Abstract
PURPOSE: Quality of sleep in patients with cancer is regarded as of utmost importance. The aim of the present study was to assess psychometric properties and feasibility of the Greek version of the Pittsburgh Sleep Quality Index (GR-PSQI). METHODS: Following a "forward-backward" procedure, the scale was translated into Greek. The GR-PSQI was administered as a self-report instrument to 209 consecutive patients with cancer during active-phase chemotherapy treatment. For stability analysis purposes, a subgroup of 60 patients completed the GR-PSQI on two occasions, 14-21 days apart. All participants also completed the Insomnia Severity Index, the Epworth Sleepiness Scale-Greek version, a Sleep Quality-Visual Analogue Scale and the Hospital Anxiety and Depression Scale-Greek version. Validity and reliability analyses were performed for GR-PSQI data. RESULTS: The Chronbach's alpha for the global GR-PSQI score was 0.76. Test-retest reliability analysis for the global GR-PSQI score yielded a high intra-class correlation coefficient of 0.82 (p < 0.001). Exploratory factor analysis generated a two-factor structure for the GR-PSQI, [quality of nocturnal sleep] and [daily disturbances and management of sleep problems]. This construct was further supported by its high correlations with similar content instruments, as well as by the instrument's ability to discriminate well between contrasting groups of patients with different levels of anxiety, depression and performance status. CONCLUSIONS: The present findings support the GR-PSQI as a reliable, stable over time and valid sleep quality instrument when administered to patients with cancer during chemotherapy treatment; however, it is suggested that the use of a two-factor scoring method (instead of the traditional unidimensional) could improve its sensitivity in this patient group.