Quality scores do not predict discrepant statistical significances among meta-analyses on different targets of
Anna M Passon1, Anna Drabik, Peter T Sawicki
1Institute of Health Economics and Clinical Epidemiology, The University Hospital of Cologne (AöR), Gleueler Straße 167-178, 50935 Cologne, Germany.
Objectives:
To examine differences in conclusions, statistical significances, and quality of systematic reviews on preventive effects of different targets of blood glucose lowering on macrovascular events in patients with type 2 diabetes.
Study Design And Setting:
We searched MEDLINE, the Cochrane Database of Systematic Reviews, and Embase until October 15, 2011. Systematic reviews using meta-analyses to compare macrovascular events in patients with type 2 diabetes randomized to different therapeutic targets of blood glucose were eligible for inclusion in this study. Effect measures were extracted and quality was assessed by Overview Quality Assessment Questionnaire, Assessment of Multiple Systematic Reviews, and Preferred Reporting Items for Systematic Reviews and Meta-Analyses. An additional checklist was used to control for completeness and suitability of included trials, correctness of data extraction, and choice of outcome parameters.
Results:
The 16 included systematic reviews were partially discrepant in statistical significances of reported outcome parameters. Twelve systematic reviews did not include available relevant trials and included eight systematic reviews that should not have been included according to the systematic review's aim and search strategies. Quality differed considerably among systematic reviews.
Conclusion:
Physicians, researchers, and policy makers should bear in mind that common quality assessment instruments are necessary but not sufficient to guarantee reliable results and conclusions obtained from meta-analyses.
Related Concept Videos
Diabetes Mellitus: Type 2 and Gestational
Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast, controlled...
Diabetes: Management and Pharmacotherapy
Insulin remains the cornerstone of treatment for most patients with type 1 and many...
Significance Testing: Overview
Accuracy and Errors in Hypothesis Testing
In hypothesis testing, the probability of making a Type I error, denoted as α, is commonly set at 0.05. This significance level indicates a 5% chance...
Type II Diabetes Mellitus III: Clinical Manifestations and Diagnosis

