Psychometric validation of the Asthma Daytime/Nighttime Symptom Diary (ADSD/ANSD) in a phase 2 trial with adults with moderate-to-severe uncontrolled asthma

<bold>Background:</bold> The Asthma Daytime/Nighttime Symptom Diary (ADSD/ANSD) were recently developed for daily assessment of asthma symptom severity in clinical trials (both feature 6 items). This study aimed to add to the measures’ growing body of psychometric evidence. <bold>Methods:</bold> Treatment-blinded ADSD/ANSD data from the rilzabrutinib phase 2 trial (<ext-link>NCT05104892</ext-link>) were analyzed (Baseline to Week 12). ADSD/ANSD item properties were evaluated, and confirmatory factor analysis (CFA) applied to assess dimensionality. ADSD/ANSD scores were assessed for internal consistency (Cronbach’s α), test-retest reliability (intraclass correlation coefficients; ICCs), and construct validity using convergent correlations (r) and known-groups (e.g., Patient Global Impression of Severity; PGIS) comparisons. Ability to detect change was examined. <bold>Results:</bold> Data from 196 patients were analyzed. ADSD/ANSD inter-item correlations were moderate to high (r:0.51-0.95). One-factor ADSD/ANSD CFA models mostly reached acceptable fit thresholds (after allowing correlated residuals between items 4 and 5). Internal consistency (α:0.94-0.97) and test-retest reliability (ICCs:0.88-0.96) were good or excellent. Construct validity evidence was shown by moderate correlations with supportive measures and large between-group effect sizes (ES) for collapsed PGIS groups. Limited ability to detect change evidence was found with PGIS. <bold>Conclusion:</bold> Findings were generally supportive of ADSD/ANSD validity and reliability in trials for patients with moderate-to-severe uncontrolled asthma. Future research should further evaluate ability to detect change and score interpretation.