Estimating the Effect of Beta-Interferon Exposure in Delaying Disease Progression in Relapsing-Remitting Multiple Sclerosis Patients Using Causal Inference Tools

• Activity Number: 468
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Mohammad Karim*+ and Paul Gustafson and Albert John Petkau and Yinshan Zhao and Afsaneh Shirani and Elaine Kingwell and Charity Evans and Mia van der Kop and Joel Oger and Helen Tremlett
• Companies: University of British Columbia and University of British Columbia and University of British Columbia and University of British Columbia and University of Texas Southwestern Medical Center and University of British Columbia and University of Saskatchewan and Karolinska Institutet and University of British Columbia and University of British Columbia
• Keywords: causal inference ; observational study ; inverse probability weighting ; survival analysis ; multiple sclerosis ; longitudinal

Abstract

We investigated the association between beta-interferon (BIFN) treatment exposure and disease progression in relapsing-remitting multiple sclerosis patients in the 'real-world' settings using a Marginal Structural Cox Model (MSCM) approach. This was a retrospective study of prospectively collected longitudinal data (1995-2008) in British Columbia, Canada. The survival outcome measure was time from BIFN treatment eligibility to a confirmed and sustained Expanded Disability Status Scale (EDSS) 6. The MSCM allows adjustment for baseline characteristics as well as time-varying confounders through the use of inverse probability of treatment and censoring (IPTC) weighting. We constructed appropriate IPTC weights suitable for this chronic disease setting to fit the MSCM. Additionally, IPTC weight adjusted Kaplan-Meier survival curves were constructed to display the survival pattern. The MSCM results indicated that time-dependent exposure to BIFN was not associated with a statistically significant difference in the hazard of reaching EDSS 6 (HR 1.36; 95% CI 0.95-1.94). The findings were consistent based on sensitivity analyses conducted to check various MSCM assumptions.