Psychometric validation of the PROMIS Fatigue-Short Form 7a in adults with newly diagnosed or recurrent Mycobacterium avium complex (MAC) lung disease: the ARISE and ENCORE studies

Fatigue symptoms contribute to the burden of Mycobacterium avium complex (MAC) lung disease. This study evaluated the psychometric properties of the Patient Reported Outcomes Measurement Information System Short Form v1.0 – Fatigue 7a (PROMIS-F SF-7a) in adults with a new or recurrent diagnosis of MAC lung disease. Data from the ARISE (NCT04677543) and ENCORE (NCT04677569) phase 3 trials were analyzed. Modern psychometric methods were employed to confirm the structural validity of the PROMIS-F SF-7a within this context of use. Classical methods were used to confirm the reliability and validity of the PROMIS-F SF-7a within this context of use. Internal consistency (McDonald’s omega, Cronbach’s alpha), test-retest reliability (two-way mixed effects intraclass correlation coefficient [ICC(2,1)]), and known-groups validity across Patient Global Impression of Severity (PGI-S) Fatigue groups were estimated. Convergent validity (Pearson correlations) was assessed by correlating PROMIS-F SF-7a scores with scores on the Exacerbations of Chronic Pulmonary Disease Tool (EXACT), EXACT Respiratory Symptoms (E-RS), St. George Respiratory Questionnaire (SGRQ), and Functional Assessment of Chronic Illness Therapy (FACIT) Fatigue Scale. Meaningful within-patient change (MWPC) thresholds were determined using anchor-based methods. The baseline sample included 231 patients (99 ARISE, 132 ENCORE). The cross-sectional validation sample comprised 230 patients (excluding 1 ARISE patient with missing item-level PROMIS-F SF-7a data). The longitudinal validation analysis sample comprised all 99 ARISE patients. Modern psychometric methods supported the relevance of all items and a unidimensional unit-weighted sum score for the PROMIS-F SF-7a. The PROMIS-F SF-7a demonstrated strong internal consistency (Cronbach’s alpha: 0.86), test-retest reliability (ICC[2, 1]: 0.76), and convergent validity (FACIT-Fatigue: −0.80, EXACT: 0.56, E-RS: 0.52, SGRQ: 0.66). Known-groups validity was demonstrated across PGI-S Fatigue groups. The MWPC analyses supported a −4.00-point median change from baseline (95% CI: −3.00 to −6.00 points) as the estimated threshold of clinically meaningful within-patient improvement for the PROMIS-F SF-7a. The PROMIS-F SF-7a is a robust, sensitive, and responsive measure of fatigue in adult patients with a new or recurrent diagnosis of MAC lung disease. It is content and psychometrically valid and appears to have interpretability to assess a threshold of MWPC for fatigue symptoms in this population. ClinicalTrials.gov. NCT04677543, registered 16 December 2020, https://clinicaltrials.gov/study/NCT04677543 . NCT04677569, registered 16 December 2020, https://www.clinicaltrials.gov/study/NCT04677569 .