Caffeine and energy drink use by combat arms soldiers in Afghanistan as a countermeasure for sleep loss and high operational demands

Abstract

Objectives

Combat deployments are characterized by high operational demands with limited opportunities for sleep leading to fatigue and degraded cognitive and operational performance. Caffeine in moderate doses is recognized as an effective intervention for physical and cognitive decrements associated with sleep loss.

Methods

This report is based on data collected by two separate, independently conducted surveys administered in Afghanistan in 2011–2012. It assessed caffeine use and sleep disruption among U.S. Army combat soldiers (J-MHAT 8; n = 518) and among deployed soldiers with different military assignments (USARIEM Deployment Survey; n = 260).

Results

Daily caffeine intake assessed in the J-MHAT 8 survey averaged 404 ± 18 mg. In the USARIEM Deployment Survey, intake was 303 ± 29 mg and was significantly higher among combat arms soldiers (483 ± 100 mg) compared to combat service support personnel (235 ± 23 mg). In both surveys, over 55% of total caffeine intake was from energy drinks. Additional sources of caffeine included coffee, tea, sodas, gum, candy, and over-the-counter medications. Higher caffeine intake was not associated with ability to fall asleep at night or wake-up in the morning (J-MHAT 8 survey). Higher caffeine consumption was associated with disrupted sleep from high operational tempo and nighttime duties of combat operations.

Discussion

Overall caffeine consumption and energy drink use in Afghanistan was greater than among non-deployed soldiers and civilians. Caffeine was frequently used as a countermeasure during night operations to offset adverse effects of sleep loss on physical and cognitive function, consistent with current Department of the Army recommendations.

Keywords:

• Coffee
• Cola beverages
• Caffeinated gum
• Military
• Stress

Introduction

Caffeine is the most widely consumed psychoactive substance in North America and is popular in many other parts of the world. About 80% of the U.S. population regularly consume caffeine.¹ Many studies have examined the behavioral effects of caffeine, and it is generally agreed that in moderate doses it enhances specific aspects of cognitive performance, including vigilance and attention, as well as decreasing self-reported fatigue (for comprehensive reviews see ^2–5). Whether these effects are present in individuals who are not regular caffeine users has been controversial, and it has been suggested that the beneficial effects of caffeine are solely due to withdrawal from caffeine.⁶ It has been hypothesized that when regular users of caffeine are not consuming it, their mood and performance is degraded and caffeine administration restores these functions to ‘normal levels.’ However, studies of individuals who are not regular consumers have found they also respond positively to caffeine administration.⁷

Combat operations are extremely stressful for deployed U.S. military personnel and are characterized by exposure to austere environments and include physical and cognitive stressors such as ascent to altitude, heavy load carriage, exposure to high ambient temperatures and long periods of reduced sleep.^8,9 Insufficient sleep because of operational requirements impacts both physical and cognitive function as well as marksmanship^10,11 and reduced sleep during deployment is associated with an increased probability of accidents or mistakes that affect mission performance.¹² Caffeine is a safe and effective strategy in many occupational settings to mitigate the physical and cognitive decrements due to restricted sleep (for a recent review, see ⁴), and is recommended by U.S. Army doctrine as a strategy to mitigate the effects of sleep loss on soldier performance, although obtaining adequate sleep is the preferred strategy.¹³ Moreover, the U.S. Department of Defense has developed and tested various caffeine-containing products, including caffeine gum, to assist service members in maintaining alertness for extended periods of time.^11,14–17 However, a concern regarding use of caffeine to mitigate the effects of sleep loss is its potential to affect subsequent sleep quality in certain circumstances.^18–20

Caffeine is found in beverages such as coffee, tea, and energy drinks with the choice of the latter beverage more frequent in younger male military personnel and civilians.^21–24 Recent findings from surveys of non-deployed and deployed U.S. military members suggest that consumption of energy drinks increases during deployment.^24,25 Prevalence for consuming at least 1 energy drink per week is reported to vary from 28 to 38% for non-deployed military service members^23,26 but higher daily rates have been reported by Toblin et al.²⁵ (45%) and Waits et al.²⁴ (81%) during combat deployments in Afghanistan. Some surveys of U.S. military personnel have indicated that consumers of energy drinks report problems with sleep disruption or difficulty falling asleep.^23,25

Combat service support personnel who do not directly participate in ground combat missions and who provide logistical support for military combat arms units may experience fewer episodes of restricted sleep during the night. As such, their caffeine consumption may be lower than combat arms soldiers who are more likely to be engaged in the high tempo, nighttime duties associated with combat and have fewer opportunities for sleep.

To assess caffeine intake in Afghanistan and whether military unit assignment influenced caffeine consumption and sleep, data from two surveys were analyzed. One survey, the Joint Mental Health Advisory Team 8 (J-MHAT 8) survey, was administered in Afghanistan in 2012 to U.S. Army brigade combat team soldiers.²⁷ Unlike previous MHAT surveys, the J-MHAT 8 survey included questions about coffee, tea, and soft drink consumption as sources of caffeine in the diet rather than only questions about energy drink consumption.²⁵ The second survey (USARIEM Deployment Survey) was conducted in 2011 with soldiers who were deployed to Afghanistan from various units and were categorized into the three standard categories of military assignment in the U.S. Army: combat arms, combat support (personnel who provide arms units direct operational support such as security or engineering), and combat service support.

The objectives of this study were to estimate total caffeine consumed from all sources of caffeinated products surveyed (including energy drinks, coffee, tea, sodas, gum, and over-the-counter medications) and examine the relationship between caffeine consumption, combat activities, and measures of operational stress (such as the high operational tempo and nighttime duties) and sleep quality among combat soldiers and those in different unit types (e.g. combat support and combat service support) in the combat theater.

Methods

J-MHAT 8 survey

This survey was approved by the Headquarters U.S. Army Medical Research and Materiel Command Institutional Review Board. The J-MHAT 8 was conducted in Afghanistan in support of Operation Enduring Freedom. During April to July 2012, anonymous paper and pencil surveys were administered in gymnasiums, theaters, classrooms, and dining facilities of forward operating bases, outposts, or joint service stations to soldiers (individuals in the U.S. Army) and Marines (individuals in the U.S. Marine Corps). Local medical brigade personnel were responsible for reading the survey instructions, distributing the participant information sheet, as well as distributing and collecting the questionnaires. Informed consent was determined based on responses to the first question on the survey which asked whether respondents received an information sheet and agreed to allow their data to be used for research purposes. No compensation was offered for participation. The J-MHAT 8 survey was conducted under the larger research protocol, the Land Combat Study, which was approved by the Walter Reed Army Institute of Research Institutional Review Board.

While data were obtained from U.S. Army soldiers (n = 994) and U.S. Marines (n = 369), the data presented in this study were obtained from the Army sample. The initial plan was to use a random cluster-based sampling of maneuver platoons, which were all combat units.²⁷ However, an error on the request resulted in some soldiers from non-maneuver platoons participating. Of the 994 soldiers who completed a questionnaire, 619 were from combat units (the combat maneuver platoons). Comparisons of key variables between the soldiers in the sample and the subset of 409 soldiers from the platoons initially identified on the sampling plan did not reveal any systematic differences. All soldiers in the sample were male because when the J-MHAT 8 was conducted females were restricted from most combat arms assignments. Five hundred and fifty-one (551; 89%) of the 619 combat soldiers provided consent. Of those 551, 518 soldiers had complete data for the questions pertaining to the use of caffeinated beverages and were included in our analyses.

In addition to collecting demographic information such as age, rank, years of service, military occupational specialty, and history of deployment and tobacco use, the survey assessed sleep quality and use of caffeinated beverages. Soldiers indicated how many energy drinks were consumed each day during their current deployment and when not deployed. Questions assessing consumption of 12-ounce (355 mL) servings of coffee, and caffeinated tea and soft drinks were also included. Daily caffeine intake was calculated using 160 mg of caffeine as the caffeine content for each 16-ounce (475 mL) energy drink, 128 mg for a 12-ounce cup of coffee, and 52 mg of caffeine for a cup of tea or soft drink consumed.^3,28,29 For all analyses, soldiers were divided into three categories defined as non-users and very low caffeine users (≤52 mg or the equivalent of 1 cup of tea or soda), moderate users (>52 and <400 mg or up to two energy drinks and one soda, or three cups of coffee) and heavy daily consumers (>400 mg). The use of 400 mg to separate moderate and heavy consumers of caffeine is consistent with a conservative upper limit for caffeine consumption proposed by Nawrot et al.³⁰ and confirmed by Wikoff et al.³¹ for lack of adverse health effects associated with daily caffeine intake.

A series of questions regarding sleep, customized for the unique nature of the combat environment, were included in the survey (Table 1). Indices of sleep quality were obtained directly from responses to questions in the survey with one exception. The extent of sleep deprivation was calculated as the difference between responses to the questions ‘On average, how many hours of sleep do you get per day’ and ‘How many hours of sleep do you need per day in order to feel well-rested.’

Table 1 Prevalence and chi-square associations for various measures of sleep and categories of intake of caffeine (mg) among participants from the Joint Mental Health Advisory Team 8 Survey

Sleep variable	Very light (0–52 mg) (n = 124)	Moderate (>52–399 mg) (n = 196)	Heavy (≥400 mg) (n = 197)	P-value
	% (N)	% (N)	% (N)
Sleep deprived
Yes	56.5 (70)	59.0 (115)	64.1 (125)	.35
No	43.5 (54)	41.0 (80)	35.9 (70)
Sleep disruption more than half the nights due to stress related to combat
Yes	7.3 (9)	5.7 (11)	10.2 (20)	.24
No	92.7 (114)	94.3 (183)	89.8 (176)
Sleep disruption more than half the nights due to stress related to personal life
Yes	9.0 (11)	6.2 (12)	12.8 (25)	.08^†
No	91.0 (111)	93.8 (183)	87.2 (170)
Sleep disruption more than half the nights due to illness
Yes	2.3 (3)	0 (0)	3.6 (7)	NP
No	97.5 (119)	100 (194)	96.4 (188)
Sleep disruption more than half the nights due to sleep environment (noise, light, temperature)
Yes	23.8 (29)	21.0 (41)	34.2 (67)***	.01
No	76.2 (93)	79.0 (154)	65.8 (129)
Sleep disruption more than half the nights due to high operational tempo
Yes	16.5 (20)	23.3 (45)	30.7 (50)	.02*
No	83.5 (101)	76.7 (148)	69.3 (133)
Sleep disruption more than half the nights due to nighttime duties
Yes	29.3 (36)	29.9 (58)	40.1 (77)	.05*
No	70.7 (87)	70.1 (136)	59.9 (115)
Sleep disruption more than half the nights due to leisure activities
Yes	5.7 (7)	4.1 (8)	4.6 (9)	.80
No	94.3 (115)	95.9 (186)	95.4 (186)
Fell asleep sometimes/often while sitting in briefings
Yes	18.0 (22)	12.2 (24)	17.4 (34)	.26
No	82.0 (100)	87.8 (172)	82.6 (161)
Fell asleep sometimes/often while riding in convoys
Yes	13.8 (17)	10.8 (21)	15.8 (31)	.34
No	86.2 (106)	89.2 (174)	84.2 (165)
Fell asleep sometimes/often while standing on guard duty
Yes	31.1 (38)	25.0 (49)***	39.1 (77)***	.01
No	68.9 (84)	75.0 (147)	60.9 (120)
Trouble falling asleep during the day following night operations
Yes	13.2 (16)	15.0 (29)	23.0 (45)	.04*
No	86.8 (105)	85.0 (164)	77.0 (151)
Trouble staying asleep during the day following night operations
Yes	15.7 (19)	15.0 (29)	27.7 (54)***	.01
No	84.3 (102)	85.0 (164)	72.3 (141)
Trouble falling asleep at night
Yes	24.8 (30)	18.7 (36)	25.0 (49)	.26
No	75.2 (91)	81.3 (157)	75.0 (147)
Trouble staying asleep at night
Yes	23.8 (29)	18.7 (36)	24.0 (47)	.38
No	76.2 (93)	81.3 (157)	76.0 (149)
Trouble waking up
Yes	38.0 (46)	42.5 (82)	45.9 (90)	.38
No	62.0 (75)	57.5 (111)	54.1 (106)
Problems waking too early
Yes	17.2 (21)	17.6 (34)	20.1 (39)	.75
No	82.8 (101)	82.4 (159)	79.9 (155)
Had an accident or made a mistake because of sleepiness that affected the mission
Yes	3.3 (4)	4.6 (9)	9.6 (19)	.04*
No	96.7 (119)	95.4 (187)	90.4 (178)

* P ≤ 0.05 but no differences among groups; *** P < 0.001;

^† 0.1 > P > 0.05; NP, not possible to calculate chi-square probability due to low cell counts.

USARIEM Deployment Survey

This survey was approved by the U.S. Army Medical Research and Materiel Command Office of Research Protections Institutional Review Board Office. The sample consisted of 250 deployed U.S. Army military personnel in Afghanistan. The participation rate for the soldiers who were asked to take the survey was approximately 80%. Sites for administering the survey were chosen based on the availability and distribution of soldiers and included individuals stationed in forward operating bases, including those located in austere environments. No compensation was offered for participation.

Prior to administering this anonymous survey, the ‘Dietary Supplement Survey for Deployed Military Personnel’,³² participants were briefed regarding its contents, provided instructions for completing all questions and with an explanation that all information obtained would remain confidential, participation was voluntary, and they were free to withdraw from the study. In addition to providing responses to demographic and lifestyle information, which included sex, age, years of service, rank, military job category (combat arms, combat support, combat service support), tobacco use, education, ratings of health and fitness, soldiers completed one multi-part question that listed 30 types of caffeine-containing products together with amounts and frequency of consumption. Products listed in the survey included coffee, tea, soft drinks, energy drinks, and caffeine-containing gums, candy, and over-the-counter medications. On an open-ended question, participants could include responses from other sources of caffeinated products not described. The survey contained one question that addressed the average number of hours of sleep obtained during a 24-hour period while deployed with eight possible choices in 1-hour increments.

Given the greater detail associated with the 30-part question in the USARIEM Deployment Survey, daily caffeine consumption could be determined for the various sizes of caffeinated beverages and frequency of consumption, using a database of soft drinks, coffees, and energy drinks based on data provided by Caffeineinformer³³ and other sources such as specific product and company websites and previously established generic caffeine values.³⁴ For all analyses, soldiers of the USARIEM Deployment Survey were classified into the same three categories (levels) of caffeine consumption used for J-MHAT 8 survey analyses.

Statistical analyses: J-MHAT 8 and USARIEM Deployment Surveys

Surveys were scanned using ScanTools Plus with ScanFlex (version 6.301; Scantron Corporation, Eagan, MN, U.S.A.) and imported to SPSS (version 20.0; SPSS Inc., Armonk, NY, U.S.A.). Continuous variables were analyzed using a one-way analysis of variance (ANOVA). Bonferroni corrections were applied for multiple comparisons. Wald chi-square tests were used to assess significant differences in mean percentages among multiple characteristic levels. Pearson-product correlations were used to examine the relationship between daily caffeine intake and hours of sleep. A P-value of ≤ 0.05 was considered statistically significant. Logistic regression models were used to examine associations between types and amounts of caffeinated beverages and demographic factors including age, rank, years of service, number of deployments, use of tobacco, and self-rating of health. Results for the logistic regressions are presented as odds ratios (OR) and 95% confidence intervals (95% CI).

Results

J-MHAT 8 survey

Sixty percent (n = 310) of respondents were between 18 and 24 years old, 27% (n = 140) were between 25 and 29 years and 13% (n = 66) were 30 or older. Two respondents did not answer this question. The soldiers averaged (mean ± SD) 4.0 ± 3.6 years of service and 2.8 ± 1.1 deployments. The majority of soldiers were infantry (78%) with artillery (11%), armor scout (7%), and medic (3%) military occupational specialities also represented in the survey. Although medics are not considered to be combat soldiers, they are often exposed to the same conditions as combat soldiers when assigned to combat (maneuver) units.

Over 80% of the soldiers in the sample reported consuming caffeinated beverages, although only 24% reported drinking coffee. In contrast, over 60% of the soldiers indicated they consumed energy drinks. Similarly, over 60% reported drinking tea or soft drinks but the survey did not separate these beverages as sources of caffeine. Daily intake of caffeine from the various beverage sources, as well as total intake, is presented in Table 2. The values presented include both users and non-users of caffeine with caffeine intake of non-users entered as zero. Age, number of deployments, and rank did not affect total daily caffeine intake, although coffee consumed per day was lower among soldiers in the youngest age group compared to the older age groups, less for soldiers with the fewest deployments versus those with multiple deployments, and less for those in the more junior ranks compared to soldiers in the more senior ranks. Total daily caffeine intake was significantly less for soldiers with 1–4 years of service compared to those with 5–9 years of service, which reflected reduced intake from both coffee and energy drinks. Coffee consumption was greatest for soldiers with 10 or more years of service, whereas energy drink consumption was the highest for those with 5–9 years of service. Users of tobacco consumed more total caffeine daily, which reflected higher caffeine intake from all types of beverages. Those soldiers who rated their health as excellent or very good consumed less daily caffeine compared to soldiers who rated their health as good. Although caffeine intake from all beverage sources was lower for this former group, only caffeine from tea and soft drinks was significantly lower compared to those soldiers who did not rate their health as excellent or very good. Daily caffeine intake was not related to average hours of sleep.

Table 2 Reported mean daily intake of caffeine (mg) of participants from the Joint Mental Health Advisory Team 8 Survey by select demographic and lifestyle characteristics

	Total caffeine	Coffee	Energy drinks	Tea/soft drinks
All respondents	404.09 ± 18.03	54.35 ± 4.90	283.69 ± 14.79	64.41 ± 3.11
Age (years)
18–24	368.66 ± 21.6	31.27 ± 4.5^a	275.62 ± 18.7	61.78 ± 4.1
25–29	467.68 ± 38.4	85.65 ± 12.0	310.86 ± 31.0	71.18 ± 6.2
30+	410.36 ± 44.3	88.91 ± 15.6	256.97 ± 33.5	64.48 ± 7.8
Service (years)
1–4	364.50 ± 20.4^b	37.93 ± 4.8^c	265.89 ± 17.3	60.68 ± 3.6
5–9	505.24 ± 42.6	70.27 ± 11.8^d	359.32 ± 35.2^e	75.65 ± 7.6
10+	391.45 ± 56.7	121.89 ± 25.0	202.67 ± 38.9	66.89 ± 10.6
Number of deployments
1	376.73 ± 22.9	35.02 ± 5.0^f	283.12 ± 19.8	58.59 ± 3.9
2	412.94 ± 30.1	74.63 ± 9.9	265.58 ± 23.6	72.73 ± 6.1
≥3	470.68 ± 57.7	89.12 ± 18.2	309.04 ± 46.0	72.52 ± 9.3
Rank
E1–E3	359.42 ± 33.5	27.21 ± 6.4	272.00 ± 28.4	60.20 ± 5.9
E4	405.69 ± 29.6	43.45 ± 7.0	299.78 ± 24.4	62.45 ± 5.0
E5–E9	441.34 ± 32.2	87.40 ± 11.0^g	283.46 ± 26.7	70.48 ± 5.8
WO/officer	348.31 ± 56.0	81.17 ± 24.5	196.36 ± 39.3	70.77 ± 15.9
Tobacco
Non-user	278.68 ± 25.9	34.85 ± 6.3	197.60 ± 22.9	46.23 ± 4.3
User	466.36 ± 22.5^h	64.35 ± 6.6^i	327.64 ± 18.4^h	74.36 ± 4.1^h
Rating of health
Excellent/very good	352.63 ± 22.4^j	46.97 ± 6.3	253.57 ± 18.8	52.09 ± 3.8^k
Good	459.91 ± 32.3	68.08 ± 9.6	315.09 ± 26.0	76.74 ± 5.9
Fair/poor	439.51 ± 52.7	49.32 ± 12.1	305.07 ± 45.9	85.12 ± 9.6

Note: Values are mean ± SE. E1–E3 (enlisted ranks 1–3), E4 (enlisted rank E4), E5–E9 (enlisted ranks 5–9), WO (Warrant Officer).

^a 18–24 age group significantly lower than the 25–29 and ≥30 groups, P < 0.001.

^b 1–4 years of service group significantly different from the 5–9 group, P = 0.003.

^c 1–4 years of service group significantly lower than the 5–9 group, P = 0.015 and from the ≥10 group, P < 0.001.

^d 5–9 years of service group significantly lower than the ≥10 group, P = 0.019.

^e 5–9 years of service group significantly higher than the 1–4 group, P = 0.025 and the ≥10 year group, P = 0.022.

^f One deployment group significantly lower than the 2 deployment group, P = 0.001, and the ≥3 deployment group, P < 0.001.

^g E5–E9 group significantly higher than the E1–E3 group, P < 0.001 and the E4 group, P = 0.001.

^h Users of tobacco significantly higher than non-users, P < 0.001.

ⁱ Users of tobacco significantly higher than non-users, P = 0.004.

^j Excellent/very good group significantly lower than the good group, P = 0.020.

^k Excellent/very good group significantly lower than both good and fair/poor groups, P < 0.001.

Associations between categories of daily caffeine intake and various sleep metrics are presented in Table 1. The amount of caffeine consumed was not associated with whether soldiers were sleep deprived or if their sleep was disrupted due to personal or combat stress, or leisure activities. In addition, caffeine intake was not associated with soldiers’ ability to fall asleep at night or wake in the morning, and caffeine intake was not associated with the frequency of falling asleep during meetings or while riding in convoys. However, significant associations were observed as the amount of caffeine consumed increased and sleep was disrupted due to the sleep environment, the high operational tempo, and nighttime duties. In addition, higher amounts of caffeine consumed were significantly associated with trouble falling or staying asleep following night operations, as well as the incidence of making a mistake because of sleepiness. Although higher caffeine intake was associated with the frequency of sleeping while on guard duty, this effect was not directly proportional to the amount of caffeine consumed since moderate users reported a lower prevalence than non-users or very light users of caffeine.

The results of the logistic regression analyses indicated that greater consumption of coffee was more likely among senior enlisted (OR = 2.66; 95% CI, 1.03–6.84) compared to junior enlisted ranks. Users of tobacco were also more likely to consume coffee (OR = 2.07; 95% CI, 1.24–3.47) versus non-users. Energy drinks were more likely to be consumed by users of tobacco (OR = 2.90; 95% CI, 1.52–3.44) but less likely to be used by those soldiers with 10 or more years of service (OR = 0.30; 95% CI, 0.10–0.90) compared to those with only 1–4 years of service. Tobacco use (OR = 1.68; 95% CI, 1.12–2.51) and good (OR = 1.58; 95% CI, 1.02–2.43) or fair or poor (OR = 1.98; 95% CI, 1.11–3.55) ratings of health were also significant predictors of increased caffeine intake from tea or soft drinks.

Approximately 43% of soldiers reported an increase in their consumption of energy drinks during the deployment, whereas 37% of the soldiers reported their consumption did not change and the remaining 20% reported a decrease in use. There were no demographic factors that were significant predictors of a propensity to change consumption during deployment.

Soldiers were grouped into categories of caffeine consumption (none/light, moderate, and heavy) and logistic regression models were generated to examine whether there were significant associations of demographic and lifestyle characteristics within these groups. The regression models for the non-user/light user and heavy user categories were significant. Age, years of service, number of deployments, and rank were not associated with any of the categories of caffeine consumption. Non-users of tobacco were more likely to be non-users/very light users of caffeine (OR = 2.23; 95% CI, 1.43–3.47) and less likely to be heavy consumers of caffeine (OR = 0.36; 95% CI, 0.23–0.56). Those soldiers who rated their health as good were less likely to be non-users/very light users of caffeine (OR = 0.60; 95% CI, 0.36–0.99) and more likely to be heavy consumers of caffeine (OR, 1.68; 95% CI, 1.09–2.58) compared to those who rated their health as excellent or very good.

USARIEM Deployment Survey

Of the 250 soldiers who completed this survey, 20.8% (n = 52) were females, with 34.8% (n = 87) respondents between 18 and 24 years of age, 21.6% (n = 54) were between 25 and 29 years, and the remaining 43.6% (n = 109) were 30 years or older. Soldiers averaged 9.1 ± 7.4 years of service and 2.1 ± 1.5 deployments. Almost 90% of the soldiers reported consuming a caffeinated product. Coffee and soft drinks were consumed by approximately 60% of respondents, followed by energy drinks (52%), tea (41%), and other products (14%) such as gum, candy, and over-the-counter medications.

Daily caffeine consumption from the various caffeinated products as well as total caffeine intake is presented in Table 3. Sex, years of service, number of deployments, and self-ratings of health and fitness did not affect total daily caffeine consumed or the amount consumed from the various caffeinated products. Combat arms soldiers consumed more caffeine per day compared to combat service support personnel due primarily to their greater use of energy drink products. Tobacco users consumed more caffeine than non-users. Soldiers of higher rank and those between 30 and 39 years of age consumed more coffee than junior enlisted personnel and soldiers 24 years or younger, respectively. Similarly, soldiers with the most education consumed more caffeine in coffee and less in soft drinks than soldiers with the least education. Finally, soldiers who indicated they averaged 4 or fewer hours of sleep consumed more daily caffeine than those soldiers reporting more than 6 hours of sleep.

Table 3 Reported daily intake of caffeine (mg) of participants from the USARIEM Deployment Survey, including non-users, for any caffeine and specific sources of caffeine by select demographic and lifestyle characteristics

	N	Total caffeine	Coffee	Tea	Soft drinks	Energy drinks	Other
All respondents	250	303.4 ± 29.2	93.5 ± 10.1	17.6 ± 2.5	38.0 ± 4.8	131.6 ± 20.7	23.0 ± 9.7
Sex
Male	194	317.5 ± 33.3	91.9 ± 11.6	17.1 ± 3.0	41.6 ± 5.8	150.2 ± 24.9	16.8 ± 5.5
Female	51	273.8 ± 61.0	108.2 ± 21.9	18.7 ± 4.2	26.3 ± 7.1	71.8 ± 33.1	48.9 ± 43.1
Assignment
Combat arms	51	482.6 ± 99.6^a	91.2 ± 27.6	25.0 ± 7.1	36.2 ± 9.7	260.5 ± 67.6^b	69.8 ± 45.1
Combat support	52	320.3 ± 68.6	102.4 ± 24.4	14.9 ± 4.7	41.2 ± 9.6	147.2 ± 52.5	14.5 ± 11.3
Combat service support	147	235.2 ± 22.9	91.1 ± 11.3	15.9 ± 3.0	37.3 ± 6.6	81.3 ± 17.0	9.7 ± 3.2
Age
≤24	87	319.5 ± 57.6	54.7 ± 12.4	19.1 ± 4.5	35.2 ± 8.6	176.7 ± 41.3	34.0 ± 25.9
25–29	54	264.9 ± 50.8	87.8 ± 23.7	16.9 ± 4.8	30.6 ± 7.7	112.6 ± 39.5	17.1 ± 7.8
30–39	65	374.1 ± 63.7	134.6 ± 24.3^c	17.2 ± 3.9	42.9 ± 10.1	153.3 ± 43.4	26.0 ± 12.5
≥40	44	214.3 ± 28.4	116.4 ± 20.4	15.7 ± 7.2	44.8 ± 12.0	33.4 ± 18.6	4.0 ± 3.0
Service (year)
1–4	99	343.9 ± 57.9	84.8 ± 17.8	13.9 ± 2.7	33.5 ± 6.7	172.6 ± 40.1	39.1 ± 23.5
5–9	45	257.2 ± 49.5	76.2 ± 18.6	24.4 ± 6.9	28.2 ± 7.8	112.4 ± 34.4	16.1 ± 8.4
≥10	97	285.4 ± 37.4	115.6 ± 16.0	15.2 ± 3.9	42.2 ± 7.7	100.5 ± 29.3	11.9 ± 56.3
Number of deployments
1	113	295.4 ± 42.9	90.4 ± 15.2	18.0 ± 3.6	37.1 ± 7.5	122.3 ± 27.0	27.6 ± 20.0
2	68	329.1 ± 58.1	101.0 ± 18.7	21.3 ± 5.9	38.2 ± 8.0	147.6 ± 46.6	21.0 ± 9.7
≥3	68	295.2 ± 52.6	92.3 ± 19.7	13.2 ± 3.5	39.3 ± 9.6	132.8 ± 39.7	17.5 ± 9.7
Rank
E1–E4	104	302.3 ± 46.4	69.9 ± 15.4	17.6 ± 3.8	42.0 ± 8.5	162.0 ± 36.6	10.8 ± 5.6
E5–E9	90	336.2 ± 53.4	90.4 ± 15.7	12.0 ± 2.9	38.6 ± 7.7	149.2 ± 35.2	46.0 ± 26.0
WO/officer	55	255.3 ± 41.9	144.6 ± 22.9^d	25.8 ± 7.2	29.7 ± 7.6	46.6 ± 23.5	8.6 ± 5.1
Education
Some HS/HS	55	389.1 ± 72.4	69.8 ± 19.1^e	13.5 ± 3.8	63.1 ± 15.2	224.4 ± 56.9	18.3 ± 11.3
Some college/assoc. degree	122	261.0 ± 36.6	77.9 ± 14.9^f	18.2 ± 3.6	31.1 ± 5.7^g	103.0 ± 20.7	30.8 ± 18.7
Bachelor/graduate degree	73	309.7 ± 54.1	137.2 ± 18.1	19.5 ± 5.3	30.2 ± 6.4^h	109.3 ± 43.3	13.5 ± 8.3
Sleep duration
<4 hours	35	484.8 ± 124.5^i	117.2 ± 33.5	18.5 ± 4.4	60.0 ± 20.0	240.3 ± 87.1	48.8 ± 22.9
5–6 hours	141	290.0 ± 34.6	92.1 ± 12.5	15.7 ± 3.1	40.4 ± 6.1	120.8 ± 25.6	21.9 ± 15.8
>6 hours	72	242.7 ± 38.7	81.9 ± 18.0	21.1 ± 5.6	23.1 ± 5.8	103.4 ± 27.8	13.2 ± 8.3
Tobacco use
User	95	396.7 ± 60.2^j	102.7 ± 18.5	17.0 ± 4.8	48.9 ± 8.8	204.6 ± 44.3^k	23.8 ± 10.2
Non-user	155	246.2 ± 27.2	87.8 ± 11.6	17.9 ± 2.7	31.3 ± 5.5	86.8 ± 18.4	22.5 ± 14.4
General health
Excellent/good	242	299.1 ± 29.0	94.4 ± 10.3	17.6 ± 2.5	38.4 ± 4.9	124.9 ± 20.2	23.7 ± 10.1
Fair/poor	8	432.7 ± 202.6	64.4 ± 33.4	15.1 ± 12.2	21.0 ± 9.0	332.3 ± 204.9	0.0 ± 0.0
General fitness
Excellent/good	208	295.0 ± 30.6	96.3 ± 11.2	18.0 ± 2.8	33.7 ± 4.4	123.8 ± 21.9	23.1 ± 11.4
Fair/poor	42	345.1 ± 79.6	79.3 ± 22.2	15.1 ± 5.2	58.4 ± 18.1	169.9 ± 57.9	22.6 ± 12.9

Note: Values are mean ± SE. E1–E4 (enlisted ranks 1–4), E5–E9 (enlisted ranks 5–9), HS (high school), WO (Warrant Officer).

^a Combat arms significantly greater than combat service support, P = 0.002.

^b Combat arms significantly greater than combat service support, P = 0.002.

^c Personnel age 30–39 significantly greater than personnel ≤24 years of age, P = 0.012.

^d Officers significantly greater than personnel of the E1–E4 rank, P = 0.014.

^e Personnel with some high school/high school significantly less than personnel with a bachelor/graduate degree, P = 0.05.

^f Personnel with some college/associate degree significantly less than personnel with a bachelor/graduate degree, P = 0.035.

^g Personnel with some college/associate degree significantly less than personnel with some high school/high school degree, P = 0.027.

^h Personnel with a bachelor/graduate degree significantly less than personnel with some high school/high school degree, P = 0.043.

ⁱ Personnel who slept less than 4 hours significantly greater than personnel who slept more than 6 hours, P = 0.029

^j Tobacco users significantly greater than non-users, P = 0.011.

^k Tobacco users significantly greater than non-users, P = 0.005.

The Pearson-product correlations found that daily caffeine intake was significantly but inversely related to the hours of sleep reported for combat arms soldiers (r = −0.38, P = 0.004). However, there was no relationship between hours of sleep and daily caffeine consumption for combat support or combat service support personnel.

Logistic regression analyses demonstrated that combat service support personnel were less likely to consume energy drinks (OR = 0.26; 95% CI, 0.12–0.57) than combat arms soldiers. Similarly, combat support personnel were less likely to consume both energy drinks (OR = 0.35; 95% CI, 0.14–0.86) and other caffeinated products such as gum, candy, or over-the-counter medications (OR = 0.13; 95% CI, 0.03–0.61) than combat arms personnel. Those soldiers 30–39 years of age were more likely to consume other caffeinated products (OR = 2.92; 95% CI, 1.02–8.38) than soldiers 24 or younger, and higher ranking soldiers were less likely to consume energy drinks (OR = 0.45; 95% CI, 0.23–0.89) than junior enlisted soldiers. Soldiers who slept 5–6 hours were less likely to consume tea (OR = 0.42; 95% CI, 0.19–0.96) compared to those who slept 4 hours or less. Finally, the use of other caffeinated products such as gum, candy, and over-the-counter medications was less likely for those who slept 5–6 hours (OR = 0.37; 95% CI, 0.14–0.94) or more than 6 hours (OR = 0.33; 95% CI, 0.11–0.96) compared to soldiers who slept 4 hours or less. When soldiers were grouped into the three categories of caffeine consumption, only the regression model for moderate consumption (>52 and <400 mg) was significant, indicating that soldiers with some college education were more likely (OR = 2.09; 95% CI, 1.06–4.11) to be a moderate consumer of caffeine than those soldiers with less education. None of the other demographic or lifestyle factors assessed were associated with the different categories of caffeine consumption.

Discussion

The use of caffeine at appropriate times and doses is an effective strategy to counter the physical and cognitive decrements associated with sleep loss in various occupational settings, such as during military operations, among shift workers in factories, long-haul truck operations, and pilots of airplanes (for a review, see ⁴). The results of the two surveys we analyzed indicate that soldiers in the combat theater of Afghanistan used caffeine to mitigate the consequences of the extremely demanding operational requirements associated with that environment. The findings from the J-MHAT 8 survey suggest soldiers used caffeine to mitigate the inevitable degradation in physical and cognitive function associated with reduced sleep during night operations and the increased operational demands of a combat theater. Caffeine use was significantly greater among soldiers who reported having a poor sleep environment, disrupted sleep due to nighttime duties, and trouble remaining awake while standing guard (Table 1). Furthermore, these soldiers stated they were more likely to have had an accident or made a mistake that affected their mission due to sleepiness. Caffeine consumption assessed by the USARIEM Deployment Survey was greatest among soldiers in the combat arms occupational classification, the soldiers most likely to be directly involved in combat, and caffeine consumption was greater for soldiers sleeping less than 4 hours per day. Therefore, the findings of both surveys are consistent with the U.S. Department of Army guidance that recommends use of caffeine when soldiers are sleep deprived and unable to increase time spent sleeping.¹³

Unlike a previous study of combat soldiers, this study assessed total caffeine intake from caffeinated beverages, such as coffee, tea, and sodas, and caffeinated products such as gum, candy, and over-the-counter medications, not just energy drinks, to quantify total consumption. Daily caffeine intake was high in the current study averaging over 400 mg for combat arms soldiers (483 ± 100 mg; USARIEM Deployment Survey) or soldiers in combat units (404 ± 18 mg; J-MHAT 8 survey) although approximately 10–15% of the soldiers in both surveys reported they did not consume caffeine. A few soldiers reported intakes of over 2000 mg. Daily caffeine intake was higher than reported previously for U.S. military personnel who were largely drawn from non-deployed Army, Navy, and Marine units and did not address the impact of military assignment and combat on caffeine consumption.^22,26

In the current study, average caffeine intake from energy drinks represented 70% of total daily intake for soldiers in combat units assessed by the J-MHAT 8 survey and 55% of total caffeine intake among combat arms personnel participating in the USARIEM Deployment Survey. Over 60% of the combat arms soldiers (USARIEM Deployment Survey) or soldiers in combat maneuver units (J-MHAT 8 survey) reported consuming these drinks. These prevalence rates appear to be higher than those observed in the preceding J-MHAT 7 survey conducted 2 years earlier when 45% of combat maneuver soldiers reported consuming energy drinks.²⁵ Consistent with previous reports, older soldiers, and those of higher rank were more likely to consume caffeine in the form of coffee.^22,26 Tobacco use was significantly associated with caffeine intake, an association reported previously for U.S. military personnel regardless of beverage source, and in multiple civilian surveys.¹

It should be noted that direct comparison of caffeine intake as assessed by the two datasets is not possible since they used somewhat different procedures. The higher intake by combat arms soldiers (483 mg/day) observed in the USARIEM survey appears largely attributable to the ‘other’ category of caffeine-containing products not assessed in the J-MHAT 8 survey (Tables 2 and 3).

Current US Department of Army guidelines recommend the use of caffeine to counter the physical and cognitive degradation associated with sleep loss.^{11,13,15–17} Insufficient sleep due to the high operational tempo of the combat environment is associated with an increased probability of accidents or mistakes that can impact mission performance.¹² Caffeinated gum, candy, and energy bars are available to deployed soldiers in rations or for purchase from military exchanges.³⁵ Combat arms soldiers surveyed by the USARIEM Deployment Survey were over seven times as likely as combat support personnel to consume other caffeinated products, such as caffeinated gum, consistent with the fact they may have limited access to other caffeine-containing products such as coffee. Analyses of the J-MHAT 8 survey dataset from soldiers in combat units also revealed higher daily caffeine intake was related to disruption of sleep due to a poor sleeping environment, high operational tempo, and nighttime duties, suggesting that caffeine was purposely used in combat to sustain physical and cognitive performance during periods of reduced sleep.^16,36

The J-MHAT 8 data demonstrate that there was a significant association between greater caffeine intake and incidence of accidents or mistakes due to sleepiness that impacted mission effectiveness. This is consistent with other findings that soldiers whose sleep is restricted by operational demands sleep less and use more caffeine²⁴ and appears consistent with, and to be in support of, current US Department of Army guidelines.¹³ This view is also consistent with the associations observed between higher daily caffeine intake and disrupted sleep due to nighttime duties and the high operational tempo. Soldiers might voluntarily choose to consume caffeine during night operations to help mitigate the consequences of falling asleep while on duty¹⁷ or a lowered state of arousal on the ability to detect and discriminate targets as friend or foe.^10,37 Higher daily caffeine intake of soldiers in combat units (J-MHAT 8) was also associated with disrupted sleep following night operations, which is consistent with the effect of caffeine on recovery sleep following its use to maintain cognitive function during an overnight period of sustained wakefulness.^19,20,38,39 Clearly, obtaining sufficient sleep would be the preferred option but if this is not possible, due to the demands of combat, then use of caffeine to counter the effects of sleep loss is one strategy consistently documented by studies conducted in various military training scenarios.^{11,15–17,36} An alternative explanation, suggested by Toblin et al.,²⁵ is greater caffeine intake impacts the quality of sleep, thereby increasing daytime sleepiness and subsequent accidents. However, the J-MHAT 8 data presented here show that caffeine intake was not associated with the ability to fall asleep at night, providing evidence that caffeine is not disrupting sleep and is being used as effective countermeasure to mitigate the unavoidable sleep loss that is typically present among combat soldiers.⁴⁰

Limitations and strengths

There are limitations with the current study design. First, for continuity purposes with past J-MHAT surveys, the random cluster sampling approach⁴¹ provides a representative sample of combat maneuver platoons but total sample numbers are limited. Although the analyses from USARIEM Deployment Survey confirmed the increased caffeine consumption among combat arms soldiers, sample numbers also were modest. Second, as with all self-reported survey data, report and recall bias cannot be ruled out. Finally, our estimates of total daily caffeine intake and caffeine consumption are based on a standardized estimate of caffeine content of beverages and standardized serving size, but these can vary.

Despite these limitations, there are several strengths of the present study, including the ability to determine estimates of total caffeine intake from all beverages sources rather than just energy drinks, categorization of caffeine intake into various groupings and analysis of demographic predictors for caffeine use in multivariate analyses. The use of two independent surveys is another strength as is the use of the random probability sampling procedure for the J-MHAT protocol.

Conclusions

Caffeine use, especially by consumption of energy drinks, was greater among soldiers directly involved with combat compared with those providing support roles and the general Army population. The amount of caffeine consumed was associated with sleep disruption due to night duties and high operational tempo as well as disrupted sleep following night operations. Greater use of caffeine appears to be a strategy that is widely employed by combat soldiers to attenuate decrements in physical and cognitive function due to the inevitable sleep loss associated with intense combat operations, including sleep loss associated with nighttime actions.

Disclaimer statements

Contributors None.

Funding This work was supported by the Center Alliance for Nutrition and Dietary Supplement Research of the Defense Medical Research and Development Program and the Military Operational Medical Research Program of the U.S. Army Medical Research and Materiel Command.

Conflict of interest There are no conflicts of interest to disclose by any author of this work.

Ethics approval The surveys were approved by Human Use/Institutional Review Boards.

Acknowledgements

The investigators have adhered to the policies for the protection of human subjects as prescribed in DOD Instruction 3216.02 and the research was conducted in adherence to the provisions of 32 CFR Part 219. The opinions or assertions contained herein are the private views of the author(s) and are not to be construed as official or as reflecting the views of the Army or the Department of Defense. Citations of commercial organizations and trade names in this report do not constitute an official Department of the Army endorsement or approval of the products or services of these organizations.

K.W. Williams was supported by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and US Army Medical Research and Materiel Command. T.M. McLellan was supported by the Oak Ridge Institute for Science and Education, as well as through a consulting agreement with the Henry M. Jackson Foundation for the Advancement of Military Medicine Inc.