Evaluating Dose Response in Flexible Dose Clinical Trials
Author Information
Author(s): Ilya Lipkovich, David H. Adams, Craig Mallinckrodt, Doug Faries, David Baron, John P. Houston
Primary Institution: Lilly Research Laboratories, Eli Lilly and Company
Hypothesis
Can the dose response relationship be accurately evaluated in flexible dose clinical trials using marginal structural models?
Conclusion
The study found that using marginal structural models with inverse probability of treatment weighting may help remove biases in evaluating dose effects in flexible dose trials.
Supporting Evidence
- The study analyzed data from two flexible-dose trials of olanzapine in patients with bipolar disorder and schizophrenia.
- Results indicated that naive analyses suggested a negative dose effect, while weighted analyses showed no strong dose effects.
- Using marginal structural models may provide a better understanding of dose response relationships in flexible dose trials.
Takeaway
This study looked at how changing doses of medication affects patients with bipolar disorder and schizophrenia, and found that the way we analyze the data can change what we think about how effective different doses are.
Methodology
The study used marginal structural models and inverse probability of treatment weighting to analyze data from two randomized, double-blind, flexible-dose trials.
Potential Biases
Potential selection bias due to non-random dose assignment and dropouts.
Limitations
The study was exploratory and had limited data, which may affect the robustness of the findings.
Participant Demographics
Participants included patients with bipolar mania (N=234) and schizophrenia (N=172), with varying ages and gender distributions.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website