Caffeine Dosing Strategies To Optimize Alertness During Slumber Loss

Naptime intelligence via Wiley Open Source:

Journal of Sleep Research

First published: 28 May 2018

Summary

Sleep loss, which affects most one‐third of the USA population, tin severely impair physical as well as neurobehavioural performance. Although caffeine, the most widely used stimulant inward the world, tin mitigate these effects, currently at that spot are no tools to guide the timing as well as amount of caffeine consumption to optimize its benefits. In this work, nosotros render an optimization algorithm, suited for mobile computing platforms, to cause upward one's hear when as well as how much caffeine to consume, then equally to safely maximize neurobehavioural functioning at the desired fourth dimension of the day, nether whatever sleep‐loss condition. The algorithm is based on our previously validated Unified Model of Performance, which predicts the number of caffeine consumption on a psychomotor vigilance task. We assessed the algorithm yesteryear comparison the caffeine‐dosing strategies (timing as well as amount) it identified alongside the dosing strategies used inward iv experimental studies, involving total as well as partial slumber loss. Through estimator simulations, nosotros showed that the algorithm yielded caffeine‐dosing strategies that enhanced functioning of the predicted psychomotor vigilance chore yesteryear upward to 64% piece using the same total amount of caffeine equally inward the master copy studies. In addition, the algorithm identified strategies that resulted inward equivalent functioning to that inward the experimental studies piece reducing caffeine consumption yesteryear upward to 65%. Our locomote provides the outset quantitative caffeine optimization tool for designing effective strategies to maximize neurobehavioural functioning as well as to avoid excessive caffeine consumption during whatever arbitrary sleep‐loss condition.

1 INTRODUCTION

Sleep loss, which is a mutual stressor for both civilians as well as armed services personnel, tin severely impair cognitive as well as physical performance, as well as thereby diminish productivity as well as compromise safety. Several studies guide hold demonstrated that, when safely used, caffeine tin aid to sustain cognitive functioning during prolonged periods of restricted slumber (Doty et al., 2017; Kamimori et al., 2015; Killgore et al., 2008; Mclellan, Bell, & Kamimori, 2004; Mclellan et al., 2005; Wesensten, Killgore, & Balkin, 2005). However, these investigations offering caffeine countermeasure guidance that is study‐specific, as well as which cannot hold upward readily adaptable to whatever arbitrary sleep‐loss condition. Providing a foundation for addressing this need, our grouping has previously developed as well as validated a mathematical model, the unified model of functioning (UMP), which tin predict the effects of slumber loss as well as caffeine, equally a component subdivision of fourth dimension of day, on objective measures of neurobehavioural functioning (i.e. the psychomotor vigilance task, PVT) across a broad make of sleep–wake schedules as well as caffeine doses (Ramakrishnan et al., 2013, 2014, 2016). More recently, nosotros guide hold built upon the UMP to educate the open‐access Web tool 2B‐Alert (Reifman et al., 2016), a determination assistance to aid users blueprint slumber studies as well as locomote schedules, as well as the smartphone 2B‐Alert app (Reifman et al., 2017) for real‐time, individualized functioning prediction (Liu, Ramakrishnan, Laxminarayan, Balkin, & Reifman, 2017).

In this work, our destination was to educate a computational tool to provide, inward existent time, effective caffeine‐dosing strategies for whatever arbitrary sleep‐loss condition. Once incorporated into a mobile computing device, such a tool could render customized caffeine‐consumption guidance to, for example, sustain the attending of sleep‐deprived armed services personnel. To this end, using the predictive might of the UMP, nosotros formulated an optimization occupation to cause upward one's hear when as well as how much caffeine to consume, then equally to safely maximize neurobehavioural functioning at the desired fourth dimension of the twenty-four hr catamenia for the desired duration. To solve this problem, nosotros developed an efficient optimization algorithm that was able to observe near‐optimal solutions inward existent time. We assessed the optimization algorithm yesteryear comparison the effects of its predicted caffeine‐dosing (timing as well as amount) strategies alongside those obtained inward iv experimental studies previously used to validate the UMP (Ramakrishnan et al., 2016). In particular, nosotros obtained caffeine‐dosing strategies that enhanced PVT functioning piece using the same total amount of caffeine equally inward the master copy studies, as well as strategies that yielded equivalent levels of functioning equally inward the master copy studies piece reducing caffeine consumption.

2 METHODS
2.1 The unified model of performance

The UMP has 2 components. The first, based on Borbély's two‐process model (Borbely, 1982), describes functioning equally a component subdivision of the circadian wheel as well as a homeostatic process. The minute is a pharmacokinetic as well as pharmacodynamic model, which estimates the caffeine marker inward the blood as well as predicts the duration as well as magnitude of the number of caffeine intake on neurobehavioural performance. For a given sleep–wake schedule as well as caffeine consumption strategy, which constitute the inputs to the model, the UMP predicts the PVT hateful answer fourth dimension (RT) for an ‘average’ individual. We advert the reader to Ramakrishnan et al. (2016) for detailed descriptions of the UMP, the parameter estimation procedure as well as model validation. We guide hold provided the UMP equations as well as parameter values inward Section I of the Supporting Information.

2.2 Optimization problem

Our destination was to observe a caffeine‐dosing strategy that would minimize neurobehavioural functioning impairment based on the PVT for a given sleep–wake schedule. To this end, nosotros sought to minimize the objective component subdivision Z (Equation 1, Table 1), which considers both the expanse nether the UMP‐predicted PVT hateful RT flexure (AUC_C) that is inward a higher identify the baseline, as well as the worst functioning (WP_C) (i.e. the divergence betwixt the meridian of the hateful RT flexure as well as the baseline) (Figure 1). As the baseline hateful RT, nosotros used the highest predicted value of hateful RT when an average private has no slumber debt, wakes upward at 07:00 as well as is awake for 16 hr. We normalized AUC_C as well as WP_C yesteryear the corresponding values for the predicted hateful RT flexure without caffeine consumption, AUC_NC as well as WP_NC, respectively. In addition, nosotros included a penalisation term inward Z to bound the accumulation of caffeine inward the blood [C(t_i, D_i)], which could outcome inward dangerous consumption (Killgore et al., 2008). This term penalizes Z when the maximum marker of caffeine inward the blood is higher than the maximum marker (C_max) achieved yesteryear a unmarried 400‐mg dose (Institute of Medicine, 2001). (Note that the value of C_max tin hold upward readily changed inward the algorithm.) Hence, without considering the penalisation term, Z varies from 0 (for a strategy that consistently maintains the hateful RT below the baseline) to 100 (for a strategy that is no ameliorate than using no caffeine). Therefore, the smaller the value of Z, the ameliorate is the dosing strategy....

...MUCH MORE
Text (HTML)

Or 10 page PDF