Validity of the epworth sleepiness scale as a screening tool for obstructive sleep apnea

Objectives/hypothesis: To determine whether the Epworth Sleepiness Scale (ESS) can be improved by having a bed partner or roommate ("partner") also fill out the questionnaire; and to assess the applicability of the eight items in the questionnaire, particularly in an urban population.

Study design: Retrospective chart review.

Methods: A retrospective chart review was conducted to identify charts of subjects with an ESS score collected from both the subject and a partner. Consensus scores were also identified. Self-reported ESS scores were compared to partner and consensus ESS scores. The number of subjects who did not drive was determined.

Results: One hundred consecutive charts were included in this study. Mean ESS score was significantly higher when scored by the partner (10.4 ± 5.8) compared to the subject (7.2 ± 4.7). In 73% of cases, the subject underestimated their score compared to the score given by the partner. The mean difference in score was 3.2 ± 3.5. Mean consensus ESS score was 12.3 ± 5.4. Using an ESS score of ≥10 as abnormal, subject scores did not correlate with the presence of obstructive sleep apnea (OSA; P = .095). Partner and consensus scores, however, did correlate with the presence of OSA (P = .003 and P =.039, respectively). Thirty-five percent of our subjects did not drive.

Conclusions: The difference in mean ESS score obtained from the subject compared to partner implies that this diagnostic form should be completed as a consensus between the patient and partner to provide a more accurate score.

Level of evidence: 4 Laryngoscope, 2016 127:525-531, 2017.