OBJECTIVES: Meta-analyses (MAs) are used to support evidence-based health care decision making, particularly when comparing multiple competing interventions. A recently published direct MA assessed the efficacy of once weekly (QW) glucagon-like peptide-1 receptor agonists (GLP-1RAs) using only placebo-controlled trials. The aim was to compare the direct MA to a network meta-analysis (NMA), which allowed for the assessment of evidence from both placebo- and active-controlled trials. METHODS: We conducted a random-effect Bayesian NMA of randomized controlled trials from 29 core GLP-1RA registration programs. The outcomes analyzed were weighted mean difference (WMD) in A1C from baseline and odds ratio (OR) for patients reaching an A1c target of <7% at six months. The direct MA included six albiglutide and seven dulaglutide trials for these outcomes. RESULTS: The WMD in A1C versus placebo for albiglutide was -0.66%, 95% CI [-1.14, -0.19], and -0.95%, 95% credible interval (CrI) [-1.21, -0.69], from the direct MA and NMA, respectively.  For dulaglutide, the WMDs were similar: -1.18%, 95% CI [-1.34, -1.02], from the direct MA, and -1.10%, 95% CrI [-1.25, -0.95], from the NMA.  The OR versus placebo for reaching target at six months were similar for albiglutide (direct MA 3.13, 95% CI [2.04,4.80]); NMA 3.85, 95% CrI [2.22, 6.73]), but not for dulaglutide (direct MA7.90, 95% CI [6.30,9.90]); NMA 5.46, 95% CrI [3.97, 7.56]). The direct MA did not provide results for exenatide QW due to study selection criteria. The WMD and OR for exenatide QW from the NMA was -1.04, 95% CrI [-1.22, -0.86] and 5.52, 95% CrI [3.83, 8.14] respectively. The NMA also provided comparisons between GLP-1RA agents. CONCLUSIONS: Meta-analyses can yield varying results based on the analytic approach and study selection. This NMA utilized both direct and indirect evidence from placebo- and active-controlled GLP-1RA trials, which provided a more comprehensive assessment of QW GLP-1RAs.