0206 Personalized Caffeine Recommendations To Maintain Alertness: You And I Need Different Doses. (12th April 2019)

• Record Type: Journal Article
• Authors: Vital-Lopez, Francisco; Ramakrishnan, Sridhar; Doty, Tracy J; Balkin, Thomas J; Reifman, Jaques
• Is Part Of: Sleep. Volume 42(2019)Supplement 1
• Page Start: A84
• Page End: A85
• Publication Date: 2019-04-12
• 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/zsz067.205
• Languages: English
• ISSNs: 0161-8105

Abstract

Abstract: Introduction: By optimizing caffeine consumption, we could maximize its recuperative benefits during sleep deprivation and reduce its use. However, to date, there are no algorithms to determine how much and when one should consume caffeine to safely maximize alertness, while considering between-individual variability. Methods: To provide individualized caffeine recommendations, we combined two recently developed algorithms: 1) an artificial intelligence algorithm that uses psychomotor vigilance task (PVT) measurements to customize the parameters of a validated alertness-prediction model to an individual's response to sleep deprivation and caffeine, and 2) an optimization algorithm that provides optimal caffeine recommendations to maximize alertness at the desired time, while minimizing caffeine use. We used the resulting algorithm to calculate personalized caffeine recommendations for 21 subjects challenged with 62 h of total sleep deprivation. The algorithm used the first 38 h (12 PVT measurements) to learn how each subject responds to sleep deprivation, and then obtained caffeine recommendations to keep each subject to an alertness level no lower than that achieved by a blood alcohol concentration of 0.08%. To place the advantages of personalizing the recommendations in perspective, we compared them to recommendations obtained for an "average" individual (i.e., a group average prediction). Results: We obtained a wide range of personalized doses: whereas the seven subjects most vulnerable to sleep deprivation required between 500 and 1000 mg of caffeine, the seven most resilient subjects required none. For the vulnerable subjects, following the group-average recommendation instead would have worsened their alertness impairment (by 49 ms, on average, in a 5-min PVT). In contrast, for the seven most resilient subjects, following the same recommendation would have led them to consume caffeine (300 mg on average) not needed to maintain their desired alertness level. Conclusion: We present the first caffeine optimization algorithm that provides personalized guidance for safe and effective use of caffeine to maximize alertness at the most needed times. Support (If Any): This work was sponsored by the Military Operational Medicine Program Area Directorate of the U.S. Army Medical Research and Materiel Command, Ft. Detrick, MD.