Grant: DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)
University of Toronto
$120,000.00
Addressing risk of bias in cohort studies of interventions
EF 2025-26
Résumé
Columbus's accidental discovery of America while aiming for India was praised as a success, yet under the lens of confirmatory research, his voyage would be considered a failure. For confirmatory studies evaluating healthcare interventions, researchers write a protocol before they begin. It prespecifies the study goals and analysis plans. If the final paper quietly alters the goals or the analysis to highlight more publishable findings, doctors get a biased picture. This is termed as selective reporting bias which, in the worst case, can promote ineffective or even harmful care.
To deter this bias, clinical trials must now have an outline posted in a public registry before launch, and many journals require protocols to be submitted alongside the final report of results. But for cohort studies of interventions-a key type of observational study that guides practice when trials are not available-these transparency practices are rarely used, and the scale of selective reporting remains unknown.
This PhD project addresses that gap in two steps. First, we will sample cohort studies of interventions conducted at Canadian and Danish institutions. By comparing their protocols archived by institutional research ethics boards with final results published in journals, we will evaluate changes in study goals and analyses, and assess whether these changes favour positive results. This work will provide evidence of selective reporting in cohort studies of interventions. Second, we will conduct an experiment where researchers assess the risk of bias for a cohort study using the final publication alone and then reassess with the protocol in hand. We will estimate how often the protocols helps with the assessment of bias.
The expected impact is to promote transparency of cohort studies of interventions. Making it clear when studies follow their original plan-and when and why they deviate-will improve evidence appraisal and ultimately support better-informed healthcare decisions.
Programme
DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
559654
Name
SONG ZEXING
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Health Services and Policy Research
Theme2
Clinical
Prcname2
DT2 (Health services and population health research) CGRS Doctoral
Projecttitle
Addressing risk of bias in cohort studies of interventions
Programtype2
Studentship Programs
Orgnameinp2
University of Toronto
Cihramount2
$120,000
Cihrequipment2
$0
Programname2
DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)
Keyworddelim
Clinical Cohort Study; Open Science; Reporting Bias; Research Integrity
Abstract
Columbus's accidental discovery of America while aiming for India was praised as a success, yet under the lens of confirmatory research, his voyage would be considered a failure. For confirmatory studies evaluating healthcare interventions, researchers write a protocol before they begin. It prespecifies the study goals and analysis plans. If the final paper quietly alters the goals or the analysis to highlight more publishable findings, doctors get a biased picture. This is termed as selective reporting bias which, in the worst case, can promote ineffective or even harmful care.
To deter this bias, clinical trials must now have an outline posted in a public registry before launch, and many journals require protocols to be submitted alongside the final report of results. But for cohort studies of interventions-a key type of observational study that guides practice when trials are not available-these transparency practices are rarely used, and the scale of selective reporting remains unknown.
This PhD project addresses that gap in two steps. First, we will sample cohort studies of interventions conducted at Canadian and Danish institutions. By comparing their protocols archived by institutional research ethics boards with final results published in journals, we will evaluate changes in study goals and analyses, and assess whether these changes favour positive results. This work will provide evidence of selective reporting in cohort studies of interventions. Second, we will conduct an experiment where researchers assess the risk of bias for a cohort study using the final publication alone and then reassess with the protocol in hand. We will estimate how often the protocols helps with the assessment of bias.
The expected impact is to promote transparency of cohort studies of interventions. Making it clear when studies follow their original plan-and when and why they deviate-will improve evidence appraisal and ultimately support better-informed healthcare decisions.
Competitiondate
202511
Approvedterm2
3 yrs 0 mth
Pinamesdelim
SONG, ZEXING
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $120,000~ $0~