Validation of the Qualms Questionnaire to Assess Health-Related Quality of Life in European and Israeli Patients with Myelodysplastic Syndromes: Results from the MDS-Right Project

Background

The assessment of health-related quality of life (HRQoL) in patients with myelodysplastic syndromes (MDS) is critical to organize more robustly patient-centered care. Therefore, availability of well-validated disease-specific HRQoL measures in this area is important.

Objectives

We investigate the validity of the Quality of Life in Myelodysplasia Scale (QUALMS), to assess HRQoL in MDS patients reported to the European MDS (EUMDS) Registry. A prior validation effort has been reported (Abel G, et al. Haematologica, 2016 June; 781-88), but this was largely a North American cohort and subscales suggested by the original principal components analysis had not been validated.

Patients and Methods

The QUALMS questionnaire was included in the EUMDS Registry (de Swart L, et al. Br J Haematol. 2015 Aug;170:372-83), a large international initiative including several European countries and Israel. We used proportions, median and interquartile range (IQR) to summarize patients' characteristics at baseline. We also assessed internal consistency and test-retest reliability by Cronbach's alpha (threshold for acceptability Cronbach's alpha ≥0.70). Confirmatory factor analysis (CFA) was used to examine the model fit for the underlying scale structure. The model performance was evaluated by Comparative Fit Index (CFI), the Tucker-Lewis Index (TLI), and the Root Mean Squared Error of Approximation (RMSEA), where acceptable values were >0.90 for CFI and TLI and RMSEA<0.08. We assessed concurrent validity calculating the Spearman's rank correlation between scales from QUALMS and EQ-5D-3L questionnaire. Discriminant validity was assessed using the Wilcoxon-Mann-Whitney test to compare the scores between patient subgroups, defined as low vs. intermediate/high MDS comorbidity index (MDS-CI), lower vs higher (< 90 vs. ≥ 90) Karnofsky performance status (KPS), anemic vs. non-anemic patients as defined by gender-specific thresholds from the World Health Organization, males vs. females and having received or not red blood cell (RBC) transfusions previously. Responsiveness to change of QUALMS scales was evaluated by Wilcoxon signed-rank test to assess changes in the scores from baseline in patients who reported an improvement in Hb-levels (≥1.5 g/dL) and in those who reported a clinically meaningful deterioration in the Visual Analogue Scale (VAS) of the EQ-5D-3L.

Results

Overall, 270 MDS patients completed the QUALMS questionnaire, from 17 centers across four countries: Austria (N = 61), Israel (N = 67), The Netherlands (N = 37) and United Kingdom (N = 105). The majority of patients were male (67.4%) with a median age at diagnosis of 74 years (IQR =68.0-80.0) and a low MDS-CI score (n=158). CFA indicated acceptable model fit, respectively CFI=0.93, TLI=0.93 and RMSEA=0.07. For all analyses, the scales QUALMS-physical burden (P), QUALMS-emotional burden (E) and the total score showed satisfactory results, while the scale QUALMS-benefit finding (BF) showed worse results. Internal consistency and test-retest reliability were acceptable for the total score (TS) and all scales but BF (see figure). Pairwise correlations between QUALMS-P, QUALMS-E, TS and all the scales from EQ-5D-3L (including VAS) showed satisfactory concurrent validity (range from -0.67 to 0.60). Only the QUALMS-BF did not correlate with any of the EQ-5D-3L scales (range from -0.07 with EQ-VAS to 0.14 with EQ-5D Anxiety/ Depression). As per discriminant validity, all known-group comparisons were statistically significant and consistent with the underlying clinical construct for all scales apart for the QUALMS-BF. For example, the QUALMS-P mean scores were 75.8 vs 54.6 (p< 0.001) for patients with a higher vs lower KPS, while the corresponding QUALMS-BF mean scores were 49.4 vs 50.9 (p=0.829). The same profile resulted for responsiveness to change, where all scales reflected significant changes in Hb levels and in EQ-VAS score in the expected direction, while QUALMS-BF did not. For example, those patients who experienced a clinically meaningful deterioration in the VAS score also had also a worsening in QUALMS-P (Δ=-6, p=0.003), QUALMS-E (Δ=-1.8, p=0.075) and TS (Δ=-3, p=0.003).

Conclusions

Current findings provide novel evidence-based data that support the validity of the QUALMS for its use in European and Israeli patients with MDS. Implementation of the QUALMS in MDS research may further raise PRO standards in this area.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal