Bayesian Estimation Improves Prediction of Outcomes after Epilepsy Surgery

Low power is a problem in many fields, as underpowered studies that find a statistically significant result will exaggerate the magnitude of the observed effect size. We quantified the statistical power and magnitude error of studies of epilepsy surgery outcomes. The median power across all studies was 14%. Studies with a median sample size or less (n<=56) and a statistically significant result exaggerated the true effect size by a factor of 5.4 (median odds ratio 9.3 vs. median true odds ratio 1.7), while the Bayesian estimate of the odds ratio only exaggerated the true effect size by a factor of 1.6 (2.7 vs. 1.7). We conclude that Bayesian estimation of odds ratio attenuates the exaggeration of significant effect sizes in underpowered studies. This approach could help improve patient counseling about the chance of seizure freedom after epilepsy surgery. SHORT SUMMARYWe estimated the statistical power of studies predicting seizure freedom after epilepsy surgery. We exacted data from a Cochrane meta-analysis. The median power across all studies was 14%. Studies with a median sample size or less (n<=56) and a statistically significant result exaggerated the true effect size by a factor of 5.4, while the Bayesian estimate of the odds ratio only exaggerated the true effect size by a factor of 1.6. We conclude that Bayesian estimation of odds ratios attenuates the exaggeration of significant effect sizes in underpowered studies. This result could improve patient counseling regarding epilepsy surgery.