This is a very interesting proposal, but more information is needed.
First, the Scientific Abstract suggests that the development and application of diagnostics for informative censoring will start with a short-term trial, but then be further tested in a long-term trial. However, the Methods text of the proposal discusses only the work to be done with the short-term trial (NCT00397033). Please clarify and provide additional information as needed.
Second, I found the description of the primary outcome measure to be confusing. How will it be dichotomously categorized? As a “positive” or “negative” mean difference? What’s the threshold for the mean difference to be “positive/negative”?