Related Experiment Video
Updated: May 29, 2025
Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Evaluating the Role of High-Dimensional Proxy Data in Confounding Adjustment in Multiple Sclerosis Research: A Case
Mohammad Ehsanul Karim1,2, Md Belal Hossain1,2, Huah Shin Ng3,4
1School of Population and Public Health, University of British Columbia, Vancouver, British Columbia, Canada.
High-dimensional data in multiple sclerosis (MS) research minimally altered disease-modifying drug (DMD) effect estimates on mortality. This suggests residual confounding in MS studies may be modest, warranting further investigation with diverse datasets.
Area of Science:
- Epidemiology
- Health Services Research
- Biostatistics
Background:
- Multiple sclerosis (MS) research often uses limited confounders in administrative health data analyses.
- High-dimensional data offers potential for more robust confounder adjustment.
Purpose of the Study:
- To evaluate the impact of incorporating high-dimensional proxy information on confounder adjustment in MS research.
- To assess changes in effect estimates for disease-modifying drugs (DMDs) and all-cause mortality using high-dimensional propensity score (hdPS) and high-dimensional disease risk score (hdDRS) methods.
Main Methods:
- Population-based retrospective study using linked administrative databases in British Columbia (BC), Canada.
- Cohort of 19,360 individuals with MS, followed from 1996 to 2017.
- Compared Cox proportional hazards models with investigator-specified covariates versus those including empirical covariates via hdPS and hdDRS.
Main Results:
- Unadjusted analysis showed DMDs associated with 69% lower mortality (HR 0.31).
- Adjusting for specified covariates yielded aHR 0.76.
- hdPS and hdDRS methods resulted in minor variations (aHRs 0.77-0.81), indicating 19%-23% lower mortality risk.
Conclusions:
- High-dimensional proxy information resulted in minor variations in effect estimates compared to traditional covariate adjustment.
- The impact of residual confounding in this MS study appears modest.
- Further research should explore additional data dimensions and replicate findings across diverse datasets.
More Related Videos
05:44Author Spotlight: Creating a Versatile Experimental Autoimmune Encephalomyelitis Model Relevant for Both Male and Female Mice
Published on: October 13, 2023
10:46A Method of Trigonometric Modelling of Seasonal Variation Demonstrated with Multiple Sclerosis Relapse Data
Published on: December 9, 2015
Related Concept Videos
01:25Strategies for Assessing and Addressing Confounding
Confounding can be addressed at both the design phase of a study and through analytical methods after data...
01:27Confounding in Epidemiological Studies
01:11Blinding
01:23Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...