Egger's Test and Publication Bias Explained

Publication bias occurs when studies with statistically significant or "positive" results are more likely to be published than studies with null or negative findings — which means the published literature your systematic review searches can systematically overstate an effect.

The funnel plot

A funnel plot graphs each study's effect size against a measure of its precision (often standard error). In the absence of bias, studies should scatter roughly symmetrically around the pooled estimate, wider at the bottom (small, imprecise studies) and narrower at the top (large, precise studies) — forming a rough funnel shape. Asymmetry suggests something's missing, often small studies with null results.

Egger's test

Egger's test statistically checks funnel plot asymmetry by regressing standardized effect estimates against their precision — a significant result (typically p < 0.10, a more lenient threshold than usual) suggests meaningful asymmetry. It requires a reasonable number of studies (commonly cited minimum: 10) to have any real statistical power.

Limitations to know

• Asymmetry can arise from causes other than publication bias — genuine heterogeneity, small-study effects unrelated to publication, or chance.
• With fewer than 10 studies, the test is underpowered and results should be interpreted cautiously, if reported at all.
• A non-significant Egger's test doesn't prove the absence of bias — it just didn't detect it.

What to do if you find bias

Report it transparently rather than hiding it. Trim-and-fill methods can estimate an adjusted pooled effect accounting for likely missing studies, and searching grey literature (theses, conference abstracts, trial registries) proactively reduces the underlying problem rather than just detecting it after the fact.