Meta-analysis: the science of review in Neurology

• Published in: Revista de neurologiá Vol. 40; no. 4; pp. 222 - 228
• Main Authors: Allam, M F, Del Castillo, A S, Navajas, R F-C
• Format: Journal Article
• Language: Spanish
• Published: Spain 16.02.2005
• Subjects: Humans; Mathematics; Meta-Analysis as Topic; Neurology; Reproducibility of Results; Research Design; Review Literature as Topic
• ISSN: 0210-0010

The review and continuous analysis of the present knowledge have always been necessary for scientific and clinical practice advance. This study presents the difference between narrative and systematic reviews, the most important points of meta-analysis, and finishes with detailed description of its phases. Our objective was to explain, in simple terms, the technique of meta-analysis and this could permit its application and use in the clinical practice and neurological investigations. The term meta-analysis was described for the first time by the psychologist Glass in 1967. Since then, the meta-analysis was utilized by many investigators as a technique to combine the results of different studies. The steps of a meta-analysis are: 1. Hypothesis of the study; 2. Localization of the studies of investigation; 3. Selection of the localized studies; 4. Qualitative meta-analysis; 5. Quantitative meta-analysis. Basically, It refers to the numeric combination of data, which were extracted by the reviewers. The mathematical method which will be used to evaluate the effect size should be chosen, also homogeneity and sensibility tests should be done. Well-designed meta-analysis is accepted as the optimum form to present the results of different studies. The meta-analysis could be of great importance for clinical practice and neurological research.