The authors of this article argue that prediction intervals should be routinely reported in meta-analyses to better understand the uncertainty of intervention effects. They highlight that while measures like τ² or I² are used to quantify heterogeneity, their clinical interpretation is unclear. In contrast, prediction intervals provide a more straightforward way to estimate the range of true effects in similar studies. The authors analyzed 479 statistically significant meta-analyses from the Cochrane Database of Systematic Reviews (2009–2013) and found that in 72.4% of these, the 95% prediction interval suggested the intervention effect could be null or even in the opposite direction. In 20.3% of these, the prediction interval showed the effect could be completely opposite to the point estimate. The prediction interval also helps calculate the probability of a new trial showing a negative effect and improve the power calculations of a new trial. The authors emphasize that prediction intervals reflect the variation in treatment effects across different settings, including what effect is expected in future patients. They argue that prediction intervals should be routinely reported to allow more informative inferences in meta-analyses. However, they note that the calculations are based on the normality assumption, which may not always hold. Additionally, prediction intervals may be imprecise if the estimates of the summary effect and heterogeneity are imprecise. The authors also mention that the prediction interval is only valid for settings similar to those in the meta-analysis. The authors provide an example of a meta-analysis on topical steroids for nasal polyps, where the prediction interval suggested that the true effect could be null or even in the opposite direction. They also discuss the limitations of using prediction intervals, including the assumption of normality and the potential for imprecision. The authors conclude that prediction intervals should be routinely reported in meta-analyses to improve clinical decision-making.The authors of this article argue that prediction intervals should be routinely reported in meta-analyses to better understand the uncertainty of intervention effects. They highlight that while measures like τ² or I² are used to quantify heterogeneity, their clinical interpretation is unclear. In contrast, prediction intervals provide a more straightforward way to estimate the range of true effects in similar studies. The authors analyzed 479 statistically significant meta-analyses from the Cochrane Database of Systematic Reviews (2009–2013) and found that in 72.4% of these, the 95% prediction interval suggested the intervention effect could be null or even in the opposite direction. In 20.3% of these, the prediction interval showed the effect could be completely opposite to the point estimate. The prediction interval also helps calculate the probability of a new trial showing a negative effect and improve the power calculations of a new trial. The authors emphasize that prediction intervals reflect the variation in treatment effects across different settings, including what effect is expected in future patients. They argue that prediction intervals should be routinely reported to allow more informative inferences in meta-analyses. However, they note that the calculations are based on the normality assumption, which may not always hold. Additionally, prediction intervals may be imprecise if the estimates of the summary effect and heterogeneity are imprecise. The authors also mention that the prediction interval is only valid for settings similar to those in the meta-analysis. The authors provide an example of a meta-analysis on topical steroids for nasal polyps, where the prediction interval suggested that the true effect could be null or even in the opposite direction. They also discuss the limitations of using prediction intervals, including the assumption of normality and the potential for imprecision. The authors conclude that prediction intervals should be routinely reported in meta-analyses to improve clinical decision-making.