PSYCHOMETRIC EVALUATION OF CLINICAL OUTCOMES ASSESSMENTS IN A PHASE II TRIAL
Author(s)
Huang VW1, Banderas B2, Sen R2
1Novartis, East Hanover, NJ, USA, 2Adelphi Values USA, Boston, MA, USA
Presentation Documents
OBJECTIVES: This study utilized in-trial data to evaluate the validity and reliability of scores of 3 patient reported outcomes (PRO): Gastrointestinal (GI) Symptom Diary, Modified Satisfaction with Iron Chelation Therapy (MSICT) Questionnaire and Palatability Questionnaire. METHODS: This was a randomized, open-label, multi-center, 2-arm, phase 2 study to evaluate the safety of Deferasirox film-coated tablet and Deferasirox dispersible tablet formulations in patients with transfusion-dependent thalassemia or myelodysplastic syndrome treated for 24 weeks (wks). PROs were administered at wks 2, 3, 13 and end of treatment (EOT). Upon completion of data collection, psychometric analyses were conducted and included reliability (i.e. test-retest, internal consistency using Cronbach’s α), construct-related validity (including convergent and discriminant validity correlation estimates and conducting a set of known‑groups analyses), and distribution-based testing. RESULTS: Data were available for 154 patients for the MSICT and Palatability Questionnaires and for 149 patients for GI Symptom Diary (at either wks 2, 3, 13 or EOT). 51% of the patient population was females with a mean age of 35 years. All PRO scores demonstrated adequate test-retest reliability (intra-class correlations ≥0.69). Internal consistency reliability estimates were moderate to high for GI Symptom Diary and MSICT domains (Cronbach’s α >0.60), and moderate for the Palatability Questionnaire (α >0.45). GI symptom scores demonstrated validity based on the moderate to high correlations between reported adverse events and GI items. Palatability scores differentiated between known-groups. Distribution-based minimal important difference (MID) results revealed approximately 1-point changes for the MSICT domains and Palatability scores to be deemed as clinically meaningful for group level comparisons. MIDs for GI Symptom overall scores ranged from 0.24 to 0.39. CONCLUSIONS: There is evidence supporting the reliability and validity for all 3 PRO measures. Scores showed relatively small MID estimates, which were not unexpected given the patient population and goals for the trial.
Conference/Value in Health Info
2016-10, ISPOR Europe 2016, Vienna, Austria
Value in Health, Vol. 19, No. 7 (November 2016)
Code
PCN210
Topic
Patient-Centered Research
Topic Subcategory
Patient-reported Outcomes & Quality of Life Outcomes
Disease
Oncology