Patients with type 2 diabetes (T2DM) who are inadequately controlled with insulin therapy may be treated by adding a non-insulin anti-diabetic agent instead of intensifying insulin therapy. However, there have been no comparative efficacy trials among various non-insulin anti-diabetic agents. We recently performed a network meta-analysis to examine the comparative efficacy and safety among of dipeptidyl peptiase-4 inhibitor (DPP4i), glucagon like peptide-1 receptor agonists (GLP-1RA), sodium-glucose co-transporter 2 inhibitor (SGLT2i) and thiazolidinedione (TZD) as adjunctive treatment in patients with poorly controlled T2DM with pre-existing insulin therapy. Out of potentially relevant 8935 published articles and 1092 registered trials, 47 RCTs were included for the analysis. GLP-1RA showed the greatest HbA1c lowering effect compared to control group (-0.84%, 95% Credible interval, -0.99 to -0.69%), followed by SGLT2i (-0.69%, -0.87 to -0.52%), TZD (-0.61%, -0.86 to -0.37%) and DPP4i (-0.53%, -0.66 to -0.40%). For FPG lowering efficacy, SGLT2i was superior to any other treatment groups. GLP-1RA and SGLT2i showed greater body weight reduction, whereas TZD increased body weight. For insulin sparing effects, TZD was ranked as the highest. The risks of hypoglycemia were increased in TZD and GLP-1RA groups. Current study provides best available evidence on comparative efficacy and safety of non-insulin anti-diabetic agents on top of insulin therapy in patients inadequately controlled T2DM.