Reliability, Validity, and Responsiveness of InFLUenza Patient-Reported Outcome (FLU-PRO (c)) Scores in Influenza-Positive Patients

摘要

Objectives: To assess the reliability, validity, and responsiveness of InFLUenza Patient-Reported Outcome (FLU-PRO (c)) scores for quantifying the presence and severity of influenza symptoms. Methods: An observational prospective cohort study of adults (= 18 years) with influenza-like illness in the United States, the United Kingdom, Mexico, and South America was conducted. Participants completed the 37-item draft FLU-PRO (c) daily for up to 14 days. Item-level and factor analyses were used to remove items and determine factor structure. Reliability of the final tool was estimated using Cronbach alpha and intraclass correlation coefficients (2-day reliability). Convergent and known groups validity and responsiveness were assessed using global assessments of influenza severity and return to usual health. Results: Of the 536 patients enrolled, 221 influenza-positive subjects comprised the analytical sample. The mean age of the patients was 40.7 years, 60.2% were women, and 59.7% were white. The final 32-item measure has six factors/domains (nose, throat, eyes, chest/respiratory, gastrointestinal, and body/systemic), with a higher order factor representing symptom severity overall (comparative fit index = 0.92; root mean square error of approximation = 0.06). Cronbach alpha was high (total = 0.92; domain range = 0.71-0.87); test-retest reliability (intraclass correlation coefficient, day 1 day 2) was 0.83 for total scores and 0.57 to 0.79 for domains. Day 1 FLU-PRO (c) domain and total scores were moderately to highly correlated (= 0.30) with Patient Global Rating of Flu Severity (except nose and throat). Consistent with known-groups validity, scores differentiated severity groups on the basis of global rating (total: F = 57.2, P 0.001; domains: F = 8.9-67.5, P 0.001). Subjects reporting return to usual health showed significantly greater (P 0.05) FLU-PRO score improvement by day 7 than did those who did not, suggesting score responsiveness. Conclusions: Results suggest that FLU-PRO scores are reliable, valid, and responsive to change in influenza-positive adults.