Brief Version of the Pittsburgh Sleep Quality Index (B-PSQI) and Measurement Invariance Across Gender and Age in a Population-Based Sample

Abstract

The Pittsburgh Sleep Quality Index (PSQI) is the most widely used questionnaire in research and clinical practice to assess sleep quality. However, a brief version of this measure would improve its efficiency and applicability. This study aimed to develop a brief form of the PSQI and to study measurement invariance across gender and age in a nonclinical population. In total, 609 participants with a mean age of 37.3 years (standard deviation [SD] 11.9) were recruited, of whom 71.8% (n 437) were women. Participants completed online versions of the PSQI and the Insomnia Severity Index (ISI). Reliability analyses were performed to reduce the number of items, followed by validity and measurement invariance analyses for the new Brief Version of the PSQI (B-PSQI). Six questions were included in the B-PSQI out of the initial 18; the brief form had adequate internal consistency ( .79 and 0.91). Confirmatory factor analysis showed optimal fit of the B-PSQI ( 2 (4) 22.428; p .01; comparative fit index (CFI) 0.99; normed fit index (NFI) 0.99; Tucker-Lewis index (TLI) 0.98; root mean squared error of approximation (RMSEA) 0.06; standardized root mean square residual (SRMR) 0.04), achieving partial scalar invariance across gender-same factorial structure, loadings, and thresholds in the majority of the items. Invariance across age was only achieved for model structure. Additionally, the B-PSQI yielded favorable sensitivity (75.82%) and specificity (76.99%) for classifying poor sleepers, similar to values for the full PSQI. In conclusion, the B-PSQI is a brief, reliable, and valid measure that can be used as a screening tool, allowing valid score comparisons between men and women of similar age.