Grant: Master's Award: Canada Graduate Scholarships
University of British Columbia
$17,500.00
Methodological concerns of stepped wedge randomized controlled trials with unequal cluster sizes
FY 2019-20
Summary
"In an effort to bridge the gap between experimental findings and practice, a special type of cluster randomized trial (CRT), the stepped-wedge trial, has gained increasing popularity over the recent years. Such trial is characterized by the sequential enrolment of clusters in intervention, where all clusters start with the control or standard treatment and at regular steps in time, clusters are randomly selected to cross over to intervention, until all clusters are enrolled in intervention. As such, stepped-wedge trials not only facilitate the implementation of practices but also offer a rigorous evaluation of their effects in the real-world. However, the design and analysis of stepped-wedge trials are complicated by additional considerations such as the confounding effect of time and the implications of cluster sizes imbalance when some clusters enroll more patients than others. While few studies have explored the consequences of cluster size variation in the context of stepped-wedge trials, several studies have compared different analysis methods for parallel-arm CRTs with small sample sizes. However, it is unknown whether the same conclusions apply to stepped-wedge trials, since they yield fundamentally different types of information on the intervention effect than conventional CRTs. This project seeks to compare the performance in terms of Type I error, power and bias of various analysis methods for data arising from stepped-wedge trials with few clusters, unequal cluster sizes and a continuous outcome, using analytical approaches and computer simulations."
Department
Program
Master's Award: Canada Graduate Scholarships
Recipient
Location
British Columbia, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
429097
Name
Xu Liang
Region
British Columbia
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Methodological concerns of stepped wedge randomized controlled trials with unequal cluster sizes
Programtype2
Studentship Programs
Orgnameinp2
University of British Columbia
Cihramount2
$17,500
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Cluster Randomized Trial; Methodology; Power; Small Sample Size; Stepped-Wedge Trial; Trial Analysis; Trial Design; Type I Error; Unequal Cluster Sizes
Abstract
"In an effort to bridge the gap between experimental findings and practice, a special type of cluster randomized trial (CRT), the stepped-wedge trial, has gained increasing popularity over the recent years. Such trial is characterized by the sequential enrolment of clusters in intervention, where all clusters start with the control or standard treatment and at regular steps in time, clusters are randomly selected to cross over to intervention, until all clusters are enrolled in intervention. As such, stepped-wedge trials not only facilitate the implementation of practices but also offer a rigorous evaluation of their effects in the real-world. However, the design and analysis of stepped-wedge trials are complicated by additional considerations such as the confounding effect of time and the implications of cluster sizes imbalance when some clusters enroll more patients than others. While few studies have explored the consequences of cluster size variation in the context of stepped-wedge trials, several studies have compared different analysis methods for parallel-arm CRTs with small sample sizes. However, it is unknown whether the same conclusions apply to stepped-wedge trials, since they yield fundamentally different types of information on the intervention effect than conventional CRTs. This project seeks to compare the performance in terms of Type I error, power and bias of various analysis methods for data arising from stepped-wedge trials with few clusters, unequal cluster sizes and a continuous outcome, using analytical approaches and computer simulations."
Competitiondate
201912
Approvedterm2
1 yr 0 mth
Pinamesdelim
Xu, Liang
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $17,500~ $0~