Prospective associations and population impact of sweet beverage intake and type 2 diabetes, and effects of substitutions with alternative beverages

Diabetologia. 2015 Jul;58(7):1474-83. doi: 10.1007/s00125-015-3572-1. Epub 2015 May 6.


Aims/hypothesis: This study aimed to evaluate the association of types of sugar-sweetened beverages (SSB) (soft drinks, sweetened-milk beverages, sweetened tea/coffee), artificially sweetened beverages (ASB) and fruit juice with incident type 2 diabetes and determine the effects of substituting non-SSB for SSB and the population-attributable fraction of type 2 diabetes due to total sweet beverages.

Methods: Beverage consumption of 25,639 UK-resident adults without diabetes at baseline (1993-1997) in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Norfolk study was assessed using 7-day food diaries. During 10.8 years of follow-up 847 incident type 2 diabetes cases were verified.

Results: In adjusted Cox regression analyses there were positive associations (HR [95% CI] per serving/day]) for soft drinks 1.21 (1.05, 1.39), sweetened-milk beverages 1.22 (1.05, 1.43) and ASB 1.22 (1.11, 1.33), but not for sweetened tea/coffee 0.98 (0.94, 1.02) or fruit juice 1.01 (0.88, 1.15). Further adjustment for adiposity attenuated the association of ASB, HR 1.06 (0.93, 1.20). There was a positive dose-response relationship with total sweet beverages: HR per 5% energy 1.18 (1.11, 1.26). Substituting ASB for any SSB did not reduce the incidence in analyses accounting for energy intake and adiposity. Substituting one serving/day of water or unsweetened tea/coffee for soft drinks and for sweetened-milk beverages reduced the incidence by 14%-25%. If sweet beverage consumers reduced intake to below 2% energy, 15% of incident diabetes might be prevented.

Conclusions/interpretation: The consumption of soft drinks, sweetened-milk beverages and energy from total sweet beverages was associated with higher type 2 diabetes risk independently of adiposity. Water or unsweetened tea/coffee appear to be suitable alternatives to SSB for diabetes prevention. These findings support the implementation of population-based interventions to reduce SSB consumption and increase the consumption of suitable alternatives.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adiposity
  • Adult
  • Aged
  • Animals
  • Beverages / adverse effects*
  • Coffee
  • Cohort Studies
  • Diabetes Mellitus, Type 2 / epidemiology*
  • Diet
  • Diet Records
  • Dose-Response Relationship, Drug
  • Female
  • Fruit
  • Humans
  • Male
  • Middle Aged
  • Milk
  • Population
  • Prospective Studies
  • Sweetening Agents / adverse effects*
  • Tea
  • United Kingdom / epidemiology


  • Coffee
  • Sweetening Agents
  • Tea