Examining the ability to detect change using the TRIM-Diabetes and TRIM-Diabetes Device measures

Abstract

Purpose: Responsiveness is defined as the ability of an instrument to accurately detect change when it has occurred and is an essential psychometric property of a patientreported outcomes (PRO) measure to understand and interpret study findings. This study examined the responsiveness of 2 Treatment Related Impact Measures (TRIMs): The TRIM-Diabetes (TRIM-D) and TRIM-Diabetes Device (TRIM-DD) as well as confirmed their measurement models in a randomized controlled trial (RCT) design. Methods: The data were collected in a multi-center, randomized, open-label (2 × 12 week), cross-over study of two prefilled pens in subjects with type 1 or type 2 diabetes, age 18 or older. Internal and external responsiveness were examined. To confirm the measurement model identified in the previous study, the Bentler comparative fit index (CFI) and internal consistency for the RCT sample scores were examined and compared. Results: Based on a priori criteria, tests of responsiveness were confirmed with patients having significant improvements over time ranging from 2.7 (Psychological Health) to 11.1 (Treatment Burden) (P<0.01) (effect sizes ranging from 0.2 to 0.8). The previous measurement model factor structure was confirmed (CFI ranging from 0.8 to 1.0), and internal consistency of the TRIMs was similar to the developmental findings. Conclusions: The total score as well as all domain scores of the TRIMs was significantly responsive over time, thus acceptable internal and external responsiveness of TRIM-D and TRIM-DD are concluded. To date, all validation evidence supports the use of these two measures in future clinical trials. © The Author(s) 2011.