Slide 1: Do you know…? What is publication bias?

Slide 2: Publication Bias. The failure to publish results of studies on the basis of the direction or strength of the study findings. How can we assess whether publication bias is at play in a body of literature?

Slide 3: Funnel Plots. A simple graphical test for publication bias. Larger trials with narrow confidence intervals (higher precision) are more likely to be closer to the actual effect size. Smaller trials with wider confidence intervals are more heterogenous due to random error.

A body of literature without publication bias should have smaller trials evenly distributed at the base of the triangle and larger trials close together at the point.

Graphic of funnel plot. Study precision (sample size) on the y axis. Study result (effect size) on the x axis.

Slide 4: A real life example. This is a meta-analysis of clinical trials of aspirin for prevention of pre-eclampsia. Despite many small trials showing a benefit to aspirin, the largest one (CLASP) failed to show a benefit, suggesting the presence of publication bias in the literature. These trials were never published!