Psychometric evaluation and normative data for the Karolinska Sleep Questionnaire

Abstract

The study objective was to validate the Karolinska Sleep Questionnaire with regard to dimensionality, internal consistency, and construct and criterion validity. Another objective was to provide normative data. Data from the cross-sectional Vasterbotten Environmental Health Study in Sweden were used. The 3406 participants in this study, 18 to 79 years old, constituted a random sample stratified for age and sex. Along with the Karolinska Sleep Questionnaire, the participants responded to the Hospital Anxiety and Depression Scale, the Perceived Stress Scale, and the Shirom Melamed Burnout Questionnaire for assessing construct validity. Factor analyses of the questions in the Karolinska Sleep Questionnaire that relate to nocturnal sleep revealed the dimensions sleep quality, non-restorative sleep, and sleep apnea. A separate factor analysis on the questions regarding day time sleepiness revealed a sleepiness dimension. The sleep quality, non-restorative sleep, and sleepiness dimensions showed approximate normal distributions, whereas the distribution for sleep apnea was positively skewed. All dimensions showed good internal consistency. Satisfactory construct validity was found for all dimensions. Using the DSM-IV criteria of insomnia, relevant questions in the Karolinska Sleep Questionnaire were combined into an index of nocturnal symptoms of insomnia as were questions of sleep apnea. Prevalences of insomnia and sleep apnea correspond well to those of other studies, indicating good criterion validity. The favorable psychometric properties of the dimensions and indexes from the Karolinska Sleep Questionnaire suggest their use for assessing sleep quality, non-restorative sleep, sleep apnea, sleepiness and nocturnal symptoms of insomnia.