Greenwood,D.C., Threapleton,D.E., Evans,C.E., et al, Br.J.Nutr. 112:725-734 (2014) 10.1017/S0007114514001329
OBJECTIVE: To review systematically and undertake a dose-response meta-analysis of the latest evidence relating to the reported association between intake of sugar-sweetened drinks and increased risk of type-2 diabetes, and whether this is related to sugar content or related lifestyle factors, whether similar associations hold for artificially sweetened soft drinks, and how these associations are related to BMI.
DESIGN: Systematic review and meta-analysis following a search of 7 databases - Medline, Medline in process, Embase, CAB abstracts, The Cochrane Library, ISI Web of science, and Biosis.
Eligible articles were published since 1990, in English language, reported >/= 3-year follow-up data, and reported type 2 diabetes risk in relation to intake of sugar-sweetened or artificially sweetened soft drinks.
Methodological quality was assessed using the Newcastle-Ottawa Scale.
SETTING: Evidence from prospective cohorts.
PARTICIPANTS: Generally healthy participants; n= ~280,000, including over 22,000 cases of type 2 diabetes.
MAIN OUTCOME MEASURES: Relative risk (RR) of type-2 diabetes.
RESULTS: Nine cohorts were identified, 5 from the USA, 5 from Europe and 2 from Asia. Six cohorts investigated the association between sugar-sweetened soft drink intake and risk of type 2 diabetes, and 4 cohorts investigated the association with artifically sweetened soft drinks. Estimated intakes of sugar-sweetened soft drinks ranged from 1 to 1000ml/day, and of artificially sweetened drinks from 0 to 500ml/day.
The pooled estimate of RR of type-2 diabetes from the linear dose-response meta-analysis was 1.2 per 330ml sugar-sweetened soft drinks per day (95% CI 1.12-1.29, P<0.001). Estimates with adjustment for BMI were slightly lower than those without adjustment (RR 1.16 vs RR 1.23).
For artificially sweetened soft drinks the pooled estimate of RR of type-2 diabetes from the linear dose-response meta-analysis was 1.13 per 330ml per day (95% CI 1.02-1.25, P<0.02).
There was substantial heterogeneity between studies and there was some evidence of mild nonlinearity in the dose response curves. The dose response curves for sugar sweetened and artificially sweetened soft drinks were similar with an intake of 330ml/day associated with ~20% increased risk.
CONCLUSIONS: The findings show a positive association between intake of sugar-sweetened soft drinks and risk of type-2 diabetes, which is attenuated by adjustment for BMI. However, there is also a trend for an association of increased risk with consumption of artificially sweetened soft drinks which may suggest an alternative explanation of the observed association such as other lifestyle factors or reverse causality. Other potential explanations include an association of artificially sweetened soft drinks with higher intakes of sugar from other components of the diet, or overcompensation for the perceived loss of energy from the artificially sweetened drink.
Note to Readers: Any opinions expressed in the recent research abstracts are those of the authors of the original scientific papers and may not reflect the views of Sugar Nutrition UK