Estimated substitution of tea or coffee for sugar-sweetened beverages was associated with lower type 2 diabetes incidence in case-cohort analysis across 8 European countries in the EPIC-InterAct study

Introduction: Beverage consumption is a modifiable risk factor for type 2 diabetes (T2D), but there is insufficient evidence to inform the suitability of substituting 1 type of beverage for another. Objective: The aim of this study was to estimate the risk of T2D when consumption of sugar-sweetened beverages (SSBs) was replaced with consumption of fruit juice, milk, coffee, or tea. Methods: In the European Prospective Investigation into Cancer and Nutrition (EPIC)-InterAct case-cohort study of 8 European countries (n = 27,662, with 12,333 cases of incident T2D, 1992-2007), beverage consumption was estimated at baseline by dietary questionnaires. Using Prentice-weighted Cox regression adjusting for other beverages and potential confounders, we estimated associations of substituting 1 type of beverage for another on incident T2D. Results: Mean +/- SD of estimated consumption of SSB was 55 +/- 105 g/d. Means +/- SDs for the other beverages were as follows: fruit juice, 59 +/- 101 g/d; milk, 209 +/- 203 g/d; coffee, 381 +/- 372 g/d; and tea, 152 +/- 282 g/d. Substituting coffee for SSBs by 250 g/d was associated with a 21% lower incidence of T2D (95% CI: 12%, 29%). The rate difference was -12.0 (95% CI: -20.0, -5.0) per 10,000 person-years among adults consuming SSBs >= 250 g/d (absolute rate = 48.3/10,000). Substituting tea for SSBs was estimated to lower T2D incidence by 22% (95% CI: 15%, 28%) or -11.0 (95% CI:-20.0, -2.6) per 10,000 person-years, whereas substituting fruit juice or milk was estimated not to alter T2D risk significantly. Conclusions: These findings indicate a potential benefit of substituting coffee or tea for SSBs for the primary prevention of T2D and may help formulate public health recommendations on beverage consumption in different populations.
The InterAct project was funded by the EU FP6 program (LSHM_CT_2006_037197). InterAct investigators acknowledge funding from the following agencies: Medical Research Council Epidemiology Unit Core Support (MC_UU_12015/1, MC_UU_12015/5); National Institute for Health Research, Biomedical Research Centre Cambridge: Nutrition, Diet, and Lifestyle Research Theme (IS-BRC-1215-20014); the German Federal Ministry of Education and Research, the German Center for Diabetes Research, the State of Brandenburg, Germany; Netherlands Agency grant IGE05012 and an Incentive Grant from the Board of the University Medical Center Utrecht (Netherlands); Dutch Ministry of Public Health, Welfare, and Sports, Netherlands Cancer Registry, LK Research Funds, Dutch Prevention Funds, Dutch ZON (Zorg Onderzoek Nederland), World Cancer Research Fund, and Statistics Netherlands; Swedish Research Council, Novo Nordisk, Swedish Heart Lung Foundation, and Swedish Diabetes Association; Danish Cancer Society; Deutsche Krebshilfe; Associazione Italiana per la Ricerca sul Cancro; Asturias Regional Government; Navarra Regional Government, Health Research Fund of the Spanish Ministry of Health; CIBER Epidemiologia y Salud Publica, Spain; Murcia Regional Government (grant 6236); and AIRE-ONLUS Ragusa, AVIS-Ragusa, and Sicilian Regional Government.