The Impact of Incorporating Clinical Relevance on the Feasibility of Clinical Trials
- PDF / 9,220,773 Bytes
- 8 Pages / 612 x 792 pts (letter) Page_size
- 37 Downloads / 160 Views
99
The Impact of Incorporating Clinical Relevance on the Feasibility of Clinical Trials In the past, data jiom placebo-contrdled trials have been analyzed to demonstmte a statistical&significanttreatment difference. Howeva the most important dmwback of this classical appmch is that looking at the presence or absence of an efl& ignores the issue of fl& size (ie, diniud relevance). A statisticalbsign$cant the information that treatment effect gives the treutment differenceis not zero but does not provide information about the clinicalrelevance. llerefm, this pmctice has been criticized by regulatory authorities and the use of point estimates and corresponding confidence intervals
on&
Dietor Horrchke Institute of Medical Biometry and Medical Informatics. University Medical Center Freiburg. Freiburg. Germany Dietrich Hafnor ALLERGOPHARMA. loachim Ganzer KG. Reinbek. Germany
on&
Koy Words Clinical relevance; COPD; SGRQ; Statistical significance
Correspondence Address Dieter Hauschke. Institute of Medical Biometry and Medical Informatics, University Medical Center Freiburg. Stefan-MeierStrafie 26. 79104 Freiburg, Germany (email: diete,: [email protected]) .
IN T R O D U C T l O N In the International Conference on Harmonization (ICH) guideline Statistical Principlesfor Clinical Trials ( l ) ,the statistical analysis of placebocontrolled trials to show superiority was addressed. According to this guideline, efficacy of a new test treatment is established by demonstrating superiority over placebo by a statistically significant test result. However, the recent guideline on the Choice of Non-InferiorityMargin, published in 2005 by the Committee for Medicinal Products for Human Use (2) of the European Medicines Agency, points out that for demonstrating superiority not only statistical significance but also clinical relevance has to be shown. Clinical relevance of a statistically significant treatment effect can be assessed either on the original scale by using the point estimate and the associated confidence interval or by performing a responder analysis (2). The latter one classifies each patient to either achieve a predefined meaningful treatment effect or not. A probably more intuitive approach to judge the clinical relevance is by considering the strength of the effects on the original scale. In this article, we focus on the assessment of clinical significance by point estimates and corresponding
has been recommended assessing additionally whether the observed treatment ef& is also of clinicalrelevance. In this article, we focus on the implications of differentcriteria of dt$ning clini d relevance, and we illustratethe consequences for dinical trials in patients with chronic obstructive pulmonary disease when the St. George’s Respimtory Questionnaire is used as a sylnptomaticclinical endpoint. In condusion, we demonstmtethat thepoint estimateto exceed an a prim’-defined threshold is an adequate method for incorporating clinical relevance required in addition to statisticalsignificance.
two-sided confid
Data Loading...
