Validity of the Pittsburgh Sleep Quality Index in Indian University Students.

Abstract

Despite the demonstrated utility of the Pittsburgh Sleep Quality Index (PSQI) in various demographic groups, it has never been validated in a sample of Indian subjects. To extend and confirm the PSQI's applicability for South Asian subjects, this preliminary study aimed to assess its psychometric and diagnostic validity in a sample of university students. Forty-seven male students were recruited from Jamia Millia Islamia, a public central university in New Delhi, India. The mean age of the students was 23.4±3.9 years, and they had a mean body mass index (BMI) of 23.3±3.3kg/m(2). The PSQI was administered to all subjects and overnight polysomnographic testing was carried out as a concurrent validation measure. Cronbach's alpha for the questionnaire was found to be 0.736. Internal homogeneity was high, with the majority of correlations between questionnaire component scores and the summed global score being significant (p<0.010). Criterion validity-correlations between the PSQI global score and polysomnography (PSG) measures were low. However, the questionnaire component scores and the related polysomnographic measures did show some significant relationships. The optimal cut-off scores for distinguishing students with/without sleep problems was >6 and was generated using receiver operating characteristic curve analysis. The area under the curve, sensitivity, specificity, positive and negative likelihood ratios at the cut-off score were 0.838 (p<0.0001), 75.0%, 88.9%, 6.75, and 0.280, respectively. The study found evidence that the PSQI had internal consistency, internal homogeneity, and diagnostic characteristics that compared well with PSG among a sample of young adult male students in India. This supports the applicability and certain aspects of the validity of the PSQI in the population.