2004
DOI: 10.1051/medsci/2004202231
|View full text |Cite
|
Sign up to set email alerts
|

Du bon usage des tests dans les essays cliniques

Abstract: 231Répondre à de multiples questions au cours d'un même essai clinique est un problème bien connu des chercheurs qui s'investissent dans la recherche clinique. Or, si multiplier les analyses est une tentation légitime, le risque d'aboutir dans ce contexte à des conclusions faussement positives est réel et affecte la crédibilité des résultats. Le contrôle de ces erreurs s'effectue par un ajustement sur l'erreur de type I [1] * . De nombreuses méthodes ont été développées dans cet objectif. Elles utilisent habit… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Year Published

2020
2020
2023
2023

Publication Types

Select...
4

Relationship

0
4

Authors

Journals

citations
Cited by 4 publications
(1 citation statement)
references
References 5 publications
0
1
0
Order By: Relevance
“…To demonstrate non-inferiority of the OptiMA or Com-PAS strategies with standard protocol, we assumed a noninferiority margin of 10%, the same as in other noninferiority trials assessing simplified protocol [11,13]. We reduced the test significant level (alpha) at 1.25 to consider the multiple testing [17], a power of 90% for the main objective and 80% for the 2 main secondary objectives were applied, and an inflation rate of 5% was used to account for unexploitable data. Calculations were performed using the SAS software (Proc Power TwoSample-Freq).…”
Section: Sample Sizementioning
confidence: 99%
“…To demonstrate non-inferiority of the OptiMA or Com-PAS strategies with standard protocol, we assumed a noninferiority margin of 10%, the same as in other noninferiority trials assessing simplified protocol [11,13]. We reduced the test significant level (alpha) at 1.25 to consider the multiple testing [17], a power of 90% for the main objective and 80% for the 2 main secondary objectives were applied, and an inflation rate of 5% was used to account for unexploitable data. Calculations were performed using the SAS software (Proc Power TwoSample-Freq).…”
Section: Sample Sizementioning
confidence: 99%