Activity Number:
|
233
- Innovations in Inferential and Design Strategies in Mental Health Research
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 31, 2017 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Mental Health Statistics Section
|
Abstract #322446
|
View Presentation
|
Title:
|
Sample Size Determination for Number of Patient Interviews When Developing a PRO Using Qualitative Research Based on the FDA PRO Guidance
|
Author(s):
|
Chris Barker*
|
Companies:
|
Statistical Planning and Analysis Services, Inc.
|
Keywords:
|
Patient Reported Outcomes ;
sample size ;
qualitative research ;
PRO instrument development ;
FDA
|
Abstract:
|
PRO's developed de novo, using the FDA guidance require an initial step of structured patient interviews or focus groups. Criteria for selecting patients includes "patient diversity". Qualitative Research is a methodology for eliciting and coding interviews and produces concepts or themes. These concepts are used to develop items in a PRO. A convention in the field is that interviews and code/concept elicitation are considered complete when subsequent interviews produces "no new concepts" -termed "saturation". FDA reviewers frequently challenge PRO developers on whether or not enough patient interviews have been conducted to confirm that saturation is achieved. Guest (2006) demonstrated "no new concepts" may occur more than once. First occurrence of no new concepts without a confirmation that saturation achieved is a Type I error in my methodology. This research presents the first formal statistical methodology for confirming saturation is achieved, specifically to address regulatory concern about saturation. I demonstrate a method for simulation of the process leading to concept saturation. I outline my ongoing statistical work for estimating the total number of interviews.
|
Authors who are presenting talks have a * after their name.