An empirical comparison of sleep-specific versus generic quality of life instruments among Australians with sleep disorders.

Abstract

Purpose: In Australian adults diagnosed with a sleep disorder(s), this cross-sectional study compares the empirical relationships between two generic QoL instruments, the EuroQoL 5-dimension 5-level (EQ-5D-5L) and ICEpop CAPability measure for Adults (ICECAP-A), and three sleep-specific metrics, the Epworth Sleepiness Scale (ESS), 10-item Functional Outcomes of Sleep Questionnaire (FOSQ-10), and Pittsburgh Sleep Quality Index (PSQI).

Methods: Convergent and divergent validity between item/dimension scores was examined using Kendall's Tau-B correlation, with correlations below 0.30 considered weak, between 0.30 and 0.50 moderate and those above 0.50 strong (indicating that instruments were measuring similar constructs). Exploratory factor analysis (EFA) was conducted to identify shared underlying constructs.

Results: A total of 1509 participants (aged 18-86 years) were included in the analysis. Convergent validity between dimensions/items of different instruments was weak to moderate. A 5-factor EFA solution, representing 'daytime dysfunction', 'fatigue', 'wellbeing', 'physical health', and 'perceived sleep quality', was simplest with close fit and fewest cross-loadings. Each instrument's dimensions/items primarily loaded onto their own factor, except for the EQ-5D-5L and PSQI. Nearly two-thirds of salient loadings were of excellent magnitude (0.72 to 0.91).

Conclusion: Moderate overlap between the constructs assessed by generic and sleep-specific instruments indicates that neither can fully capture the complexity of QoL alone in general disordered sleep populations. Therefore, both are required within economic evaluations. A combination of the EQ-5D-5L and, depending on context, ESS or PSQI offers the broadest measurement of QoL in evaluating sleep health interventions.