The paper discusses the empirical assessment of publication bias in meta-analyses, focusing on its impact on the results and conclusions of systematic reviews. The study analyzed 48 meta-analyses from the Cochrane Database of Systematic Reviews that considered binary endpoints and contained 10 or more individual studies. Using the trim and fill method, the authors estimated that 54% of reviews had missing studies, with 48% showing significant effects. In four cases, statistical inferences regarding the intervention's effect changed after adjusting for publication bias. The study concludes that publication bias is common but often does not affect the conclusions, emphasizing the need for researchers to routinely check for and address potential non-random selection mechanisms. The authors also highlight the simplicity and effectiveness of the trim and fill method in detecting and adjusting for publication bias.The paper discusses the empirical assessment of publication bias in meta-analyses, focusing on its impact on the results and conclusions of systematic reviews. The study analyzed 48 meta-analyses from the Cochrane Database of Systematic Reviews that considered binary endpoints and contained 10 or more individual studies. Using the trim and fill method, the authors estimated that 54% of reviews had missing studies, with 48% showing significant effects. In four cases, statistical inferences regarding the intervention's effect changed after adjusting for publication bias. The study concludes that publication bias is common but often does not affect the conclusions, emphasizing the need for researchers to routinely check for and address potential non-random selection mechanisms. The authors also highlight the simplicity and effectiveness of the trim and fill method in detecting and adjusting for publication bias.