Over 200 publications, including the composite endpoints regulators now recognize. We publish our methodology in the open so reviewers meet it before your submission does.
Alzheimer's Association International Conference (AAIC) · 2024:Poster Tue-17 · Hendrix, Dickson et al.
Companies recently identified clinical study sites engaging in fraudulent practices, reducing success in Alzheimer's disease (AD) clinical trials. Successful trials are reliable…
When the sample size a conventional design demands is larger than the trial you can run, a Global Statistical Test lets the clinical-endpoint hypotheses be tested anyway. The paper sets out when GST applies, how it is pre-specified, and how it has been received in review.
1 publication
Sorted newest first
1996 · Dissertation
Abstract: Many clinical trials are designed to compare treatment groups after a pre-determined length of time on treatment in order to allow the full treatment effect to be observed. If the scheduled duration of treatment is more than a few days, patients may withdraw from the study and appropriate handling of these patients may become critical to the proper interpretation of the efficacy results. We consider a protocol compliant (PC) or evaluable patients analysis, including only study completers, and an intent-to-treat endpoint analysis (ITT) using the last available value carried forward. We compare these two methods to an exposure adjusted (EA) method that includes all patients’ endpoint values in an analysis of covariance adjusting for varying exposure to study medication. These three methods are compared in terms of bias and efficiency under various dropout mechanisms including those unrelated to both the treatment group and the patient’s latent final response, as well as mechanisms related to these variables. We use a linear relationship between endpoint response and exposure to study medication. We show that under certain types of dropout, the ITT approach using the last available value carried forward is overly conservative and even biased at times. The PC analysis and the EA analysis preserve the type I error in most cases, and increase the power of the analysis over the ITT approach. We discuss the patterns of dropout that present the most difficulty, and indicate the best analysis for these situations.
Loading more publications…
Looking for older work, or a specific methodology paper? Our research team can point you to the right reference.
Ask our team
