Pilot Sleep in Long-Range and Ultra-Long-Range Commercial Flights.

INTRODUCTION: Despite the clear need for understanding how pilot sleep affects performance during long-range (LR; 12-16h) and ultra-long-range (ULR; 16+h) flights, the scientific literature on the effects of sleep loss and circadian desynchronization on pilots' sleep in commercial aviation is sparse.METHODS: We assessed pilots' sleep timing, duration, and post-trip recovery on two LR and two ULR nonstop California to Australasia routes. Pilot's sleep/wake history was measured with actigraphy and verified by logbook across 8-9 d.RESULTS: Pilots averaged 8.210 ± 1.687 SD hours of sleep per 24 h across the study period. A logistic model of the circadian timing of sleep indicated that time of day and phase of trip are significant predictors of pilots being asleep. Significant two- and three-way interactions were found between time of day, phase of trip, and route. A significant difference in average sleep time was observed between baseline and recovery day 1 for one route. All other recovery days and routes were not significantly different from baseline.DISCUSSION: For the four routes, the average amount of sleep per 24-h period during the study period was within the normal range with the circadian rhythm aligned to home-base time pre- and post-trip. Flight segments and layover conditions were associated with a misalignment of sleep relative to circadian rhythm, with layover sleep appearing to shift toward the local night. Full post-trip sleep duration recovery appears to occur for all routes within 1-2 d.Lamp A, McCullough D, Chen JMC, Brown RE, Belenky G. Pilot sleep in long-range and ultra-long-range commercial flights. Aerosp Med Hum Perform. 2019; 90(2):109-115.

Equal to or better than: The application of statistical non-inferiority to fatigue risk management.

In December 2014, the Federal Aviation Administration (FAA) completed a major revision of the rules and regulations governing flight and duty time in commercial aviation (Federal Aviation Regulation (FAR) Part 117). Scientists were included in the revision process and provided insights into sleep, sleep loss, the circadian rhythm, and their effects on performance that were incorporated into the new rule. If a planned flight was non-compliant with the regulation, for example if it exceeded flight and duty time limits, it could only be flown under an FAA-approved Fatigue Risk Management System (FRMS) as meeting an Alternative Method of Compliance (AMOC). One method that a flight could qualify as an AMOC is if it could be demonstrated empirically that it was as safe as or safer than a similar flight, designated the Safety Standard Operation (SSO), that was compliant with the regulation. In the present paper, we demonstrate the FRMS process using a comparison between a non-compliant AMOC flight from the US west coast to Australia and a compliant SSO flight from the US west coast to Taiwan. The AMOC was non-compliant because it exceeded the flight time limits in the prescriptive rule. Once a data collection exemption was granted by the FAA, both the outbound and inbound AMOC and SSO routes were studied on four Safety Performance Indicators (SPIs). The SPIs studied were inflight sleep, cognitive performance, self-reported fatigue, and self-reported sleepiness. These measures were made at top of descent (TOD), a critical phase of flight. The study was designed as a paired comparison. Forty volunteer pilots studied flew both the AMOC and the SSO flights for a total of 80 studied flights. Using statistical non-inferiority applied to the AMOC and SSO SPIs, we demonstrated, as required by the new rule, that the US-Australia AMOC flight was "as safe as, or safer than" the US-Taiwan SSO flight. In the context of FRMS, statistical non-inferiority is a concept and technique of great utility, straightforward in application, producing clear visual representations of the findings, and providing a direct answer to the question posed by the regulation - is the AMOC flight "as safe as, or safer than" the SSO.