Grant: Doctoral Research Award: Canada Graduate Scholarships
University of Ottawa
$120,000.00
Methodological issues in the design and analysis of longitudinal cluster randomized trials
FY 2024-25
Summary
Cluster randomized trials (CRTs) randomize groups (clusters), rather than individuals to study arms. This study design introduces unique challenges, as outcome measurements from individuals within the same clusters tend to be correlated. In longitudinal CRTs, outcomes are measured repeatedly over time, introducing additional complexity as correlations also exist between responses from individuals within the same cluster in different periods. Additionally, in the stepped-wedge (SW) CRT, all clusters begin in the control condition then sequentially cross over to the intervention, introducing further complexity as the intervention effect is partially confounded by time. This project will consist of a series of three methodological papers aiming to highlight and address the challenges of longitudinal CRT designs. The first paper will be a systematic review of longitudinal CRTs published within the past five years to identify gaps in conduct and reporting. Details of analytic methods used and completeness of reporting will be extracted and summarized. The second paper will analyze real trial data of a longitudinal CRT being conducted in pediatric emergency departments across Canada to assess the effect of an innovative approach to acute mental healthcare on patient wellbeing. I will use a variety of analytic methods to showcase how different approaches can lead to different conclusions. The third paper will analyze multiple real SW-CRT datasets to demonstrate the consequences of failing to account for time-varying treatment effects. This project will identify areas for improvement in the design, analysis, and reporting of longitudinal CRTs and will demonstrate real-world implications of using inappropriate methods. It will ultimately inform the development of practical guidance to improve the quality of the design and analysis of longitudinal CRT designs.
Department
Program
Doctoral Research Award: Canada Graduate Scholarships
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
517978
Name
Shaw Julia F
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Population and Public Health
Theme2
Clinical
Prcname2
Doctoral Research Awards - B
Projecttitle
Methodological issues in the design and analysis of longitudinal cluster randomized trials
Programtype2
Studentship Programs
Orgnameinp2
University of Ottawa
Cihramount2
$120,000
Cihrequipment2
$0
Programname2
Doctoral Research Award: Canada Graduate Scholarships
Keyworddelim
Clinical Epidemiology; Clinical Trials; Cluster Randomized Trials; Methodological Review; Reporting; Stepped Wedge Trial; Trial Design
Abstract
Cluster randomized trials (CRTs) randomize groups (clusters), rather than individuals to study arms. This study design introduces unique challenges, as outcome measurements from individuals within the same clusters tend to be correlated. In longitudinal CRTs, outcomes are measured repeatedly over time, introducing additional complexity as correlations also exist between responses from individuals within the same cluster in different periods. Additionally, in the stepped-wedge (SW) CRT, all clusters begin in the control condition then sequentially cross over to the intervention, introducing further complexity as the intervention effect is partially confounded by time. This project will consist of a series of three methodological papers aiming to highlight and address the challenges of longitudinal CRT designs. The first paper will be a systematic review of longitudinal CRTs published within the past five years to identify gaps in conduct and reporting. Details of analytic methods used and completeness of reporting will be extracted and summarized. The second paper will analyze real trial data of a longitudinal CRT being conducted in pediatric emergency departments across Canada to assess the effect of an innovative approach to acute mental healthcare on patient wellbeing. I will use a variety of analytic methods to showcase how different approaches can lead to different conclusions. The third paper will analyze multiple real SW-CRT datasets to demonstrate the consequences of failing to account for time-varying treatment effects. This project will identify areas for improvement in the design, analysis, and reporting of longitudinal CRTs and will demonstrate real-world implications of using inappropriate methods. It will ultimately inform the development of practical guidance to improve the quality of the design and analysis of longitudinal CRT designs.
Competitiondate
202411
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Shaw, Julia F
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $120,000~ $0~