Nonrestorative sleep scale: reliable and valid for the Chinese population

Purpose: To conduct a linguistic and psychometric evaluation of a Chinese version of the Nonrestorative Sleep Scale (NRSS). Methods: The Chinese NRSS was created from a standard forward–backward translation and trialed on 10 Chinese adults. Telephone interviews were then conducted with 100 adults, who completed the Chinese NRSS, the Pittsburgh Sleep Quality Index (PSQI), the Athens Insomnia Scale (AIS), the Center for Epidemiological Studies Depression Scale (CES-D), and the Toronto Hospital Alertness Test (THAT). A household survey was conducted with 20 subjects, followed by a confirmatory factor analysis (CFA), and a bifactor model was developed to evaluate the reliability and validity of the NRSS. Results: The bifactor model had the root mean square error of approximation (RMSEA), standardized root mean square residual (SRMR), and comparative fit index (CFI) of 0.06, 0.06, and 0.97, respectively. Convergent validity was shown from the moderate associations with PSQI (r = − 0.66, P < 0.01), AIS (r = − 0.65, P < 0.01), CES-D (r = − 0.54, P < 0.01), and THAT (r = 0.68, P < 0.01). The coefficient omega (0.92), omega hierarchical (0.81), factor determinacy (0.93), H value (0.91), explained common variance (0.63), and percentage of uncontaminated correlations (0.80) derived from the bifactor CFA supported the essential unidimensionality of NRSS. Conclusions: The Chinese NRSS is a valid and reliable essential unidimensional tool for the assessment of nonrestorative sleep in the Chinese population.