{"title":"Selecting the best meta-analytic estimator for evidence-based practice: a simulation study.","authors":"S. Doi, L. Furuya-Kanamori","doi":"10.1097/XEB.0000000000000207","DOIUrl":null,"url":null,"abstract":"Studies included in meta-analysis can produce results that depart from the true population parameter of interest due to systematic and/or random errors. Synthesis of these results in meta-analysis aims to generate an estimate closer to the true population parameter by minimizing these errors across studies. The inverse variance heterogeneity (IVhet), quality effects and random effects models of meta-analysis all attempt to do this, but there remains controversy around the estimator that best achieves this goal of reducing error. In an attempt to answer this question, a simulation study was conducted to compare estimator performance. Five thousand iterations at 10 different levels of heterogeneity were run, with each iteration generating one meta-analysis. The results demonstrate that the IVhet and quality effects estimators, though biased, have the lowest mean squared error. These estimators also achieved a coverage probability at or above the nominal level (95%), whereas the coverage probability under the random effects estimator significantly declined (<80%) as heterogeneity increased despite a similar confidence interval width. Based on our findings, we would recommend the use of the IVhet and quality effects models and a discontinuation of traditional random effects models currently in use for meta-analysis.","PeriodicalId":55996,"journal":{"name":"International Journal of Evidence-Based Healthcare","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"33","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Evidence-Based Healthcare","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1097/XEB.0000000000000207","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Medicine","Score":null,"Total":0}
引用次数: 33
Abstract
Studies included in meta-analysis can produce results that depart from the true population parameter of interest due to systematic and/or random errors. Synthesis of these results in meta-analysis aims to generate an estimate closer to the true population parameter by minimizing these errors across studies. The inverse variance heterogeneity (IVhet), quality effects and random effects models of meta-analysis all attempt to do this, but there remains controversy around the estimator that best achieves this goal of reducing error. In an attempt to answer this question, a simulation study was conducted to compare estimator performance. Five thousand iterations at 10 different levels of heterogeneity were run, with each iteration generating one meta-analysis. The results demonstrate that the IVhet and quality effects estimators, though biased, have the lowest mean squared error. These estimators also achieved a coverage probability at or above the nominal level (95%), whereas the coverage probability under the random effects estimator significantly declined (<80%) as heterogeneity increased despite a similar confidence interval width. Based on our findings, we would recommend the use of the IVhet and quality effects models and a discontinuation of traditional random effects models currently in use for meta-analysis.
期刊介绍:
The International Journal of Evidence-Based Healthcare is the official journal of the Joanna Briggs Institute. It is a fully refereed journal that publishes manuscripts relating to evidence-based medicine and evidence-based practice. It publishes papers containing reliable evidence to assist health professionals in their evaluation and decision-making, and to inform health professionals, students and researchers of outcomes, debates and developments in evidence-based medicine and healthcare.
The journal provides a unique home for publication of systematic reviews (quantitative, qualitative, mixed methods, economic, scoping and prevalence) and implementation projects including the synthesis, transfer and utilisation of evidence in clinical practice. Original scholarly work relating to the synthesis (translation science), transfer (distribution) and utilization (implementation science and evaluation) of evidence to inform multidisciplinary healthcare practice is considered for publication. The journal also publishes original scholarly commentary pieces relating to the generation and synthesis of evidence for practice and quality improvement, the use and evaluation of evidence in practice, and the process of conducting systematic reviews (methodology) which covers quantitative, qualitative, mixed methods, economic, scoping and prevalence methods. In addition, the journal’s content includes implementation projects including the transfer and utilisation of evidence in clinical practice as well as providing a forum for the debate of issues surrounding evidence-based healthcare.