Methodological improvements are needed in network meta analyses of antidiabetic drugs for type 2 diabetes mellitus

Numerous network meta-analyses have compared the efficacy of antidiabetic drugs for type 2 diabetes mellitus. However, a systematic assessment of their methodological and reporting quality remains lacking.

We searched the PubMed, Embase, and Cochrane Library databases for relevant studies from inception to July 21, 2025 for published network meta-analyses that compares the efficacy of antidiabetic drugs for type 2 diabetes mellitus. The methodological and reporting quality of included studies was assessed by the Assessment of Multiple Systematic Reviews 2 checklist and the PRISMA-NMA checklist. Regression analyses were performed to identify factors potentially associated with the methodological quality of the included studies.

A total of 185 NMAs were included. Over half of the studies (58.38%) originated from developing countries. Most studies focused on comparison of multiple antidiabetic drugs (122, 65.95%), and the majority used Bayesian methods (98, 52.97%). The overall methodological quality was low, with only one study rated as high quality. Critical items of AMSTAR 2 were poorly reported, with only 2.70% of studies adequately reporting item 7 and 58.38% reporting item 2. Significant associations were observed between study quality and analytical framework, outcome, economic development level, journal impact factor, and protocol registration status.

Network meta-analyses that compares antidiabetic drugs for T2DM exhibit substantial methodological limitations and advancements in methodology are critically needed in future research.