Objective: To determine the sensitivity, specificity and the positive and negative predictive value of taking a detailed sleep history for making the diagnosis of psychophysiological insomnia.
Design: Retrospective case file study.
Method: We examined 767 patients referred to the Amsterdam Centre for Sleep and Wake Disorders, and who underwent polysomnography for the first time between 1 January and 31 December 2010. We compared the probable diagnosis made following history-taking with the final diagnosis made after polysomnography. In this we differentiated between organic and non-organic insomnia. The sensitivity, specificity, positive and negative predictive values of the sleep history were calculated.
Results: In 24.8% of the 303 patients whose histories did not indicate organic insomnia, polysomnography showed there to be an organic cause. Primary causes were obstructive sleep apnoea (13.2%), upper airway resistance syndrome (5.4%), and periodic limb movement disorder (4.0%) or a combination of these. In the histories of 464 patients there were indications that the insomnia had an organic cause and in 325 of them this was confirmed by polysomnography. The sensitivity of detailed history taking to psychophysiological insomnia was 62.1%, the specificity 81.3%, the positive predictive value was 75.2% and the negative predictive value was 70.0%. In patients under the age of 40 with a score on the Epworth sleepiness scale < 10 (i.e. no hypersomnolence), a BMI < 25 kg/m2 and indications of psychophysiological insomnia, organic insomnia could not be demonstrated, with the exception of one parasomnia.
Conclusion: History-taking only meant that the organic cause was missed in a substantial percentage of patients with insomnia, in particular in older patients with hypersomnolence and a high BMI.