I have not been involved in early phase / feasibility trials, but have been asked to give a five minute presentation on the statistical aspects of them. Any pointers for what statistical aspects are critical in this area?
In my opinion, You should discuss the sample size in clinical trials, as it is crucial for the rest of analysis.
I would say try as much as possible to clearly define both the metrics or measurable outcomes you might possibly get, as well as events that would be reasons for early stopping (adverse events of various sorts).
Thanks both... I have taken those both into consideration. I have also been looking at effect sizes, magnitude of efficacy, blinding strategies, case studies, discovery-oriented single-subject designs, and small group cohort control studies.
So, if You wish to discuss so many, find a place to discuss proper methods of randomization to avoid selection bias.
Aye.. that is true Alexsander. If I decide to focus on just one, I agree with you that sample size or method of randomisation would be the way to go. Its either that, or paint a very brief outline which includes a wider range of topics.
I of course agree. It should be just some flash watchwords for further widening of statisitcal issues.
Ensure the primary outcome measure addresses the research hypothesis; keep the number of predictor variables to a minimum; identify potential confounding variables; adequate sample size; randomization.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics