Assessment of predictive ability of Epworth scoring in screening of patients with sleep apnoea

Abstract

Method: Numerous studies have considered the benefits, and the disadvantages, of the Epworth Sleepiness Scale. Following an extensive literature review, we found that the evidence was inconclusive as regards the diagnostic efficacy of Epworth scoring for obstructive sleep apnoea syndrome. We undertook a retrospective study of 343 patients who underwent a sleep assessment over a 10-year period at the Monklands Hospital.

Analysis and results: A total of 238 patients did not have sleep apnoea whereas 105 patients did. The mean Epworth score in patients with obstructive sleep apnoea syndrome was 10.94 (95 per cent confidence interval 9.46-11.42), and in the non-apnoeic group it was 7.73 (95 per cent confidence interval 7.04-8.41). Logistic regression and receiver operating characteristic curves were used to assess the predictive ability of Epworth scoring. The scores only explained 7-10 per cent of the variation in the probability of occurrence of obstructive sleep apnoea syndrome. The odds ratio for Epworth scoring was 1.118, and only 69 per cent of cases were correctly classified by the Scale.

Conclusions: The literature review suggested that the Epworth Sleepiness Scale is associated with a low effect size and/or low predictive value when correlated or regressed on the Apnoea-Hypopnoea Index or Respiratory Disturbance Index, thus limiting its value as a screening test. Our study concluded that the Epworth Scale is only marginally useful in predicting the occurrence of obstructive sleep apnoea syndrome. We believe that every patient with a direct or witnessed history of sleep apnoea with obstructive symptoms have some form of sleep assessment.