Dairy product intake is inversely associated with the risk of type 2 diabetes (T2D) in numerous cohort studies; yet, the beneficial effects of increased dairy product intake on T2D risk factors such as fasting plasma glucose, fasting insulin, insulin resistance with the homeostasis model assessment, and glycated hemoglobin (HbA1c) remain inconclusive in clinical trials. The objective of this study was to systematically review clinical trials observing the effects of elevated compared with minimal intake of dairy products on T2D risk factors in subjects without diabetes. Five databases [Medline, EMBASE, Central, CINAHL, AMED (Allied and Complementary Medicine)] were searched to identify randomized controlled trials that used elevated quantities of dairy products from ruminant sources in comparison with a lower intake in control groups. The review outcomes were fasting blood glucose, fasting insulin, homeostasis model assessment of insulin resistance (HOMA-IR), and HbA1c. Risk of bias and quality of evidence according to Grading of Recommendations Assessment, Development, and Evaluation were addressed. From the 10,627 citations screened, 44 studies (3016 participants) were included, 38 of which were used in the meta-analyses. Fasting glucose was positively associated with elevated dairy intake [34 studies, n = 2678; mean difference (MD): 0.07 mmol/L; 95% CI: 0.01, 0.12 mmol/L; P = 0.01, I2 = 23%]. Fasting insulin (29 studies, n = 1902; MD: -2.97 pmol/L; 95% CI: -7.05, 1.10 pmol/L; P = 0.15, I2 = 21%) and HOMA-IR (13 studies, n = 840; standardized MD: -0.07; 95% CI: -0.26, 0.12; P = 0.49, I2 = 38%) were not associated with elevated dairy consumption. HbA1c was negatively associated with elevated dairy product intake in 4 studies (n = 512; MD: -0.09%; 95% CI: -0.09%, -0.03%; P = 0.005, I2 = 0%). Most studies had high risk of bias and the quality of evidence was very low or low. In conclusion, evidence suggests that elevated dairy product intake is associated with increased fasting plasma glucose concentrations together with reduced HbA1c in nondiabetic subjects. Hence, the clinical significance of these results remains uncertain. Additional well-designed, long-term studies are required.
