Reliability, construct validity and responsiveness to change of the PROMIS-29 in systemic sclerosis-associated interstitial lung disease

OBJECTIVES

PROMIS-29 is a generic health-related quality of life instrument. Our objective was to assess the reliability, construct validity, and responsiveness to change of PROMIS-29 in systemic sclerosis-associated interstitial lung disease (SSc-ILD).

METHODS

Seventy-three participants with SSc-ILD were administered patient reported outcomes (PROs) at baseline and follow-up visits which included PROMIS-29 and other measures of generic health, dyspnea, and cough instruments. We assessed internal consistency reliability using Cronbach's α, an alpha of ≥ 0.70 was considered satisfactory. We assessed the responsiveness to change using linear regression models.

RESULTS

Mean age of the participants was 51.9 years and the mean disease duration was 7.9 years after first non-Raynaud's symptom. Of the 73 participants, 56.2% were classified as diffuse SSc and 26% limited SSc. The baseline (mean ± SD) FVC % predicted was 73.9±15.5 with a DLCO % predicted of 57.7±21.1; 95.9% had fibrotic NSIP pattern on HRCT. PROMIS-29 scores were 0.2 to 0.9 SD below the US population. Cronbach's α reliability was acceptable for all domains (ranged from 0.77 to 0.98). All scales showed statistically significant correlations with hypothesised PROMIS-29 domains (p≤0.05 for all comparisons). PROMIS-29 showed none-to-small discriminatory ability in comparison with physiologic measures (FVC and DLCO). There was no significant relationship between the change in FVC versus the change in PROMIS-29 measures over time.

CONCLUSIONS

PROMIS-29 has adequate reliability and construct validity for evaluation in SSc-ILD. It has moderate-to-large correlations with other PROs. The PROMIS-29 domains were not found to change over time in this cohort, likely due to stable nature of the observational cohort.