DATA POOLING OF PATIENT-REPORTED OUTCOMES IN CLINICAL TRIALS- EVALUATION OF STATISTICAL TECHNIQUES FOR ASSESSING MEASUREMENT EQUIVALENCE
Author(s)
Nixon MQuintiles, Bracknell, Berkshire, United Kingdom
Presentation Documents
OBJECTIVES: This analysis describes the development, application and comparison of three different approaches to evaluate measurement equivalence properties of a patient reported outcome (PRO) questionnaire applied to two treatment groups for gastroesophageal reflux disease (GERD). The data used in this analysis was obtained from an on-line patient community, iGuard.org. Patients using either of the two treatments were randomly invited to complete a measure of treatment satisfaction, the Treatment Satisfaction Questionnaire for Medication (TSQM). METHODS: Three statistical approaches were used to evaluate the measurement equivalence of the TSQM across the two patient populations: 1) Classical Test Theory (CTT) to assess the internal consistency of the TSQM items within each of the three factors using Cronbach’s alpha; 2) Confirmatory Factor Analysis (CFA) using a special case of structural equation modelling (SEM); and 3) Item Response Theory (IRT) – based technique of Differential Item Functioning (DIF). RESULTS: All three statistical methods indicated measurement equivalence had been achieved across the two treatment populations for all the three domains of the TSQM. The effectiveness and global satisfaction domains exhibited the strongest significant results amongst all three tests. However, while the convenience domain exhibited strongly significant measurement equivalence for the CTT, it only exhibited significant results for the SEM and DIF. CONCLUSIONS: While all three methods indicated the same overall results, there is some suggestion of differing sensitivity amongst the tests.
Conference/Value in Health Info
2010-11, ISPOR Europe 2010, Prague, Czech Republic
Value in Health, Vol. 13, No. 7 (November 2010)
Code
PMC43
Topic
Patient-Centered Research
Topic Subcategory
Patient-reported Outcomes & Quality of Life Outcomes
Disease
Multiple Diseases