Background: Medical and public health decisions are informed by reviews, which makes the quality of reviews an important scientific concern.
Objective: We systematically assessed the quality of published reviews on sugar-sweetened beverages (SSBs) and health, which is a controversial topic that is important to public health.
Design: We performed a search of PubMed and Cochrane databases and a hand search of reference lists. Studies that were selected were published reviews and meta-analyses (June 2001 to June 2011) of epidemiologic studies of the relation between SSBs and obesity, type 2 diabetes, metabolic syndrome, and coronary heart disease. A standardized data-abstraction form was used. Review quality was assessed by using the validated instrument AMSTAR (assessment of multiple systematic reviews), which is a one-page tool with 11 questions.
Results: Seventeen reviews met our inclusion and exclusion criteria: obesity or weight (16 reviews), diabetes (3 reviews), metabolic syndrome (3 reviews), and coronary heart disease (2 reviews). Authors frequently used a strictly narrative review (7 of 17 reviews). Only 6 of 17 reviews reported quantitative data in a table format. Overall, reviews of SSBs and health outcomes received moderately low-quality scores by the AMSTAR [mean: 4.4 points; median: 4 points; range: 1-8.5 points (out of a possible score of 11 points)]. AMSTAR scores were not related to the conclusions of authors (8 reviews reported an association with a mean AMSTAR score of 4.1 points; 9 reviews with equivocal conclusions scored 4.7 points; P value = 0.84). Less than one-third of published reviews reported a comprehensive literature search, listed included and excluded studies, or used duplicate study selection and data abstraction.
Conclusion: The comprehensive reporting of epidemiologic evidence and use of systematic methodologies to interpret evidence were underused in published reviews on SSBs and health.