Sleep quality, the neglected outcome variable in clinical studies focusing on locomotor system; a construct validation study
1 Institute for Evaluative Research in Medicine, University of Bern, Stauffacherstrasse 78, 3014 Bern, Switzerland
2 Department Rheumatology and Institute of Physical Medicine, University Hospital of Zürich, Rämistrasse 100, 8091 Zürich, Switzerland
BMC Musculoskeletal Disorders 2010, 11:224 doi:10.1186/1471-2474-11-224Published: 28 September 2010
In addition to general health and pain, sleep is highly relevant to judging the well-being of an individual. Of these three important outcome variables, however, sleep is neglected in most outcome studies.
Sleep is a very important resource for recovery from daily stresses and strains, and any alteration of sleep will likely affect mental and physical health, especially during disease. Sleep assessment therefore should be standard in all population-based or clinical studies focusing on the locomotor system. Yet current sleep assessment tools are either too long or too specific for general use.
Based on a literature review and subsequent patient-based rating of items, an expert panel designed a four-item questionnaire about sleep. Construct validation of the questionnaire in a random sample of the German-speaking Swiss population was performed in 2003. Reliability, correlation, and tests for internal consistency and validity were analyzed.
Overall, 16,634 (70%) out of 23,763 eligible individuals participated in the study. Test-retest reliability coefficients ranged from 0.72 to 0.87, and a Cronbach's alpha of 0.83 indicates good internal consistency. Results show a moderate to good correlation between sleep disturbances and health perception, and between sleep disturbances and overall pain.
The Sleep Standard Evaluation Questionnaire (SEQ-Sleep) is a reliable and short tool with confirmed construct validity for sleep assessment in population-based observational studies. It is easy to administer and therefore suitable for postal surveys of the general population. Criterion validity remains to be determined.