137 Recovery Dynamics in a Biomathematical Model of Fatigue. (3rd May 2021)

• Record Type: Journal Article
• Authors: McCauley, Mark; McCauley, Peter; Van Dongen, Hans
• Is Part Of: Sleep. Volume 44(2021)Supplement 2
• Page Start: A56
• Page End: A56
• Publication Date: 2021-05-03
• Subjects: Sleep -- Physiological aspects -- Periodicals; Sleep disorders -- Periodicals; Sommeil -- Aspect physiologique -- Périodiques; Sommeil, Troubles du -- Périodiques; Sleep disorders; Sleep -- Physiological aspects; Sleep -- physiological aspects; Sleep Wake Disorders; Psychophysiology; Electronic journals; Periodicals; 616.8498
• DOI: 10.1093/sleep/zsab072.136
• Languages: English
• ISSNs: 0161-8105

Abstract

Abstract: Introduction: In commercial aviation and other operational settings where biomathematical models of fatigue are used for fatigue risk management, accurate prediction of recovery during rest periods following duty periods with sleep loss and/or circadian misalignment is critical. The recuperative potential of recovery sleep is influenced by a variety of factors, including long-term, allostatic effects of prior sleep/wake history. For example, recovery tends to be slower after sustained sleep restriction versus acute total sleep deprivation. Capturing such dynamics has proven to be challenging. Methods: Here we focus on the dynamic biomathematical model of McCauley et al. (2013). In addition to a circadian process, this model features differential equations for sleep/wake regulation including a short-term sleep homeostatic process capturing change in the order of hours/days and a long-term allostatic process capturing change in the order of days/weeks. The allostatic process modulates the dynamics of the homeostatic process by shifting its equilibrium setpoint, which addresses recently observed phenomena such as reduced vulnerability to sleep loss after banking sleep. It also differentiates the build-up and recovery rates of fatigue under conditions of chronic sleep restriction versus acute total sleep deprivation; nonetheless, it does not accurately predict the disproportionately rapid recovery seen after total sleep deprivation. To improve the model, we hypothesized that the homeostatic process may also modulate the allostatic process, with the magnitude of this effect scaling as a function of time awake. Results: To test our hypothesis, we added a parameter to the model to capture modulation by the homeostatic process of the allostatic process build-up during wakefulness and dissipation during sleep. Parameter estimation using previously published laboratory datasets of fatigue showed this parameter as significantly different from zero (p<0.05) and yielding a 10%–20% improvement in goodness-of-fit for recovery without adversely affecting goodness-of-fit for pre-recovery days. Conclusion: Inclusion of a modulation effect of the allostatic process by the homeostatic process improved prediction accuracy in a variety of sleep loss and circadian misalignment scenarios. In addition to operational relevance for duty/rest scheduling, this finding has implications for understanding mechanisms underlying the homeostatic and allostatic processes of sleep/wake regulation. Support (if any): Federal Express Corporation