Ortalama Farklar İçin Meta-Analizlerin Gücünün Değerlendirilmesi: Metodolojik Araştırma

Abstract

Objective: Meta-analysis is a method that summarizes the results of various independent studies on the same subject and is used to obtain more reliable and accurate results. An important issue that is a sign of quality in metaanalyses, which is at the top of the evidence pyramid, is power, and the power of a meta-analysis can be evaluated under the two model approaches: fixed effect or random effect, with the existence of heterogeneity. Despite this importance, power details are not included in many meta-analysis studies. Whatever the model, power is influenced by effect size, the smallest sample size among studies, level of heterogeneity and the number of articles included in the meta-analysis. This study aims to determine the minimum sample size for a given number of articles in order to reach an acceptable power for a meta-analysis. Material and Methods: The power values for different numbers of articles (2, 5, 7, 10, 15, 18) and minimum of sample sizes in articles were determined, taking into consideration different effect sizes (0.1-0.2, 0.3-0.4, 0.5-0.7) under fixed and random effect models of low, medium and high heterogeneity. Results: In order for a meta-analysis study with a higher heterogeneity and/or a lower effect size to reach an acceptable power, the minimum sample size in the included articles should be high. Conclusion: This study, which considers the factors affecting power, can provide guidance to clinicians on how to calculate power according to heterogeneity and what is necessary or sufficient to achieve an acceptable quality in their studies.