Leaving the Lab: Lessons Learned in Cognitive Testing of Tobacco Use Questions in a Clinical Trials Setting (303550)

Thomas Brandon, H. Lee Moffitt Cancer Center 
Veronica Chollette, National Cancer Institute 
*Jennifer Lynn Crafts, Westat 
Sonia Duffy, The Ohio State University 
Jasmine Folz, Westat 
Dorothy Hatsukami, Masonic Cancer Center - University of Minnesota 
Stephanie Land, National Cancer Institute 
Jamie Ostroff, Memorial Sloan Kettering Cancer Center 
Benjamin Toll, Yale School of Medicine 
Graham Warren, Medical University of South Carolina 
Gordon Willis, National Institutes of Health 

Keywords: cognitive testing, questionnaire development, questionnaire pretesting, special populations

To investigate the relationship between tobacco use and the effectiveness of a broad variety of cancer treatments and therapies, cancer researchers have been in need of a set of standard longitudinal measures of tobacco use. The objective of this study was to test a set of questions designed specifically for use with cancer patients and survivors. Several prior studies have used cognitive testing techniques to develop survey questions for patient populations. However, these studies did not investigate unique methodological circumstances associated with recruiting clinical trials patients as interview participants, identifying potential interview candidates through accessing medical record data, or conducting cognitive interviews with cancer patients and survivors in a clinical setting.

The questions were initially designed through an extensive literature review and an expert review by cancer researchers and survey methodologists. Three rounds of 10 interviews each were conducted at NIH’s Center for Cancer Research with patients enrolled in clinical trials. This poster will provide lessons learned in preparing for and conducting the cognitive interviews within a clinical setting---a realistic context for administration of the finalized set of questions. We will identify similarities and differences between testing in a clinical setting vs. a more typical cognitive lab setting. We will compare and contrast these two settings on factors including (1) protocol planning and development steps, (2) formal approval processes, (3) participant sources, (4) recruiting techniques, (5) consent procedures, (6) confidentiality requirements, (7) interview session procedures, (8) data analysis approach, and (9) generalizability of findings.