Reliability, validity, and responsiveness of Severity of Dyspepsia Assessment (SODA) in a randomized clinical trial of a COX-2-specific inhibitor and traditional NSAID therapy

Abstract

OBJECTIVES: We aimed to assess the Severity of Dyspepsia Assessment (SODA) scales as measures of change in dyspepsia-related health in a blinded, randomized, controlled trial in arthritis patients treated with nonsteroidal anti-inflammatory drugs. METHODS: Three thousand nine hundred seven arthritis patients completed SODA at baseline and weeks 4, 13, 26, and 52 and/or at early termination. Using baseline and 4-wk data, reliability was evaluated with Cronbach’s α and the intraclass correlation coefficient (ICC). Dyspepsia adverse events were defined based on a combined set of World Health Organization Adverse Reaction Terminology terms. The ability of SODA to measure change in dyspepsia-related health was evaluated by comparing SODA change scores by dyspepsia adverse event severity level and withdrawal status. Responsiveness was further evaluated by the area under the curve (AUC) from receiver operating characteristic curves using withdrawal due to dyspepsia as the criterion. RESULTS: The SODA scales—Pain Intensity (α = 0.93), Non Pain Symptoms (α = 0.82), and Satisfaction (α = 0.89)—demonstrated excellent internal consistency reliability using baseline data. Reproducibility was fair to good: Pain Intensity ICC = 0.49, Non Pain Symptoms ICC = 0.61, and Satisfaction ICC = 0.45. SODA change scores (4-wk score − baseline score) increased, or worsened, with increasing dyspepsia severity and differentiated between adjacent levels of dyspepsia severity for eight of nine adjacent comparisons ( p < 0.05). SODA change scores also differentiated between those who did and did not withdraw ( p < 0.001). Responsiveness was highest with the Pain Intensity scale (AUC = 0.78), followed by the Non Pain Symptoms (AUC = 0.74) and Satisfaction (AUC = 0.75) scales. CONCLUSIONS: SODA is a reliable, valid instrument for use as a measure of dyspepsia tolerability in future clinical trials involving cyclo-oxygenase-2-specific and/or traditional nonsteroidal anti-inflammatory drugs.