OBJECTIVES: According to the literature, there is a gap between the results of clinical trials (efficacy) and the effects of the same intervention in real-life (effectiveness). Although this so-called “efficacy-effectiveness gap” is often mentioned in literature, there is little if any attempt to quantify it. We aim to quantify the difference between the efficacy and effectiveness of Metformin in terms of reducing HbA1c in patients with type-2-diabetes. METHODS: First, a systematic review was carried out to identify relevant randomized controlled trials (RCTs) and non-interventional studies (NIS). CENTRAL, MEDLINE (via PubMed) and clinicaltrials.gov were searched for relevant articles published within the last 20 years. RCTs and NIS which evaluated the treatment effects of Metformin in adult patients with type-2-diabetes and analyzed glycaemic control by means of change in HbA1c value were included. Studies in which Metformin was used as a concomitant antidiabetic, studies with a study period of less than three months and studies analyzing subpopulations were excluded from further research. Only German and English papers were eligible. Second, mean values of HbA1c reduction were aggregated in a meta-analysis using a random-effects model. Study heterogeneity was assessed by the I-parameter. To test for publication bias we used funnel plots. RESULTS: In total 1151 articles were identified. 21 RCTS and 6 NIS were included in the quantitative analysis. Overall HbA1c was reduced by 1.012% (95%-CI: -1.186%; -0.839%). Iwas 99.911% (RCTs: 99.932; NIS: 97.769) indicating high heterogeneity. The comparison of the two settings resulted in a small difference of 0.130% in HbA1c decrease between RCTs (-0.953%;  95%CI: -1.188%, -0.717%) and NIS (-1.083%; 95% CI: -1.34%, -0.826%). CONCLUSIONS: The results of our analysis indicate that the often-mentioned gap between clinical studies and real-world application does not exist in the treatment of type-2-diabetes with Metformin.