Self-reported sleep tendency poorly predicts the presence of obstructive sleep apnea in commercial truck drivers

Background and objectives: Drowsy driving increases the risk of motor vehicle crashes in those with untreated obstructive sleep apnea (OSA). Although previous studies indicated that excessive daytime sleepiness (EDS) might not predict OSA, they were not conclusive due to their small study sizes or restricted participants to sleep clinic patients. The overall objective was to determine whether self-reported EDS can be used for case identification of OSA among commercial truck drivers. Methods: Commercial truck drivers (N = 19,699) were screened for OSA-related symptoms. EDS was determined using the Epworth Sleepiness Scale (ESS) >= 11 and all participants completed the home sleep apnea test using a type 4 portable monitor to derive the respiratory event index (REI). Regression analyses were used to characterize the association between EDS and REI. Results: EDS was associated with OSA severity (p for trend <0.001). The sensitivity and specificity values of EDS for identifying moderate-to-severe OSA (REI >= 15 events/hour) were 0.10 and 0.93, respectively, and 0.48 and 0.71 if BMI >= 25 kg/m(2) was added. Those using BMI >= 25 kg/m(2) with OSA-related signs yielded the best sensitivity and specificity of 0.77 and 0.50, which were not improved by the addition of EDS. Conclusions: Despite the associations between EDS and OSA severity and between OSA and lethal crash, case-identification of OSA using the ESS in commercial truck drivers is poor. Thus, OSA screening strategy may need a special approach, including a hierarchical combination of screening tools (Swiss Cheese Model approach), and incorporation of home sleep apnea testing.