Alternative Statistical Methods for Estimating Efficacy of Interferon Beta-1b in MS Trials
Author Information
Author(s): Mieno Makiko N, Yamaguchi Takuhiro, Ohashi Yasuo
Primary Institution: Jichi Medical University
Hypothesis
This study aimed to investigate the performance of various regression models for estimating the treatment effect of interferon beta-1b in multiple sclerosis clinical trials.
Conclusion
The use of alternative models that include recurrence event data may provide better analyses for estimating treatment effects.
Supporting Evidence
- The hazard ratios of real data were estimated for each model including the effects of other covariates.
- The annual relapse rates in the high- and low-dose groups were 0.763 and 1.069, respectively.
- The percentage of patients who experienced a relapse more than once during follow-up was 55.8% in the high-dose group and 65.6% in the low-dose group.
Takeaway
Researchers looked at different ways to analyze data from multiple sclerosis trials to see which method gives the best results for treatment effects.
Methodology
A simulation study was conducted using various regression models, including extended Cox regression and Poisson regression with Generalized Estimating Equations, applied to real clinical trial data.
Potential Biases
The WLW model showed a tendency to overestimate treatment effects.
Limitations
The assumptions needed in the extended Cox regression models may not hold true for the data, and the precision of estimates in some models was relatively poor.
Participant Demographics
The study included 188 patients with relapsing-remitting MS, consisting of 55 males and 133 females.
Statistical Information
P-Value
0.006
Confidence Interval
95% CI 0.560-0.910
Statistical Significance
p=0.006
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website