A large meta-analysis, recently published in the BMJ, has shown that drinking sugar sweetened beverages raises the risk of type 2 diabetes, independent of obesity.

PubMed, Embase, Ovid and Web of Knowledge were searched for eligible papers. Studies were graded for quality and discussed using a standard protocol.

From a total of 1,937 articles, 33 were reviewed in full text and 21 articles, following 16 cohorts of patients, were included in the analysis. Unpublished but eligible data on a further cohort were added giving an overall total of 17 cohorts. The analysis covered 38,253 cases of type 2 diabetes over 10,126,754 person-years in total. None of the studies analysed was funded by industry.

Information was gathered about the intake of sugar sweetened drinks, artificially sweetened drinks and fruit juices. The data were standardised to measure the association of one serving of 250 ml per day on the risk of incident diabetes.

For sugar sweetened drinks, one 250 ml drink per day was associated with an 18% greater incidence of type 2 diabetes (95% CI: 8.8-28%). When adjusted for obesity the risk was attenuated to a 13% increase (95% CI: 5.8-21%), see table 1.

Artificially sweetened drinks were also associated with a greatly increased incidence of diabetes; 25% (95% CI: 18-33%), but this was attenuated to 8% after adjustment for obesity (95% CI: 2.1-15%).

Fruit juice intake was also associated with a raised risk of incident diabetes, 5% (95% CI: -1 to 11%) and this was increased after adjustment for obesity, to 7% (95% CI: 0.8-14%).

Further statistical analysis for confounding factors suggested that the sugar sweetened drink risk was likely to be a true finding, but the increased risk from artificially sweetened drinks and fruit juices was likely to be exaggerated by publication bias and confounding variables.

Assuming a causal relationship between increased intake of sugar sweetened drinks and type 2 diabetes, the authors estimated the increased population risk for the UK and the USA. Over a decade, they predicted 79,000 extra cases in the UK and 1.8 million in the USA. A large proportion of the cases would be in younger people and men.

The novel aspect of this paper is that the researchers have quantified the size of the problem for public health.

The results would suggest that in the UK, between 2010 and 2020, 2-6% of new cases of type 2 diabetes would be attributable to the consumption of sugar sweetened beverages.

This figure is higher in the USA, where more sugar sweetened drinks are sold, estimated as 4-13% of new cases in the same period.

The publication of this study comes at a time when the WHO has recommended a reduction in the percentage of energy in our diet from sugar to less than 10%.1 There is also a current political debate about whether taxing sugar sweetened drinks would be a way of modifying dietary intake in the UK. The potential impact of the predicted extra cases of type 2 diabetes on NHS resources adds to this debate.

Dr Matthew Lockyer