2006D-0044
  

  
 Guidance for Industry on Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims

  


  

  

  


  

  


  

  


  

  
FDA Comment Number : 
  

  
EC4
  



  

  
Submitter : 
  

  
Mr. Allen Meyerhoff
  

  
Date & Time: 
  

  
04/07/2006 05:04:00
  



  

  
Organization : 
  

  
Capitol Outcomes Research, Inc.
  



  

  
Category : 
  

  
Health Professional
  



  

  
Issue Areas/Comments 
  


  

  


  

  
GENERAL
  


  

  



  

  
GENERAL
  


  

  



  

  
 In the first paragraph of IV.D., the document indicates that measurement properties of an existing PRO instrument that has been revised with minor changes may be performed in a non-randomized study, but that more significant changes, such as use in a different patient population, may require a small, randomized study.  



Being very familiar with use of PRO instruments and issues surrounding their psychometric properties, I am at an utter loss to understand what kind of research design would involve randomization in this context.  I have seen many studies evaluating PRO instrument measurement properties, and they are either one-group cross-sectional, or better yet, one-group longitudinal designs.



Accordingly, if one desires to assess psychometric properties of an existing instrument in a population different that the one the instrument was developed for, what instrument(s) would be randomized to what populations?



Thank you.