Objective The child sleep habits questionnaire (CSHQ) has been developed in the USA for measuring medical and behavioural sleep disorders in school-aged children. The aim of this study was to assess the reliability and cross-cultural validity of a Dutch version of the CSHQ.
Methods 2453 questionnaires were distributed among 4–14-year-old children in primary schools to be completed by the parent/guardian at home. An identical second questionnaire was distributed for test–retest (n = 159) and inter-observer (n = 210) reliability, which were assessed using intraclass correlation (ICC) and compared with results that were found in US and Chinese samples. Validity was analysed using internal consistency (Cronbach’s α) and confirmative factor analysis.
Results In total, 1552 (63%) parents returned the questionnaire; 75 parents (47%) for test–retest and 67 parents (32%) for inter-observer reliability. Test–retest ICC was moderate to good, ranging from 0.47 (bedtime resistance) to 0.83 (sleep onset delay). With the exception of sleep duration, inter-observer ICC was moderate to good, ranging from 0.54 (sleep onset delay) to 0.83 (sleep disordered breathing). Cronbach’s alpha ranged from 0.47 to 0.68. In confirmatory factor analysis the domain structure of the original CSHQ could not be confirmed. Preliminary analysis suggests a smaller number of subscales than the original eight domains.
Conclusion The CSHQ has an adequate reliability and internal consistency in a western European population with different socio-cultural characteristics than the US population in which it has been devised. Rearrangement of items in a smaller number of subscales might further improve the CSHQ.