DEVELOPMENT AND VALIDATION OF PATIENT-REPORTED OUTCOMES (PRO) TOOL FOR EYELASH CHARACTERISTICS
Author(s)
Somali M. Burgess, PhD, Director1, Geoffery C. Hammond, PhD, Associate Scientist2, Jason C. Cole, PhD, Director, Business Development & Senior Scientist2, Min Yang, MD, PhD, Senior Scientist2, Jan E Hansen, PhD, Vice President3, John G Walt, MBA, Director31Allergan, Inc, Irvine, CA, USA; 2 QualityMetric Health Outcome Solutions, Lincoln, RI, USA; 3 Allergan, Inc., Irvine, CA, USA
OBJECTIVES The development and validation of a PRO tool identifying and measuring domains of interest in eyelash characteristics. METHODS Development of the conceptual framework and candidate questions was based on subject focus groups followed by clinician feedback. Psychometric assessment of the item- and scale-level reliability and validity was undertaken on a non-treated, cross-sectional convenience sample of female participants using two methods of administration (68 paper version and 924 web-based). Confirmatory factor analysis, alpha statistics, convergent and discriminant validity were calculated to assess overall model fit. Model fit was also assessed in a sample receiving Latisse (bimatoprost), a prescription treatment thought to increase the length, thickness, and darkness of eyelashes (n=277). RESULTS Three conceptual domains were identified in four focus groups conducted in various areas of the U.S. with women of different ages. Physician feedback confirmed and refined the three conceptual domains: satisfaction with physical eyelash attributes (length, fullness, overall satisfaction); subjective feelings attributed to eyelash satisfaction (confidence, attractiveness, looking professional); and daily routine in making eyelashes presentable. The finalized questionnaire structure (nine questions over three domains) showed good overall model fit characteristics on the non-clinical population (x2(24)=62.9, Comparative Fit Index (CFI):0.99, Root Mean Square Error of Approximation (RMSEA) [90% CI]:0.05 [0.04-0.06]). Item-total correlations ranged from 0.55 to 0.88, with most over 0.60, indicating a large effect size. Internal consistency ranged from 0.92 to 0.98 for the three domains. Convergent validity was significantly higher than discriminant validity for these domains. No differential item functioning between methods of administration was observed. Similar fit statistics were observed in the clinical sample. CONCLUSIONS The multi-step development and validation processes identified a three-domain conceptual framework. This was also validated by psychometric assessment of the questionnaire properties in both a convenience sample and one receiving the prescription treatment.
Conference/Value in Health Info
2009-05, ISPOR 2009, Orlando, FL, USA
Value in Health, Vol. 12, No. 3 (May 2009)
Code
PSS19
Topic
Patient-Centered Research
Topic Subcategory
Patient-reported Outcomes & Quality of Life Outcomes
Disease
Sensory System Disorders