Previous studies have shown that the incidence of diabetes is higher when women come to menopause. This study was carried out to examine the effects of combined estrogen replacement therapy (ERT) on diabetes in postmenopausal women. PubMed/MEDLINE was searched for English-language articles published between January 1997 and June 2011. Studies that examined ERT on the incidence of diabetes and randomized clinical trials that evaluated combined ERT (estrogen plus progesterone) on diabetic indices in postmenopausal women were included. Pooled relative risks were calculated using a random- or a fixed-effects model. Sixteen studies comprising 17,971 cases were included. Based on the pooled data, ERT significantly reduced the incidence of diabetes [odds ratio (OR), 0.61; 95% confidence interval (CI), 0.55–0.68, ERT past/current/continuous use vs. never use; OR, 0.57; 95% CI, 0.51–0.65, ERT current/continuous use vs. past/never use]. Women with combined ERT have significantly lower levels of fasting plasma glucose (mean difference, –1.41 mM/L; 95% CI, –2.49 to −0.33 mM/L) and HbA1c (mean difference, –0.73%; 95% CI, from −1.28 to −0.18%) compared with placebo. Furthermore, combined ERT dramatically reduced plasma total cholesterol (mean difference, –0.34 mM/L; 95% CI, from −0.53 to −0.15 mM/L) and low-density lipoprotein (mean difference, –0.43 mM/L; 95% CI, from −0.71 to −0.14 mM/L) but slightly increased high-density lipoprotein (mean difference, 0.02 mM/L; 95% CI, from −0.07 to 0.12 mM/L) levels as compared with placebo control. This systemic review and meta-analysis provides evidence that postmenopausal women taking low-dose combined ERT have a decreased risk of developing diabetes and have better diabetic control.