There is a large amount of texts about this topic and it is hard to know where to begin. Are there any that are of particular quality or highly recommended?
Recommended reference texts:
ClinicalTrials: Design, Conduct and Analysis, Second Edition. Curtis L. Meinert
Calculating sample size for a minimum effect size of interest
Analysis strategies to maximise study power
Common pitfalls regarding pre-specification; particularly the choice of primary, secondary endpoints, covariates and statistical modeling, and how pre-specification is reported and made available prior to study initiation
Modern implementations of randomisation procedure and common pitfalls
Testing the of the quality of blinding and how this is reported
Allocation concealment, how the robustness of concealment should be monitored, and how it should be reported
Logistics of conducting a hospital based clinical RCT; best-practices regarding patient flow through evaluation for inclusion and exclusion criteria, randomisation and monitoring
Reporting and publishing the results of the RCT
Any and all suggestions are appreciated and I would be more than happy to monitor the responses, and edit this post to add the suggestions under the relevant subtopics, with links.
How to deal with what happens between randomization and evaluation of the endpoint. In the language of the ICH E9 addendum think about potential “intercurrent events” and which strategy to use to handle them.
I am finally in a position to start building infrastructure at my institution for conducting randomised controlled trials. We have decided to start building competence at my institution by participating in a multicenter multinational pragmatic RCT, but our (my) end-goal is designing and running trials myself. I wanted to revive this topic with a slightly different emphasis: can anyone point me to articles or textbooks that discuss the workflows and electronic components that are needed to conduct a good embedded pragmatic RCT? Are there any trialists here on datamethods?
some aro’s put their standard operating procedures on their website. Those would tell you a lot, they will point to the infrastructure, how data are handled, archived, systems used etc. The first thing you will need to do is develop standard operating procedures. If you have a good statistician he/she could help a lot because they would have seen sops in other places and it is the data person who is in the middle of it all. Frank Harrell mentioned here, or somewhere else, redcap. He praised it. Are you familiar with it?