Table 2:

Observational Studies measuring associations between caffeinated products and sleep and related characteristics

First Author (Year)	Data Source/Study population	Goals	Measurement/Scales	Findings
CDC (2012)	A random cluster sample of US Service Members on a combat deployment to Afghanistan (N=1,249)	To determine the association between ED consumption and sleep measures	Questions on sleep and ED use embedded within the Deployment Well-Being Survey of the Joint Mental Health Advisory Team (JMHAT) 7.	*About 45% drank ≥1, and 13.9% drank ≥3 ED per day. *Among those who slept 3-4 hrs./ night: - 38.2% drank ≥3 drinks/day, - 18.4% drank 1-2 drinks/day, and - 23.9% abstained from caffeine. * Disrupted sleep > 50% of the nights was related to higher ED use and psychological stress or illness
Jacobson et al (2012)	Millennium Cohort Longitudinal Study (2007-2008, N=106,698)	To determine association between deployment experience, physical activity, and use of body building, energy or weight loss supplements	Self-reported data from the Millennium Cohort Study, collected between 2007-2008.	*About 46.7% of the participants had used ≥1 type of supplements. * 22.0% used multiple supplements. *Deployed males preferred bodybuilding supplements, whereas females preferred weight-loss supplements. *Physically active and younger subjects reported all types of supplement use. *Sleep duration ≤5 hours a night was linked to the use of energy supplements.
Waits et al(2014)	A random sample of service members deployed in Afghanistan 2010-2011	To quantify the sleep habits and energy products use among deployed service personnel.	A survey questionnaire created by the authors and incorporating the Combat Exposure Scale, and the Insomnia Symptom Questionnaire.	*Insomnia was prevalent in 83% of them; *28% were using prescription or over-the-counter sleep aids *81% consumed ≥1 energy product daily. *Use of energy products increased during deployment *There was no relationship between insomnia and the quantity of energy products used.
Matsangas et al (2018)	A sample of US Navy personnel on an aircraft carrier (N=767)	To assess fatigue and sleepiness as a result of working at sea.	A retrospective survey involving data from their semi-annual physical fitness assessment measures, Epworth Sleepiness Scale, and Fatigue Severity Scale.	*The mean±SD sleep duration was 6.12 ± 1.10 hours, and 70% slept <7 hours/day. *Those who were fatigued and sleepy consumed more caffeine than those who were non-fatigued/non-sleepy, and fatigued/non-sleepy.
McLean et al (2017)	A sample of active duty military personnel seeking treatment for PTSD after deployment to Iraq or Afghanistan (N=366)	To examine the relationship between caffeine consumption, sleep disturbances, and PTSD symptoms.	A general health Interview, Insomnia Severity Index, Self-Assessment of Sleep, Epworth Sleepiness Scale, and PTSD Symptom-Scale Interview.	*A higher caffeine use was associated with lower insomnia symptom severity, although insomnia severity was elevated in those who reported no caffeine use. *Caffeine use was not associated with any other sleep disturbance measures or PTSD symptoms.
Shattuck et al (2017)	A sample of US Navy crew members stationed on an aircraft carrier (n=82)	To evaluate the association between actigraphic sleep duration and caffeinated beverage consumption and other variable.	Daily log including activity, caffeine consumption, and exposure to sunlight, Epworth Sleepiness Scale score, and actigraphy, over a 12-day period.	*Their mean actigraphic sleep duration was 6.64±0.95 hours *About 96% of crewmembers reported daily use of a caffeinated beverage *Sleep duration was inversely associated with the number of caffeinated drinks consumed in a bivariate, as well as a multivariable model.
McLellan et al (2018)	Data from two independent surveys administered in Afghanistan. Survey # 1 involved U.S. Army combat soldiers (JMHAT-8, N=518). Survey # 2 involved soldiers deployed with different military assignments (USARIEM, N=260).	To estimate the caffeine consumption, and to examine the relationship between caffeine consumption and sleep quality, combat activities and measures of operational stress, in the combat theater.	The MHAT-8 survey questions included in this study included their demographic data, caffeine consumption, required sleep duration, and reported sleep duration. The USAREIM Deployment Survey also collected demographic data, caffeine consumption, and sleep duration.	*The mean daily caffeine intake in the J-MHAT 8 survey was 404 ± 18 mg, and 80% consumed caffeinated products. *In the JMHAT-8 survey, higher caffeine use was linked to disruptions in their sleep environment, operation tempo, and trouble falling or staying asleep during the day after night operations. *A higher caffeine intake was not associated with DFA, DSA, EMA, trouble waking up, or with being sleep deprived.. *In the USARIEM Deployment survey, intake was 303 ± 29 mg *Caffeine use was significantly higher among combat arms soldiers, than in the combat service support personnel. *Soldiers who slept ≤4 hrs used more caffeine than those who slept >6 hrs.
Toblin et al (2018)	A random sample of active duty U.S Army Soldiers who returned from a 12-month combat deployment to Afghanistan (N=627)	To quantify the prevalence of ED use. The study also examined the association between ED use and post-deployment insomnia, and other mental health problems.	Self-reported survey that collected data on ED use, insomnia (ISI), PCL, PHQ-9, GAD-7, and AUDIT-C scales	*The prevalence of ED use was 75.7%, with 29.2% reporting daily use, and 16.7% consuming high levels of ED (≥2/day) *When compared to low frequency of use, high frequency of ED users were more likely to complain of insomnia and alcohol misuse.

Legend: ED = Energy Drinks; JMHAT = Joint Mental Health Advisory Team; DFSA = Difficulty Falling Asleep; DSA = Difficulty Staying Asleep; EMA = Early Morning Awakening; hrs = hours; PCL = PTSD check list; PHQ-9 = 9-item Patient Health Questionnaire; GAD-7 = 7-item Generalized Anxiety Disorder Scale; UDIT-C = 3-item version of the Alcohol Use Disorder Identification Test;

¹Only study with longitudinal study design. All other studies are cross-sectional.