Research brief · 2026-09-09
What Is Publication Bias, and How Would You Spot It?
A meta-analysis pools every study its authors can find and combines them into one estimate. That estimate is only as trustworthy as the studies feeding it, and studies do not reach publication at random. A trial that finds a drug works is more attractive to journals and more likely to be written up quickly. None of this requires dishonesty. It is a structural tilt, and it means the visible literature is a filtered sample of everything that was tried.
The standard check is the funnel plot: each study's effect size on one axis, its precision on the other. Large, precise studies should cluster near the pooled estimate, and small ones should scatter in both directions, forming an inverted funnel. If small unfavorable studies are missing, the funnel is lopsided. Because that is a visual judgment, researchers pair it with Egger's test, a regression that checks whether the asymmetry exceeds chance. A significant result points to small-study effects, of which publication bias is one plausible cause.
A 2025 meta-analysis in Cureus by Kolli and colleagues examined weight regain after stopping four prescription anti-obesity drugs, semaglutide, liraglutide, exenatide and orlistat, pooling 36 studies. For liraglutide, Egger's test was significant at p equals 0.013. For the other three, no major bias was found. A 2026 meta-analysis in Diabetes, Obesity and Metabolism by Eisa and Barood pooled 20 randomized trials and 15,782 participants on lean mass change with incretin therapy, and its Egger's test was non-significant at p equals 0.23. A 2023 network meta-analysis in EClinicalMedicine likewise reported low publication bias.
Both outcomes are normal. The point of the test is not to always find bias but to look, and a meta-analysis that never mentions a funnel plot or Egger's test left the question unasked or unreported. What the test cannot do is name the missing studies or explain why liraglutide's evidence showed asymmetry, and it performs poorly with fewer than about ten studies. Decisions about these prescription medications belong with a physician. Sources are listed below.
Sources
- Kolli RT et al., Rebound or retention: meta-analysis of weight regain after discontinuation of GLP-1 receptor agonists and other anti-obesity drugs (36 studies)
- Cureus, 2025
- Eisa N, Barood O, Lean mass changes with incretin therapy versus lifestyle intervention: systematic review and meta-analysis of RCTs (20 trials, n=15,782)
- Diabetes, Obesity and Metabolism, 2026
- Caruso I et al., Glucometabolic outcomes of GLP-1 receptor agonist-based therapies in type 2 diabetes: network meta-analysis (40 trials)
- EClinicalMedicine, 2023
Peptide Proof reports what studies found. We do not give medical advice, dosing guidance, or treatment recommendations.