ABSTRACT
Epidemiological studies on sleep often use questionnaires, and measurement of validity provides necessary guidance in selection of valid single sleep questions. Twenty-five items assessing different aspects of sleep, including overall sleep quality, specific sleep parameters, nocturnal restoration, and exposure-related questions, were tested. This involved coherence with objective polysomnographic (PSG) laboratory measurements of sleep in 47 participants and application of selected items under field conditions in over 3,000 participants. Items on overall sleep quality correlated significantly with PSG data. For specific sleep parameter questions, tiredness in the morning, time to fall asleep, difficulties to sleep and estimated number of awakenings were correlated to PSG data. Questions asking specifically about the effect of potential sleep disturbances correlated poorly with PSG data, but showed highest effects between environmental exposure (noise and vibration) and control nights in the laboratory and highest correlation with the dose of exposure in the field. In conclusion, healthy participants seem to be able to access their sleep reliably; and sleep questions asking about specific sleep parameters can be recommended for the assessment of sleep.
