Grant: Travel Awards - Institute Community Support
McGill University
$1,000.00
Using Causal Diagrams to Visualize Solutions for the Built-In Selection Bias of the HR in the Context of Time-Varying Exposures: An example from asthma research
FY 2016-17
Summary
The Hazard Ratio (HR) is a relative measure of risk that takes time into account and is often reported in research studies evaluating the safety and/or effectiveness of drugs or other health interventions. However, the way the HR is computed can lead to misleading results, also called "bias", and this is why most researchers use caution when interpreting the HR in a causal context. In this presentation, we review the limitations of the HR when the risk factor of interest for a health outcome varies over time, and propose a new method to mitigate the inherent bias associated with the HR. We also demonstrate the usefulness of this method using an example in asthma research.
Department
Program
Travel Awards - Institute Community Support
Recipient
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
364575
Name
Longo Cristina M
Region
Québec
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Circulatory and Respiratory Health
Theme2
Social / Cultural / Environmental / Population Health
Prcname2
Institute of Circulatory and Respiratory Health
Projecttitle
Using Causal Diagrams to Visualize Solutions for the Built-In Selection Bias of the HR in the Context of Time-Varying Exposures: An example from asthma research
Programtype2
Undefined
Orgnameinp2
McGill University
Cihramount2
$1,000
Cihrequipment2
$0
Programname2
Travel Awards - Institute Community Support
Keyworddelim
Asthma; Causal Diagrams; Causal Inference; Ics Travel Award; Marginal Structural Models; Pharmacoepidemiology; Time-Varying Exposures
Abstract
The Hazard Ratio (HR) is a relative measure of risk that takes time into account and is often reported in research studies evaluating the safety and/or effectiveness of drugs or other health interventions. However, the way the HR is computed can lead to misleading results, also called "bias", and this is why most researchers use caution when interpreting the HR in a causal context. In this presentation, we review the limitations of the HR when the risk factor of interest for a health outcome varies over time, and propose a new method to mitigate the inherent bias associated with the HR. We also demonstrate the usefulness of this method using an example in asthma research.
Competitiondate
201701
Approvedterm2
1 yr 0 mth
Pinamesdelim
Longo, Cristina M
Approvedterm
1 an 0 mois
Cihrcontribution2
Inst of Circ &Respiratory Hlth~ $1,000~ $0~
