BACKGROUND: A growing number of cohort studies suggest a potential role of dairy consumption in type 2 diabetes (T2D) prevention. The strength of this association and the amount of dairy needed is not clear. OBJECTIVE: We performed a meta-analysis to quantify the associations of incident T2D with dairy foods at different levels of intake. DESIGN: A systematic literature search of the PubMed, Scopus, and Embase databases (from inception to 14 April 2015) was supplemented by hand searches of reference lists and correspondence with authors of prior studies. Included were prospective cohort studies that examined the association between dairy and incident T2D in healthy adults. Data were extracted with the use of a predefined protocol, with double data-entry and study quality assessments. Random-effects meta-analyses with summarized dose-response data were performed for total, low-fat, and high-fat dairy, (types of) milk, (types of) fermented dairy, cream, ice cream, and sherbet. Nonlinear associations were investigated, with data modeled with the use of spline knots and visualized via spaghetti plots. RESULTS: The analysis included 22 cohort studies comprised of 579,832 individuals and 43,118 T2D cases. Total dairy was inversely associated with T2D risk (RR: 0.97 per 200-g/d increment; 95% CI: 0.95, 1.00; P = 0.04; I2 = 66%), with a suggestive but similar linear inverse association noted for low-fat dairy (RR: 0.96 per 200 g/d; 95% CI: 0.92, 1.00; P = 0.072; I2 = 68%). Nonlinear inverse associations were found for yogurt intake (at 80 g/d, RR: 0.86 compared with 0 g/d; 95% CI: 0.83, 0.90; P < 0.001; I2 = 73%) and ice cream intake (at ∼10 g/d, RR: 0.81; 95% CI: 0.78, 0.85; P < 0.001; I2 = 86%), but no added incremental benefits were found at a higher intake. Other dairy types were not associated with T2D risk. CONCLUSION: This dose-response meta-analysis of observational studies suggests a possible role for dairy foods, particularly yogurt, in the prevention of T2D. Results should be considered in the context of the observed heterogeneity.
- type 2 diabetes