Publication bias — MRCPsych Paper A MCQ

Correct answer: D — Publication bias

The correct answer is **B, publication bias**. The key discriminator is that completed small trials with non-significant or unfavourable findings were not published, leaving published small positive studies over-represented and producing funnel-plot asymmetry. Funnel asymmetry alone is not proof of publication bias because clinical or methodological heterogeneity can also produce small-study effects; the trial-registry findings make publication bias the best explanation here. Selective outcome reporting involves omission or selective presentation of outcomes within a published study rather than non-publication of entire trials. Language bias results from restricting inclusion by publication language, while citation bias occurs when studies with favourable findings are more likely to be cited and therefore located. Clinical heterogeneity could cause genuine differences in effects between smaller and larger trials but does not account for confirmed non-publication.

Reference: Guyatt G, Wang Y, Eachempati P, et al. Core GRADE 4: rating certainty of evidence—risk of bias, publication bias, and reasons for rating up certainty. BMJ. 2025;389:e083864. https://www.bmj.com/content/389/bmj-2024-083864