Statistical Quality Scale 6 (SQS-6)

Abstract

ContextThe statistical analysis is an important part of the process of assessing the quality of randomized controlled trials, unfortunately it tends to be underestimated or even omitted by editors, in this sense, scales that cover this gap becomes necessary.\n\nObjectiveTo build a definition and a scale for assessing the quality of statistical analysis in randomized controlled trials.\n\nMethodsA content analysis on 16 biostatistics texts was considered in the building of the definition and of the scale. The indicators of quality were based on the description and presentation of results of articles.\n\nResultsWe identify 32 quality indicators grouping in six dimensions: randomization and management of lost; sample size computation; use of mean and median; statistical test; use and interpretation of confidence interval and P-values. The scale range from 0 to 6 and score greater than three identify article with appropriate statistical analysis.\n\nConclusionsThe scale presented three dimensions linked to quality of description and three linked to quality of used strategy.