case-Control Studies; Cohort Studies; Colorectal cancer; Colorectal Neoplasms - epidemiology; Colorectal Neoplasms - etiology; Diet records; Dietary fats; Dietary Fats - administration & dosage; Epidemiology; Female; Hematology, Oncology and Palliative Medicine; Humans; Internal Medicine; Logistic Models; Male; Middle Aged; Motor Activity; Multivariate Analysis; Prevention of cancer; Prospective Studies; United Kingdom - epidemiology
• ISSN: 1877-7821; 1877-783X
• DOI: 10.1016/j.canep.2010.07.008

Abstract Introduction : Epidemiologic evidence for an association between colorectal cancer (CRC) risk and total dietary fat, saturated fat (SF), monounsaturated fat (MUFA) and polyunsaturated fat (PUFA) is inconsistent. Previous studies have used food frequency questionnaires (FFQ) to assess diet, but data from food diaries may be less prone to severe measurement error than data from FFQ. Methods : We conducted a case–control study nested within seven prospective UK cohort studies, comprising 579 cases of incident CRC and 1996 matched controls. Standardized dietary data from 4- to 7-day food diaries and from FFQ were used to estimate odds ratios for CRC risk associated with intake of fat and subtypes of fat using conditional logistic regression. We also calculated multivariate measurement error corrected odds ratios for CRC using repeated food diary measurements. Results : We observed no associations between intakes of total dietary fat or types of fat and CRC risk, irrespective of whether dietary data were obtained using food diaries or FFQ. Conclusion : Our results do not support the hypothesis that intakes of total dietary fat, SF, MUFA or PUFA are linked to risk of CRC.