Skip to Main content Skip to Navigation
Master Thesis

Méthodes d’analyse statistique des schémas d’étude en stepped wedge : Application au projet OREANE

Abstract : Stepped wedge trials are a type of Cluster Randomised Trials (CRT). In stepped wedge trials, the clusters are sequentially randomised until the point at which all clusters are exposed to the intervention. Most of these trials are grouped within two types : cross-sectional and cohorts (opened or closed). The differences between these two types imply different levels of correlation which need to be taken into account in the statistical analysis. The aim of the OREANE study was to determine the effect of a complex training on medication errors in nursing in Neonatal Intensive Care Units. Different statistical methods have been explored in order to find the most adequat model for this cohort stepped wedge study, according to its specificities. The use of Generalized Linear Mixed Model that take into account the different levels of correlation (at the cluster and the individual levels) has shown that the efficacy of the intervention was dependent on the hospital center after adjusting on several covariates. However, the low number of clusters and their strong heterogeneity couldn’t lead to a global conclusion on the effect of the intervention.
Document type :
Master Thesis
Complete list of metadata

https://dumas.ccsd.cnrs.fr/dumas-03377937
Contributor : Isped Ub - Connect in order to contact the contributor
Submitted on : Thursday, October 14, 2021 - 1:18:32 PM
Last modification on : Friday, October 15, 2021 - 3:17:09 AM

File

M2BIOSTAT_2021_QUINTIN.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : dumas-03377937, version 1

Collections

Citation

Olivier Quintin. Méthodes d’analyse statistique des schémas d’étude en stepped wedge : Application au projet OREANE. Santé publique et épidémiologie. 2021. ⟨dumas-03377937⟩

Share

Metrics

Record views

32

Files downloads

24