Sugar-Sweetened Beverage Intake and Diabetes Risk: A systematic Review and Meta-analysis

Abstract

Background: The consumption of sugar-sweetened beverages has been increasing continuously worldwide. Drinking sugar-sweetened beverages has been suspected to be a risk factor of developing diabetes mellitus but results from previous were conflicting. Objective: To performed a systematic review and meta-analysis aiming to estimate the effect size of sugar-sweetened beverages on the risk of developing diabetes mellitus. Methods: Medicine and Scopus databases were comprehensively searched for relevant studied. Two reviewers selected studies based on titles and abstracts. Observational studies published in English were selected, if they met all of these following criteria: 1) considered sugar-sweetened beverages as an interested risk factor, 2) measured the outcomes as having or not having diabetes mellitus, 3) provided adequate data for pooling the effect size. Odds ratios or relative riskd of having type 2 diabetes mellitus of each study were pooled by using random effect model, if heterogeneity between studies presents. If not, the fixed effect model with inverse variance method was used. Sources of heterogeneity were assessed by fitting co-variables (such as age of patient, study design, assessment method) one by one in meta-regression. Results: Eight out of 1439 studies were finally eligible in our review. Compared with never or seldom drinking, drinking sugar-sweetened beverages equal or more than on serving per week significantly increased risk of having diabetes mellitus with pooled odds ratio of 1.30 (95% CI, 1.05 - 1.62). Pooling odds ratio adjusted with body mass index still showed the significant risk effect of drinking sugar-sweetened beverages equal or more than one serving per week with pooled adjusted odds ratio of 1.25 (95% CI, 1.19 - 1.31). Conclusions: Results from our study suggested that drinking sugar-sweetened beverages equal or more than one serving per week significantly increased risk of having diabetes mellitus. This risk effect still persisted even after adjusting the effect of body mass index.