May need to revise because the small site may not be able to participate due to IRB issues. Should know by next Wednesday when we meet again.
Sent from my iPhone > On Mar 18, 2015, at 10:45 AM, "GPC Informatics" <[email protected]> wrote: > > #244: Generate randomized BC patient sets > ------------------------------+----------------------------------- > Reporter: vleonardo | Owner: bchrischilles > Type: task | Status: assigned > Priority: major | Milestone: bc-survey-cohort-def > Component: data-sharing | Resolution: > Keywords: bc-research-team | Blocked By: > Blocking: 227 | > ------------------------------+----------------------------------- > > Comment (by vleonardo): > > Per 11 March GPC BC meeting, Betsy asked that we use a more evenly > distributed method for patient rates, than using a site-specific > percentage-based method. The newly proposed method is a three step > process. > > Step 1) For sites with less than 200 eligible patients, survey all > patients eligible. > > Step 2) Calculate the total number of add'l patients desired by > subtracting 'Step 1' from 2000 (which represents 1000 received surveys at > an expected 50% response rate. > > Step 3) Divide 'Step 2', the 'residual total needed' by the remaining > number of sites, evenly. > > This means all sites that have more than 200 eligible patients will be > surveying the same number of patients. > > At this point, there is just 1 site that has less than 200 eligible > patients. > > -- > Ticket URL: > <http://informatics.gpcnetwork.org/trac/Project/ticket/244#comment:6> > gpc-informatics <http://informatics.gpcnetwork.org/> > Greater Plains Network - Informatics _______________________________________________ Gpc-dev mailing list [email protected] http://listserv.kumc.edu/mailman/listinfo/gpc-dev
