-- JoAnnAlvarez - 24 Mar 2011

2011 April 4

Present: JoAnn, Warren, Frank Harrell

Topic: Randomization plan

• We will use random permuted blocks of size 4 and 6.
• Frank recommends doing the randomization centrally at VUMC and having the sites call in.
• JoAnn will use the block rand package in R using site as a stratifying variable to make the list.
• JoAnn will need the names of the sites.

2011 March 24

Present: JoAnn, Rosemary, Brooke

Topic: Need to come up with and program a randomization scheme for the upcoming Moon Shoulder Post Op Study.

• The study will start in a couple of months and be submitted to the IRB in a couple of weeks, ideally.
• There are two treatment groups: immediate and delayed.
• There will be about 9 sites, with 2 of the sites having much more enrollment. Some sites will have a delayed entry into the study (maybe over a three-month period).
• Enrollment will continue until 214 patients are enrolled. This is anticipated to take 6 months to one year.
• The protocol states the following about the randomization:

D.6.2. Randomization procedures Participants will be randomized with equal probability into one of the two study arms. Prior to study initiation, a biostatistician will generate a randomization scheme using the statistical software R. Permuted-block randomization will be used to ensure equal numbers of patients in the two study arms.

• Randomization goals:
  • Overall group balance
  • Group balance within site
  • Group balance over time