P1 Psychometric properties of health-related quality of life tools for idiopathic pulmonary fibrosis

Background Assessing health-related quality of life (HRQOL) in idiopathic pulmonary fibrosis (IPF) is important clinically and for research. As there is no universally agreed HRQOL tool for IPF, a variety of different tools have been used. We aimed to compare the psychometric properties of various HRQOL tools used IPF, assess their relationship with 1-year mortality and determine minimal important clinical difference (MCID). Methods This was an observational prospective longitudinal multicentre study, involving 238 people with IPF. Participants were asked to complete HRQOL tools including EuroQol 5 dimension (EQ-5D-5L), King’s brief interstitial lung disease questionnaire (K-BILD) and St George’s Respiratory questionnaire (SGRQ), at approximate three-monthly intervals over a 12 month period. Physiological measurements including spirometry and 6 minute walking distance were captured and matched with questionnaires. Results There were 778 patient assessments with each individual having an average of 3.3 sets of questionnaires. All questionnaires showed good internal consistency with Cronbach’s alpha coefficients of >0.8. There were strong correlations between questionnaires but not with physiological measurements. People with FVC% predicted ≤70% had higher mean SGRQ and MRC scores, and lower mean EQ5D and K-BILD score. People in upper tercile of baseline KBILD and EQ-5D-5L (better health status) had significantly reduced risk of deaths than those in the lower tercile (HR 0.06; 95% CI 0.01–0.42 and HR 0.27; 95CI 0.09–0.81, respectively). Those in the upper tercile of SGRQ (worse health status) had more than 3-fold increased risk of mortality than those in the lower tercile (HR 4.65; 95% CI 1.32–16.62). The MCID (anchor method) for K-BILD was 2.3 and SGRQ was 3.9. Conclusion We recommend using the MRC dyspnoea scale rather than UCSD SOBQ, given it brevity and better known groups validity. Both the K-BILD and SGRQ were appropriate disease specific HRQOL tools for assessing people with IPF but we recommend the use of K-BILD, given its brevity and stronger relationship to mortality.