Real-World vs. Randomized Trial Evidence for Incretin-Based Therapies: A Narrative Review

Incretin-based therapies have transformed the management of obesity and type 2 diabetes (T2D). Pivotal randomized controlled trials (RCTs) report substantial weight loss and glycemic improvements; the extent to which these translate to clinical populations remains uncertain. Seven 2026-published RCT meta-analyses (comprising 127 RCTs and 58 976 participants) were compared with 26 real-world studies identified through structured PubMed search. Primary outcomes were body weight loss and HbA1c reduction; secondary outcomes included major adverse cardiovascular events (MACE), neurological end-points, and safety. The comparison is descriptive. For T2D, real-world semaglutide achieved 4.7-10.5 kg weight loss (6-12 months) and 1.0-1.3 percentage-point HbA1c reductions, lower than RCT meta-analytic estimates (tirzepatide mean difference [MD] -9.55% vs. placebo). Without diabetes, real-world persister analyses (SHAPE study: semaglutide -14.1%, tirzepatide -16.5%) approximated per-protocol RCT estimates (semaglutide -14.9%, tirzepatide 15 mg -20.9%), whereas intention-to-treat real-world estimates were substantially lower. Just 13% of semaglutide users reached 2.4 mg and 25.9% of tirzepatide users reached 15 mg. Real-world cardiovascular data showed 20%-46% MACE reductions, broadly consistent with the RCT reduction. Differences between T2D and non-T2D populations likely reflect channeling bias and differential follow-up. Real-world incretin-based therapy effectiveness is broadly consistent with RCT efficacy when dose attainment is accounted for. High discontinuation rates (20%-50% within 1 year) and prior GLP-1 RA exposure track with the observed efficacy gap. Clinicians can expect RCT-level outcomes in patients who tolerate and persist with target doses. Head-to-head cardiovascular outcome trials comparing tirzepatide and semaglutide in non-diabetic populations are needed.