Clinical studies in restorative dentistry: Design, conduct, analysis

Publication date: January 2018
Source:Dental Materials, Volume 34, Issue 1
Author(s): Falk Schwendicke, Niek Opdam
ObjectiveClinical studies should be one main aspect underlying dentists' decision-making towards dental materials. Study design, conduct, analysis and reporting impact on the usefulness of studies. We discuss problems with current studies and highlight areas where improvement might be possible.MethodsBased on systematically and non-systematically collected data, we demonstrate where and why current studies in clinical dentistry deliver less-than-optimal results. Lending from general medicine, we suggest ways forward for clinical dental material science.ResultsRandomized controlled (efficacy) trials remain a major pillar in dental material science, as they reduce selection bias and, if well-designed and conducted, have high internal validity. Given their costs and limited external validity, alternatives like practice-based or pragmatic controlled trials or observational studies can complement the evidence-base. Prior to conduct, researchers should focus on study comparators and setting (answering questions with relevance to clinical dentistry), and pay attention to statistical power, considering the study aim (superiority or non-inferiority trial), the expected event rate, and attrition. Study outcomes should be chosen on the basis of a core outcome set or, if not available, involving patients and other stakeholders. Studies should be registered a priori, and reporting should adhere to standards. Possible clustering should be accounted for during statistical analysis.SignificanceMany clinical studies in dental material science are underpowered, and of limited validity and usefulness for daily decision-making. Dental researchers should mirror existing efforts in other medical fields in making clinical studies more valid and applicable, thus contributing to better dental care.



from #ORL-AlexandrosSfakianakis via ola Kala on Inoreader http://ift.tt/2ByNumC