Article type
Abstract
Background: Newcastle-Ottawa Scale (NOS) for assessing the quality of non-randomised studies in meta-analysis is known as one of many useful tools being implemented in the field. However, the ‘star system’ for judgement on three broad perspectives including selection, comparability and outcome need further improvements to be more successfully incorporated with interpretation of meta-analysis results.
Objectives: To develop a summary presentation tool based on judgements on each items in Newcastle-Ottawa Quality Assessment Scale using RevMan5.3 software.
Methods: To develop a summary presentation tool for the methodological quality in non-randomised cohort studies, 4-staged process is executed. First, the quality assessment for 7 cohort studies under a systematic review is performed using NOS Scale. Second, there is a reform process for each item of risk of bias tables using properties in the characteristics of included studies using RevMan5.3 software. Third, after grading judgements and assigning colors, a summary presentation tool for non-randomised cohort studies is developed. Fourth, a reformed risk of bias items is then applied to the judgement.
Results: The judgement of 9 items of NOS Scale for cohort studies was presented with forest plot to be incorporated with meta-analysis results. Especially, it was easy to identify ‘was follow-up long enough for outcomes to occur’ in a sub-group analysis by follow-up years.
Conclusions: A reformed risk-of-bias summary presentation tool for non-randomised cohort studies is a simple and user-friendly tool that is well-incorporated with the results of meta-analysis.
Objectives: To develop a summary presentation tool based on judgements on each items in Newcastle-Ottawa Quality Assessment Scale using RevMan5.3 software.
Methods: To develop a summary presentation tool for the methodological quality in non-randomised cohort studies, 4-staged process is executed. First, the quality assessment for 7 cohort studies under a systematic review is performed using NOS Scale. Second, there is a reform process for each item of risk of bias tables using properties in the characteristics of included studies using RevMan5.3 software. Third, after grading judgements and assigning colors, a summary presentation tool for non-randomised cohort studies is developed. Fourth, a reformed risk of bias items is then applied to the judgement.
Results: The judgement of 9 items of NOS Scale for cohort studies was presented with forest plot to be incorporated with meta-analysis results. Especially, it was easy to identify ‘was follow-up long enough for outcomes to occur’ in a sub-group analysis by follow-up years.
Conclusions: A reformed risk-of-bias summary presentation tool for non-randomised cohort studies is a simple and user-friendly tool that is well-incorporated with the results of meta-analysis.