Available via license: CC BY 4.0
Content may be subject to copyright.
Caffeine consumption among active duty United States Air Force
personnel
Joseph J. Knapik
a
,
b
,
*
, Krista G. Austin
a
,
b
, Susan M. McGraw
a
, Guy D. Leahy
c
,
Harris R. Lieberman
a
a
Military Nutrition Division, US Army Research Institute of Environmental Medicine, Natick, MA 01760, United States
b
Oak Ridge Institute for Science and Education, Belcamp, MD 21017, United States
c
Health Promotion Flight/Aerospace Medicine Squadron, Kirtland Air Force Base, Albuquerque, NM 87117, United States
article info
Article history:
Received 12 January 2017
Received in revised form
27 April 2017
Accepted 30 April 2017
Available online 3 May 2017
Keywords:
Coffee
Tea
Cola
Soda
Energy drink
abstract
Data from the National Health and Nutrition Examination Survey (NHANES) indicated that 89% of
Americans regularly consumed caffeinated products, but these data did not include military personnel.
This cross-sectional study examined caffeine consumption prevalence, amount of daily consumption, and
factors associated with caffeine intake in active duty United States (US) Air Force personnel. Service
members (N ¼1787) stationed in the US and overseas completed a detailed questionnaire describing
their intake of caffeine-containing products in addition to their demographic, lifestyle, and military
characteristics. Overall, 84% reported consuming caffeinated products 1 time/week with caffeine
consumers ingesting a mean ±standard error of 212 ±9 mg/day (224 ±11 mg/day for men, 180 ±12 mg/
day for women). The most commonly consumed caffeinated products (% users) were sodas (56%), coffee
(45%), teas (36%), and energy drinks (27%). Multivariate logistic regression modeling indicated that
characteristics independently associated with caffeine consumption (1 time/week) included older age,
ethnicity other than black, tobacco use, less aerobic training, and less sleep; energy drink use was
associated with male gender, younger age, tobacco use, and less sleep. Compared to NHANES data, the
prevalence of caffeine consumption in Air Force personnel was similar but daily consumption (mg/day)
was higher.
Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.
org/licenses/by/4.0/).
1. Introduction
Caffeine is a mildly psychoactive substance that is widely
consumed. Data from nationally representative samples indicated
that about 89% of American adults consume caffeinated products
with virtually no difference between men and women in how
frequently the products are consumed (Fulgoni et al., 2015; Mitchell
et al., 2014). Comprehensive reviews conducted for the United
States (US) Department of Agriculture (USDA) and Health Canada
have concluded that consumption of caffeine <400 mg/day is
generally safe and may even confer some health benefits (Nawrot
et al., 2003; USDA, 2015). The USDA commissioned report indi-
cated that in healthy adults there was evidence that moderate
coffee consumption was associated with reduced risk of
cardiovascular disease, liver and endometrial cancers, Type 2 dia-
betes, Parkinson's disease, and overall mortality (USDA, 2015). Both
the USDA and Health Canada commissioned reports recommended
lower daily caffeine consumption for pregnant women, <300 mg in
the Canadian report and 200 mg/day in the USDA report. Some
concern was expressed in the USDA report because of the increased
consumption of energy drinks by young adults (USDA, 2015). Data
from the National Health and Nutrition Examination Survey
(NHANES) indicated a trend of increased consumption of caffeine
from energy drinks from 2001 to 2010 in 19e22 year olds, although
this was largely offset by a reduction in caffeine from sodas so that
there was little change in overall caffeine consumption (Branum
et al., 2014).
Investigations that have obtained representative data on
caffeine intake in Americans (Branum et al., 2014; Drewnowski and
Rehm, 2016; Frary et al., 2005; Fulgoni et al., 2015; Mitchell et al.,
2014) have not provided information on US military service
members (SMs). Air Force personnel have a number of physically
*Corresponding author. Research Physiologist, USARIEM, 10 General Greene Ave,
Natick, MA 01760, USA.
E-mail address: joseph.j.knapik.ctr@mail.mil (J.J. Knapik).
Contents lists available at ScienceDirect
Food and Chemical Toxicology
journal homepage: www.elsevier.com/locate/foodchemtox
http://dx.doi.org/10.1016/j.fct.2017.04.050
0278-6915/Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Food and Chemical Toxicology 105 (2017) 377e386
and cognitively demanding tasks that include intelligence gath-
ering, tactical planning, search and rescue, air traffic control, and
combat actions that can require long periods of complex and time-
intense operations. Other physical demands include early morning
physical training and limited sleep during training, operations, and
deployments. These tasks may lead SMs to consume more
caffeinated substances than the general population. Previous
studies have investigated caffeine consumption in Army
(Lieberman et al., 2012), Navy, and Marine Corps personnel (Knapik
et al., 2016) but studies have not been conducted in the Air Force to
date. The purpose of this report was to complete the examination of
US military personnel by examining the caffeine consumption
prevalence, daily consumption, and characteristics associated with
intake among Air Force personnel.
2. Materials and methods
This investigation was a cross-sectional survey of caffeine con-
sumption among US active-duty Air Force personnel approved by
the institutional review board of the US Army Institute of Envi-
ronmental Medicine. Investigators and adhered to US Army regu-
lation 70-25 and US Army Medical Research and Material
Command Regulation 70-25 on the use of volunteers in research.
Questionnaires were distributed to SMs at ten US and two
overseas installations. Survey sites were selected based on the
availability of health care professionals available to assist with
questionnaire administration. Individuals in basic training, on
leave, in transition to another duty station, and/or incarcerated
were not surveyed. No incentives were offered to SMs for
completing the questionnaire. Prior to administration, participants
were briefed by healthcare providers and told the questionnaire
was anonymous, participation was voluntary, and all information
would remain confidential. Data were collected in 2010 and 2011.
2.1. Survey (questionnaire) description
The first section of the questionnaire was designed to charac-
terize participants. Questions included items on demographic
(gender, age, height, weight, marital status, race/ethnicity, educa-
tional level), lifestyle (tobacco use, frequency and duration of aer-
obic training, frequency of resistance training, and average hours of
sleep per night), and military (time in service, pay grade [rank],
special operations status) characteristics. This descriptive section
was followed by a section listing 31 types of caffeine-containing
substances, including coffee, teas, soft drinks (sodas), energy
drinks, and caffeinated gums and medications. The survey used a
standard food frequency questionnaire method in which SMs
selected among names of commonly consumed products or could
write in caffeine-containing products not listed. SMs were asked to
provide the number of times they consumed a product (per day,
week, month) in the last 6 months and the serving size. Serving
sizes were as follows: for coffee, teas, and soft drinks, 8, 12, 16, 20
and 24 þfluid ounces; for energy drinks, the number of cans or
bottles with serving size listed with the product; for gums and
medications, sticks of gum or number of pills. This questionnaire
was the same one used in previous studies of Army, Navy, and
Marine Corps personnel (Knapik et al., 2016; Lieberman et al.,
2012).
2.2. Data analysis
Caffeine consumers were defined as those using any caffeinated
product 1 time/week. Caffeine consumption (mg/day) was
calculated based on the self-reported product listed by consumers,
its serving size, and frequency of consumption. Sources of
information on caffeine content of specific products included
product and company websites and a database of caffeine content
in coffees, teas, sodas, and energy drinks (Caffeine Informer, 2015).
For generic coffee, tea, and cola, the values in Knight et al. were
used (Knight et al., 2004). Body mass index (BMI) was calculated as
weight/height
2
(kg/m
2
). Weekly duration of aerobic training (mi-
nutes/week) was calculated by multiplying weekly exercise fre-
quency (sessions/week) by the duration of training (minutes/
session). Tobacco users were defined as those reporting use of any
tobacco products in the last week while former users were those
who reported they used tobacco products in the past but had quit
within the last year or earlier.
Statistical analysis was conducted using the Statistical Package
for the Social Sciences (SPSS) (Version 21.0.0.0, 2012, IBM Corpo-
ration). Caffeine products were grouped into 7 categories that
included: 1) coffee, 2) hot tea, 3) other tea-based beverages, 4)
colas, 5) other sodas, 6) energy drinks, and 7) caffeinated gums and
medications. For some analyses, hot tea and other tea-based bev-
erages were combined, as were colas and other sodas. All categories
were combined to arrive at an aggregated caffeine intake (i.e., any
caffeine consumption). Definitions of the caffeine categories are
provided in Table 1.
Prevalences of consumption 1 time/week (%) with standard
errors (SE) were calculated. Chi-square statistics were used to
examine prevalence differences across various strata of de-
mographic (sex, age, education level, marital status, race/ethnicity,
BMI), lifestyle (tobacco use weekly duration of aerobic training,
frequency of resistance training, sleep duration) and military (time
in service, rank, special operations status) characteristics. A one-
way analysis of variance (ANOVA) was used to examine differ-
ences in daily average caffeine consumption (mg/day) across strata
of these characteristics. Since some participants did not complete
all of the questions, the number of participants is shown for each
variable. Multivariate logistic regression was used to examine as-
sociations between the dependent variable “caffeine consumer”
(1 time/week) and independent variables that included the de-
mographic (sex, age, education level, marital status, race/ethnicity,
and BMI), and lifestyle (tobacco use, weekly duration of aerobic
training, frequency of resistance training, and sleep duration)
characteristics. Six separate regression models were developed for
specific caffeine sources including any caffeine, coffee, tea (hot and
other teas combined), soda (colas and other soda combined), en-
ergy drinks, and caffeinated gums and medications. A one-way
ANOVA compared caffeine consumption across age groups in men
and women separately.
Demographic data on the entire population of Air Force
personnel were obtained from Defense Medical Epidemiological
Database (DMED). These data included sex, age, marital status, and
race. These demographics were compared to those of the volun-
teers in this study to examine representativeness of the sample.
3. Results
The final sample included 1787 active duty Air Force personnel
with 1323 reporting they were male, 437 reporting female, and 27
who did not report their gender. The mean ±standard deviation
age, height, weight, and BMI of the men was 28 ±7 years,
179 ±7 cm, 84 ±13 kg and 26.2 ±3.4 kg/m
2
, respectively; for
women these values were 29 ±8 years, 164 ±8 cm, 67 ±11 kg, and
24.8 ±3.8 kg/m
2
, respectively.
3.1. Caffeine-containing product prevalence
Table 2 provides prevalence of reported caffeine consumption
by demographic, lifestyle, and military characteristics. Overall, 84%
J.J. Knapik et al. / Food and Chemical Toxicology 105 (2017) 377e386378
Table 1
Caffeine categories in studies of active duty military personnel.
Category Definition
Any caffeine All caffeine-containing beverages, gums, and medications as listed below
Coffee Hot or cold brewed coffee, espresso, cappuccino, frozen blended coffee drinks, and other coffee-based beverages that contain caffeine
Hot tea
a
Hot brewed tea of any type that contains caffeine
Other tea
a
Other teas including iced tea and cold tea blends that include caffeine
Cola
b
All brands of cola-type beverages that contain caffeine
Other soda
b
Sodas that are not colas but are carbonated and contain caffeine including root beers, orange soda, and other flavored sodas
Energy drink All beverages labeled as energy drinks of any kind that contain caffeine
Gum or medication Chewing gums, prescription medications, weight control aids, and other over-the-counter medications that contain caffeine
a
Hot tea and other tea were combined for some analyses.
b
Cola and other soda were combined for some analyses.
Table 2
Prevalence (% ±SE) of reported caffeine consumption (1 time/week) among Air Force personnel by demographic, lifestyle, and military characteristics (p-values from chi-
square analyses).
Variable Strata Any
caffeine
Coffee Hot tea Other tea Cola Other
soda
Energy
drink
Gum or
medication
Group All (n¼1787) 83.8 ±0.9 45.3 ±1.2 12.1 ±0.8 29.3 ±1.1 50.6 ±1.2 18.7 ±0.9 27.2 ±1.1 5.4 ±0.5
Gender Men (n¼1323)
Women (n¼437)
84.1 ±1.0
83.5 ±1.8
43.8 ±1.4
50.8 ±2.4
9.6 ±0.8
19.9 ±1.9
30.0 ±1.3
27.7 ±2.1
52.5 ±1.4
45.5 ±2.4
20.7 ±1.1
12.4 ±1.6
29.9 ±1.3
19.5 ±1.9
4.7 ±0.6
7.8 ±1.3
p-value 0.77 0.01 <0.01 0.36 0.01 <0.01 <0.01 0.01
Age 18-24 years (n¼652)
25-29 years (n¼507)
30-39 years (n ¼443)
40 years (n¼185)
79.9 ±1.6
83.6 ±1.6
86.5 ±1.6
91.9 ±2.0
34.5 ±1.9
46.2 ±2.2
54.6 ±2.4
58.4 ±3.6
11.5 ±1.2
12.6 ±1.5
10.6 ±1.5
16.2 ±2.7
29.9 ±1.8
27.4 ±2.0
26.6 ±2.1
38.4 ±3.6
46.2 ±2.0
50.5 ±2.2
54.6 ±2.4
57.3 ±3.6
23.8 ±2.0
16.6 ±1.7
16.7 ±1.8
11.4 ±2.3
29.6 ±1.7
32.1 ±2.1
24.4 ±2.0
11.9 ±2.4
4.4 ±1.8
5.3 ±1.0
6.3 ±1.2
6.5 ±1.8
p-value <0.01 <0.01 0.24 0.02 0.01 <0.01 <0.01 0.51
Education Some high school/High school graduate (n¼284)
Some college/Associate's degree (n ¼1064)
Bachelor's/Graduate degree (n ¼419)
82.4 ±2.3
83.0 ±1.2
86.9 ±1.6
31.0 ±2.7
43.3 ±1.5
60.4 ±2.4
9.5 ±1.7
10.6 ±1.0
17.2 ±1.8
33.1 ±2.8
29.4 ±1.4
26.3 ±2.2
52.5 ±3.0
49.6 ±1.5
51.6 ±2.4
27.5 ±2.7
18.5 ±1.2
12.9 ±1.6
32.0 ±2.8
30.1 ±1.4
17.2 ±1.8
4.6 ±1.2
6.3 ±0.7
3.8 ±0.9
p-value 0.15 <0.01 <0.01 0.15 0.62 <0.01 <0.01 0.13
Marital status Single (n¼840)
Married (n¼945)
82.1 ±1.3
85.3 ±1.1
39.9 ±1.7
50.2 ±1.6
12.0 ±1.1
12.1 ±1.1
28.7 ±1.6
29.7 ±1.5
49.6 ±1.7
51.5 ±1.6
21.2 ±1.4
16.4 ±1.2
28.2 ±1.6
26.3 ±1.4
4.6 ±0.7
6.0 ±0.8
p-value 0.07 <0.01 0.98 0.63 0.43 0.01 0.38 0.19
Race/ethnicity White (n¼1150)
Black (n¼236)
Hispanic (n ¼231)
Other (n¼156)
86.2 ±1.0
71.6 ±2.9
81.8 ±2.5
88.5 ±2.6
47.8 ±1.5
29.7 ±3.0
45.5 ±3.3
51.3 ±4.0
11.9 ±1.0
11.9 ±2.1
12.1 ±2.1
12.8 ±2.7
29.8 ±1.3
27.5 ±2.9
26.4 ±2.9
30.1 ±3.7
53.2 ±1.5
38.1 ±3.2
49.4 ±3.3
51.3 ±4.0
20.5 ±1.2
17.4 ±2.5
13.4 ±2.2
14.7 ±2.8
27.8 ±1.3
21.6 ±2.7
29.9 ±3.0
27.6 ±3.6
6.2 ±0.7
3.0 ±1.1
3.9 ±1.3
5.1 ±1.8
p-value <0.01 <0.01 0.99 0.69 <0.01 0.04 0.19 0.16
Body mass index <25.0 kg/m
2
(n¼712)
25.0e29.9 kg/m
2
(n¼830)
30.0 kg/m
2
(n¼225)
83.7 ±1.4
84.9 ±1.2
80.9 ±2.6
44.9 ±1.9
47.7 ±1.7
37.8 ±3.3
14.3 ±1.3
10.7 ±1.1
10.7 ±2.1
30.3 ±1.7
28.9 ±1.6
27.6 ±3.0
50.0 ±1.9
51.2 ±1.7
51.1 ±3.3
21.9 ±1.6
16.6 ±1.3
16.4 ±2.5
26.3 ±1.7
28.9 ±1.6
24.9 ±2.9
4.4 ±0.8
5.4 ±0.8
8.0 ±1.8
p-value 0.33 0.03 0.07 0.68 0.89 0.02 0.34 0.10
Tobacco use Current (n ¼399)
Former (n¼416)
Never (n¼969)
90.2 ±1.5
86.1 ±1.7
80.3 ±1.3
52.6 ±2.5
53.1 ±2.4
38.9 ±1.6
10.5 ±1.5
13.0 ±1.6
12.4 ±1.1
33.1 ±2.4
29.3 ±2.2
27.7 ±1.4
55.6 ±2.5
48.6 ±2.5
49.4 ±1.6
25.1 ±2.2
14.7 ±1.7
17.9 ±1.2
40.1 ±2.5
29.8 ±2.2
20.7 ±1.3
6.3 ±1.2
6.0 ±1.2
4.7 ±0.7
p-value <0.01 <0.01 0.52 0.13 0.07 <0.01 <0.01 0.43
Aerobic exercise
duration
<136 min/wk (n¼460)
136-225 min/wk (n¼532)
226-345 min/wk (n¼363)
346 min/wk (n¼426)
88.3 ±1.5
85.5 ±1.5
82.6 ±2.0
78.2 ±2.0
51.1 ±2.3
50.8 ±2.2
43.3 ±2.6
38.5 ±2.4
11.7 ±1.5
12.2 ±1.4
11.8 ±1.7
12.4 ±1.6
30.4 ±2.1
29.7 ±2.0
27.8 ±2.4
29.1 ±2.2
56.5 ±2.3
53.2 ±2.2
49.7 ±2.6
42.3 ±2.4
22.2 ±2.0
17.7 ±1.7
17.1 ±2.0
17.6 ±1.8
27.8 ±2.1
23.9 ±1.8
29.5 ±2.4
28.9 ±2.2
5.4 ±1.1
5.5 ±1.0
6.3 ±1.3
4.5 ±1.0
p-value <0.01 <0.01 0.99 0.87 <0.01 0.17 0.20 0.71
Resistance training
frequency
<2 sessions/wk (n¼474)
2 sessions/wk (n¼1307)
86.9 ±1.6
82.8 ±1.0
47.9 ±2.3
44.2 ±1.4
12.2 ±1.5
12.0 ±0.9
29.3 ±2.1
29.4 ±1.3
56.3 ±2.3
48.7 ±1.4
19.2 ±1.8
18.5 ±1.1
23.8 ±2.0
28.2 ±1.2
4.4 ±0.9
5.7 ±0.6
p-value 0.04 0.17 0.90 0.98 <0.01 0.74 0.05 0.28
Sleep duration 4 h/night (n¼79)
5-6 h/night (n¼790)
7 h/night (n¼861)
90.0 ±3.4
87.1 ±1.2
81.3 ±1.3
48.2 ±5.6
45.6 ±1.8
46.5 ±1.7
8.9 ±3.2
11.6 ±1.1
13.0 ±1.1
34.2 ±5.3
31.4 ±1.7
27.1 ±1.5
53.2 ±5.6
52.3 ±1.8
49.2 ±1.7
24.1 ±4.8
19.9 ±1.4
17.3 ±1.3
43.0 ±5.6
31.4 ±1.7
22.6 ±1.4
12.7 ±3.7
5.7 ±0.8
4.1 ±0.7
p-value <0.01 0.35 0.46 0.10 0.43 0.19 <0.01 <0.01
Time in service 1 year (n¼386)
2-5 years (n¼527)
6-12 years (n¼474)
13 years (n¼400)
80.1 ±2.0
82.0 ±1.7
84.4 ±1.7
89.3 ±1.5
38.1 ±2.5
38.3 ±2.1
48.7 ±2.3
57.3 ±2.5
8.8 ±1.4
13.9 ±1.5
11.2 ±1.5
14.0 ±1.7
27.7 ±2.3
30.2 ±2.0
26.6 ±2.0
32.8 ±2.3
45.9 ±2.5
47.4 ±2.2
55.1 ±2.3
57.0 ±2.5
25.9 ±2.2
19.2 ±1.7
14.3 ±1.7
13.0 ±1.7
29.8 ±2.3
29.6 ±2.0
30.4 ±2.1
17.8 ±1.9
3.9 ±1.0
5.5 ±1.0
6.1 ±1.1
5.8 ±1.2
p-value <0.01 <0.01 0.07 0.20 <0.01 <0.01 <0.01 0.51
Rank Junior enlisted (n¼729)
Senior enlisted (n¼813)
Junior officer (n¼167)
Senior officer (n¼68)
81.5 ±1.4
84.7 ±1.3
84.4 ±2.8
95.6 ±2.5
35.8 ±1.8
49.0 ±1.8
56.9 ±3.8
75.0 ±5.3
10.3 ±1.1
11.6 ±1.1
19.8 ±3.1
20.6 ±4.9
30.3 ±1.7
29.2 ±1.6
26.3 ±3.4
26.5 ±5.4
45.1 ±1.8
55.1 ±1.7
55.1 ±3.8
41.2 ±6.0
24.7 ±1.6
15.1 ±1.3
14.4 ±2.7
5.9 ±2.9
32.5 ±1.7
26.1 ±1.5
16.2 ±2.9
13.2 ±4.1
5.1 ±0.8
6.3 ±0.9
3.6 ±1.4
2.9 ±2.0
p-value 0.02 <0.01 <0.01 0.72 <0.01 <0.01 <0.01 0.36
Special operations
qualified
Yes (n¼54)
No (n¼1660)
85.2 ±4.8
84.2 ±0.9
37.0 ±6.6
45.8 ±1.2
5.6 ±3.1
12.4 ±0.8
33.3 ±6.4
29.2 ±1.1
44.4 ±6.8
51.1 ±1.2
22.2 ±5.7
18.7 ±1.0
42.6 ±6.7
26.8 ±1.1
1.9 ±1.9
5.5 ±0.6
p-value 0.84 0.20 0.13 0.51 0.34 0.52 0.01 0.24
J.J. Knapik et al. / Food and Chemical Toxicology 105 (2017) 377e386 379
of participants reported using products containing caffeine 1 time
per week, with cola-type beverages and coffee being the most
popular. When colas and other sodas were combined, preva-
lence±SE was 56.4 ±1.2%; when hot and other teas were combined,
prevalence was 36.0 ±1.1%.
A larger proportion of men reported drinking colas, other sodas,
and energy drinks, while a larger proportion of women reported
consuming coffee, hot teas, and caffeinated gums/medications. The
proportion of SMs using any caffeinated product increased with
age, especially for coffee, colas, and other sodas. Consumption of
other teas were higher in older individuals (40 years) while intake
of energy drinks was higher in younger individuals (<40 years). The
proportion of SMs consuming coffee and hot tea increased as
educational level increased, while the proportion of SMs
consuming other sodas and energy drinks increased as educational
level decreased. A larger proportion of married individuals
consumed coffee, while a larger proportion of single individuals
consumed other sodas. SMs of black race/ethnicity had the lowest
intake of any caffeinated products, especially for coffee and colas,
while Hispanics and others were least likely to consume other non-
cola sodas. Lower BMI was associated with a greater consumption
of coffee and other non-cola sodas.
A larger proportion of current tobacco users reported con-
sumption of any caffeine, especially for other sodas and energy
drinks, while a larger proportion of both current and former
smokers were coffee consumers. As aerobic exercise duration
increased, prevalence of consumption for any caffeine, coffee, and
colas declined. Those performing resistance training 2 times/wk
had lower intake of any caffeinated products and colas, but higher
consumption of energy drinks. As sleep duration decreased, con-
sumption of any caffeine, energy drinks, and caffeinated gums and
medications increased.
As time in service increased so did consumption of any caffeine,
coffee, colas, other sodas. Those with 12 years of service were
more likely to consume energy drinks than those with more time in
service. Senior officers had higher overall caffeine intake, while
both senior and junior officers had a higher consumption of coffee
and hot tea compared to enlisted personnel. Senior enlisted
personnel and junior officers consumed more colas while ingestion
of other sodas and energy drinks declined as rank increased. Sol-
diers who were special operations qualified were more likely to
consume energy drinks.
3.2. Caffeine containing-product consumption
Table 3 provides the estimated daily consumption (mg/day)
among caffeine consumers (1 time/week) by their demographic,
lifestyle, and military characteristics. The average daily caffeine
consumption for consumers was 212 mg/day with coffee, teas (hot
and others), soda (colas and others), energy drinks, and caffeinated
gums/medications accounting for 37%, 14%, 16%, 30%, and 4% of
caffeine consumption, respectively. For men, these values were
33%, 13%, 17%, 33% and 3%, respectively; for women, these values
were 49%, 15%, 13%, 17%, and 5%, respectively. About 11.2% of
caffeine consumers had an overall intake 400 mg/day (12.1% of
men and 8.5% of women); 18.5% had an overall consumption
300 mg/day (20.1% of men and 13.7% of women); and 33.7% had
an overall consumption 200 mg/day (34.7% of men and 30.7% of
women). The types of products ingested by the higher caffeine
consumers differed from those of the entire group: among the male
higher consumers (400 mg/day), coffee, teas (hot and others),
sodas (colas and others), energy drinks, and caffeinated gums/
medications accounted for 29%, 11%, 10%, 47%, and 3% of caffeine
consumption, respectively; among female higher caffeine con-
sumers (300 mg/day), these values were 51%,10%, 9%, 25% and 5%,
respectively; among female higher caffeine consumers (200 mg/
day), these values were 52%, 13%, 10%, 20% and 5%, respectively.
Men consumed more total caffeine than women due to a greater
consumption from other tea, cola, non-cola soda and energy drinks;
women consumed more caffeine from hot tea. When total caffeine
consumption was determined based on a body weight basis, con-
sumption was similar among men and women (2.66 vs.
2.76 mg day
1
kg
1
, respectively, p ¼0.66). Caffeine consumption
from coffee increased with age and compared to older SMs (>40
years) younger SMs consumed more caffeine from energy drinks.
Caffeine consumption from coffee increased as education level
increased, but caffeine consumed from other tea, non-cola soda,
and energy drinks were higher among those of lower education
levels. Married SMs consumed more caffeine from coffee but not
from other caffeinated substances. SMs reporting black race/
ethnicity consumed less caffeine from coffee than those of other
race/ethnicities, while those of Hispanic and other ethnicities
consumed less caffeine from non-cola sodas than those of white
and black ethnicity. Total caffeine consumption increased as BMI
increased, due largely to increased consumption from coffee and
energy drinks; gum/medication.
Current and former smokers had greater overall caffeine con-
sumption than those who never smoked, largely accounted for by
greater consumption of coffee; current smokers consumed more
caffeine from colas and other sodas. SMs reporting more aerobic
exercise consumed less caffeine from coffee than those performing
less aerobic exercise. Resistance training frequency was not related
to caffeine consumption. As sleep duration decreased, there was
increased consumption of caffeine from other teas, colas, and other
sodas.
Those with longer time in service consumed more caffeine
overall due primarily to greater coffee consumption; those with
2e5 years in service consumed less caffeine from other teas
compared to other service lengths. Senior enlisted and senior of-
ficers consumed more caffeine overall than junior officers and
enlisted. Coffee consumption increased with rank while con-
sumption from other sodas decreased with rank. Special operations
status was not associated with any difference in caffeine
consumption.
3.3. Characteristics independently associated with caffeine
consumption
Table 4 presents results of the multivariate logistic regression
examining factors associated with consuming caffeine 1 time per
week. The results are shown for the full model with all character-
istics entered. Characteristics associated with higher overall
caffeine consumption included older age, race/ethnicity other than
black, current or former tobacco use, engaging in less aerobic ex-
ercise, and sleeping less. Higher coffee intake was independently
associated with female gender, older age, higher education level,
race/ethnicity other than black, lower BMI, former or current to-
bacco use, and moderate aerobic exercise duration (136e225 min/
wk). Greater consumption of tea was associated with older age,
lower BMI, and less sleep. Higher soda intake was independently
associated with male gender, older age, race/ethnicity other than
black, current tobacco use, and less aerobic exercise. Consuming
more energy drinks was associated with being male, younger age,
current or former tobacco use, and less sleep. Caffeinated gum/
medication intake was independently associated with female
gender, race/ethnicity other than black, and less sleep.
3.4. Caffeine consumption by age and sex
Fig. 1 presents daily caffeine consumption (mg/day) from all
J.J. Knapik et al. / Food and Chemical Toxicology 105 (2017) 377e386380
sources by age and sex. Older men and women consumed more
caffeine from coffee compared to younger men and women,
respectively (p <0.01 men; p ¼0.02 women). Younger men
consumed more caffeine from energy drinks compared to older
men (p ¼0.04), but not between younger and older women
(p ¼0.10).
3.5. Representativeness of sample
Table 5 compares demographics in the volunteers in this study
with those of the entire population of Air Force personnel in the
same year the study was conducted. The comparison indicated that
in the present study, women, younger SMs, singles, and whites
were slightly overrepresented. Men, older SMs, married SMs, and
blacks/other SMs were slightly underrepresented. However, the
largest difference in any strata was no greater than 6.1%.
4. Discussion
The present study documented that 84% of Air Force SMs
consumed caffeinated products with an average consumption of
212 mg/day among consumers. Among SMs who regularly
consumed caffeinated products, men consumed 24% more caffeine
than women (224 vs. 180 mg/day), but when caffeine consumption
was calculated on a per weight basis, consumption was similar in
men and women. Sodas had the highest consumption prevalence
Table 3
Caffeine consumption (mean ±SE mg/day) of Air Force consumers (1 time/week) by demographic, lifestyle, and military characteristics (p-values from one-way analysis of
variance).
Variable Strata Any
caffeine
Coffee Hot tea Other
tea
Cola Other
soda
Energy
drink
Gum or
medication
Group All (n ¼1478) 212 ±978±37±122±125±19±163±88±1
Gender Men (n ¼1113)
Women (n ¼365)
224 ±11
180 ±12
75 ±4
89 ±8
6±1
11 ±2
24 ±2
16 ±2
27 ±2
20 ±2
10 ±1
4±1
74 ±10
32 ±8
7±1
9±2
p-value 0.03 0.08 <0.01 0.01 0.01 <0.01 0.02 0.59
Age 18-24 years (n ¼521)
25-29 years (n ¼424)
30-39 years (n ¼383)
40 years (n ¼170)
195 ±17
202 ±14
230 ±16
251 ±20
46 ±4
73 ±5
103 ±8
137 ±15
7±1
8±2
5±1
8±2
24 ±3
18 ±2
20 ±3
31 ±5
25 ±2
24 ±2
26 ±3
28 ±4
11 ±1
7±1
8±2
8±3
75 ±8
64 ±6
69 ±6
33 ±7
7±2
7±2
9±2
6±3
p-value 0.08 <0.01 0.54 0.13 0.81 0.22 0.03 0.83
Education Some high school/High school graduate (n ¼234)
Some college/Associate's degree (n ¼883)
Bachelor's/Graduate degree (n ¼364)
194 ±17
219 ±13
209 ±11
41 ±6
70 ±4
123 ±9
8±3
6±1
9±2
29 ±4
23 ±2
15 ±2
29 ±4
25 ±2
23 ±2
16 ±3
7±1
7±2
65 ±14
78 ±12
27 ±7
6±2
10 ±2
4±1
p-value 0.59 <0.01 0.15 <0.01 0.32 <0.01 0.02 0.07
Marital status Single (n ¼690)
Married (n ¼806)
200 ±14
223 ±11
67 ±1
88 ±4
7±1
7±1
22 ±2
22 ±2
26 ±2
25 ±2
9±1
8±1
63 ±13
64 ±9
7±1
8±2
p-value 0.18 <0.01 0.71 0.93 0.75 0.94 0.94 0.45
Race/ethnicity White (n ¼991)
Black (n ¼169)
Hispanic (n ¼189)
Other (n ¼138)
223 ±10
147 ±21
216 ±24
212 ±39
87 ±4
33 ±5
72 ±8
79 ±11
7±1
6±2
8±2
9±3
24 ±2
21 ±3
19 ±4
15 ±3
25 ±1
21 ±4
32 ±5
20 ±3
10 ±1
9±3
5±1
3±1
62 ±9
49 ±17
77 ±21
76 ±37
8±1
8±4
3±1
11 ±5
p-value 0.06 <0.01 0.81 0.24 0.09 0.04 0.78 0.34
Body mass index <25.0 kg/m
2
(n ¼596)
25.0e29.9 kg/m
2
(n ¼705)
30.0 kg/m
2
(n ¼182)
183 ±10
223 ±13
271 ±40
68 ±5
87 ±5
79 ±10
8±1
7±1
4±1
20 ±2
23 ±2
26 ±5
25 ±2
23 ±2
31 ±6
9±1
8±1
9±2
47 ±8
66 ±11
109 ±38
4±1
9±2
14 ±5
p-value <0.01 0.04 0.38 0.42 0.18 0.62 0.05 0.02
Tobacco use Current (n ¼360)
Former (n ¼358)
Never (n ¼778)
250 ±15
213 ±14
195 ±14
92 ±7
94 ±7
65 ±5
6±1
7±1
8±1
28 ±4
20 ±3
21 ±2
30 ±3
21 ±2
25 ±2
14 ±2
6±1
8±1
71 ±12
58 ±11
62 ±12
10 ±3
8±2
6±1
p-value 0.04 <0.01 0.43 0.08 0.04 <0.01 0.82 0.44
Aerobic exercise
duration
<136 min/wk (n ¼406)
136-225 min/wk (n ¼455)
226-345 min/wk (n ¼300)
346 min/wk (n333)
235 ±21
206 ±12
191 ±15
213 ±20
86 ±7
90 ±6
65 ±6
65 ±7
5±1
6±1
10 ±3
9±2
23 ±3
22 ±3
20 ±3
23 ±3
27 ±3
25 ±2
26 ±3
23 ±3
10 ±2
7±1
9±2
8±2
78 ±20
47 ±9
54 ±13
77 ±17
7±2
9±2
7±2
8±2
p-value 0.35 <0.01 0.07 0.89 0.78 0.51 0.31 0.90
Resistance
training
<2 sessions/wk (n ¼412)
2 sessions/wk (n ¼1082)
205 ±12
215 ±11
86 ±7
75 ±4
8±2
7±1
26 ±3
22 ±2
28 ±3
24 ±1
10 ±2
8±1
47 ±9
70 ±10
4±1
9±1
p-value 0.61 0.14 0.75 0.87 0.13 0.43 0.17 0.06
Sleep duration 4 h/night (n ¼66)
5-6 h/night (n ¼688)
7 h/night (n ¼700)
293 ±37
221 ±12
202 ±13
116 ±23
76 ±5
80 ±5
2±1
7±1
8±1
46 ±12
23 ±2
20 ±2
38 ±8
27 ±2
22 ±2
16 ±6
9±1
7±1
60 ±12
70 ±12
59 ±11
15 ±7
8±2
6±1
p-value 0.09 0.07 0.34 <0.01 0.01 0.03 0.78 0.30
Time In service 1 year (n ¼309)
2-5 years (n ¼432)
6-12 years (n ¼400)
13 years (n ¼357)
187 ±14
181 ±16
234 ±20
249 ±17
58 ±7
53 ±5
83 ±7
122 ±8
5±1
9±2
6±1
7±1
28 ±4
17 ±2
21 ±3
25 ±3
26 ±3
22 ±2
27 ±2
27 ±3
11 ±2
8±1
7±1
9±2
53 ±10
63 ±15
82 ±18
52 ±13
6±2
8±2
9±3
7±2
p-value <0.01 <0.01 0.26 0.05 0.47 0.19 0.48 0.68
Rank Junior enlisted, E1-E4 (n ¼594)
Senior enlisted, E5-E9 (n ¼689)
Junior officer, O1-O3 (n ¼141)
Senior officer, O4-O9 (n ¼65)
190 ±14
234 ±14
180 ±14
274 ±36
46 ±4
92 ±5
102 ±12
194 ±28
7±1
7±1
11 ±3
11 ±4
23 ±2
23 ±2
16 ±3
12 ±3
23 ±2
28 ±2
25 ±3
18 ±4
11 ±1
8±1
4±1
3±2
72 ±13
68 ±12
19 ±4
34 ±23
8±2
9±2
4±2
1±1
p-value 0.03 <0.01 0.22 0.23 0.18 0.02 0.20 0.30
Special Operations
Qualified
Yes (n ¼46)
No (n ¼1397)
200 ±62
214 ±9
44 ±13
80 ±4
2±1
7±1
20 ±7
22 ±2
16 ±5
26 ±1
8±3
8±1
109 ±57
63 ±8
2±2
8±1
p-value 0.77 0.07 0.21 0.80 0.17 0.86 0.31 0.40
J.J. Knapik et al. / Food and Chemical Toxicology 105 (2017) 377e386 381
(56%), followed by coffee (45%), teas (36%) and energy drinks (27%).
Consuming any caffeinated product was independently associated
with older age, race/ethnicity other than black, current or former
tobacco use, less aerobic training, and less sleep. Energy drink
consumption was independently associated with male gender,
younger age, current or former tobacco use, and less sleep.
4.1. Caffeine prevalence and daily consumption
Previous studies have been conducted on the caffeine con-
sumption of Army (Lieberman et al., 2012), Navy, and Marine Corps
(Knapik et al., 2016) personnel. All military studies (Knapik et al.,
2016; Lieberman et al., 2012) including the present one used a
similar questionnaire and the same definitions for caffeine sources.
The Army study (Lieberman et al., 2012) used a sampling technique
similar to the one used here where volunteers were contacted in
face-to-face encounters at installations across the US and overseas;
however, the Navy and Marine Corps study (Knapik et al., 2016)
identified a random sample and solicited volunteers by e-mail.
Table 6 compares caffeine consumption prevalence and daily con-
sumption of caffeine users among the military services. Prevalence
of consumption from any source (1/week) was greatest in Navy
personnel and lowest in the Army, although the difference in
prevalence was only 6%. Despite the lower consumption prevalence
in the Army, soldiers who did consume caffeine ingested the
greatest daily amount (mg/day), 1.5 to 1.6 times higher than the
other services. In the Army, Navy, and Marine Corps the highest
consumption prevalence was for coffee but Air Force personnel
were unique in that cola was the most ingested product with coffee
ranking second. Despite this, SMs in all services consumed the most
total caffeine (mg/day) from coffee with energy drinks ranking
second. In multivariate analysis, the Air Force and Navy/Marine
data were similar in that overall caffeine consumption prevalence
was independently associated with older age, ethnicity other than
black, and less sleep; energy drink intake was independently
associated with male gender, younger age, and less sleep. Thus,
Table 4
Characteristics associated with consumption (1 time/week) of specific caffeine products among Air Force personnel. Multivariable logistic regression, data presented as odds
ratios with 95% confidence intervals.
Variable Strata Caffeine beverages or gum/medication consumed 1 time/week
Any caffeine (model 1) Coffee
(model 2)
Tea
a
(model 3)
Soda
b
(model 4)
Energy drink
(model 5)
Gum or medication
(model 6)
Gender Men
Women
1.00
1.15 (0.82e1.61)
1.00
1.48 (1.15e1.90)
1.00
1.16 (0.91e1.48)
1.00
0.72 (0.57e0.92)
1.00
0.64 (0.48e0.86)
1.00
2.18 (1.34e3.56)
Age 18-24 years
25-29 years
30-39 years
40 years
1.00
1.32 (0.92e1.88)
1.71 (1.13e2.61)
3.52 (1.75e7.08)
1.00
1.25 (0.95e1.64)
1.80 (1.32e2.44)
2.14 (1.40e3.27)
1.00
1.01 (0.76e1.32)
1.04 (0.76e1.41)
1.81 (1.20e2.72)
1.00
1.14 (0.87e1.49)
1.59 (1.18e2.15)
1.83 (1.20e2.77)
1.00
1.19 (0.89e1.59)
0.77 (0.55e1.07)
0.39 (0.23e0.69)
1.00
1.02 (0.56e1.86)
1.11 (0.59e2.10)
1.20 (0.50e2.87)
Education Some HS/HS grad
Some college
College degree
1.00
0.92 (0.63e1.36)
1.07 (0.64e1.76)
1.00
1.52 (1.11e2.09)
3.01 (2.05e4.41)
1.00
0.80 (0.60e1.07)
0.74 (0.51e1.07)
1.00
0.81 (0.60e1.09)
0.73 (0.50e1.05)
1.00
1.10 (0.80e1.51)
0.69 (0.45e1.04)
1.00
1.07 (0.56e2.06)
0.61 (0.26e1.43)
Marital status Single
Married
1.00
1.07 (0.79e1.44)
1.00
1.24 (0.99e1.56)
1.00
1.09 (0.88e1.37)
1.00
0.98 (0.79e1.22)
1.00
1.09 (0.85e1.39)
1.00
1.41 (0.87e2.28)
Race/
Ethnicity
White
Black
Hispanic
Other
1.00
0.38 (0.26e0.55)
0.81 (0.54e1.21)
1.40 (0.79e2.50)
1.00
0.51 (0.36e0.72)
1.08 (0.79e1.47)
1.21 (0.84e1.75)
1.00
0.93 (0.67e1.28)
1.00 (0.73e1.36)
1.02 (0.71e1.46)
1.00
0.53 (0.39e0.73)
0.83 (0.61e1.12)
0.90 (0.63e1.29)
1.00
0.69 (0.47e1.02)
1.16 (0.83e1.61)
1.02 (0.68e1.53)
1.00
0.39 (0.17e0.90)
0.54 (0.25e1.15)
0.88 (0.41e1.91)
Body mass index <25.0 kg/m
2
25.0e29.9 kg/m
2
30.0 kg/m
2
1.00
0.96 (0.71e1.32)
0.70 (0.44e1.10)
1.00
1.05 (0.83e1.32)
0.65 (0.45e0.93)
1.00
0.78 (0.62e0.98)
0.71 (0.50e1.01)
1.00
0.94 (0.75e1.18)
0.85 (0.60e1.21)
1.00
1.08 (0.84e1.39)
1.00 (0.68e1.49)
1.00
1.36 (0.81e2.27)
1.92 (0.98e3.92)
Tobacco use Current
Former
Never
2.14 (1.43e3.18)
1.52 (1.07e2.15)
1.00
2.31 (1.76e3.02)
1.99 (1.53e2.57)
1.00
1.26 (0.97e1.64)
1.16 (0.90e1.49)
1.00
1.40 (1.08e1.83)
0.90 (0.70e1.15)
1.00
2.10 (1.60e2.77)
1.33 (1.00e1.76)
1.00
1.25 (0.73e2.17)
1.20 (0.70e2.06)
1.00
Aerobic exercise duration <136 min/wk
136-225 min/wk
226-345 min/wk
346 min/wk
1.97 (1.31e2.96)
1.56 (1.08e2.25)
1.28 (0.97e1.89)
1.00
1.23 (0.90e1.66)
1.44 (1.08e1.91)
1.14 (0.83e1.56)
1.00
0.95 (0.70e1.27)
0.93 (0.70e1.23)
0.94 (0.69e1.27)
1.00
1.70 (1.27e2.29)
1.43 (1.08e1.88)
1.20 (0.89e1.62)
1.00
1.14 (0.82e1.58)
0.87 (0.63e1.19)
1.01 (0.72e1.41)
1.00
1.33 (0.68e2.59)
1.24 (0.66e2.32)
1.28 (0.66e2.49)
1.00
Resistance training <2 sessions/wk
2 sessions//wk
1.13 (0.80e1.59)
1.00
1.06 (0.83e1.35)
1.00
0.93 (0.73e1.18)
1.00
1.27 (0.99e1.62)
1.00
0.78 (0.60e1.03)
1.00
0.65 (0.38e1.12)
1.00
Sleep duration 4 h/night
5-6 h/night
7 h/night
1.62 (1.21e2.17)
1.30 (0.66e2.55)
1.00
0.77 (0.46e1.29)
0.97 (0.78e1.20)
1.00
1.27 (0.78e2.10)
1.26 (1.02e1.55)
1.00
1.58 (0.94e2.64)
1.19 (0.97e1.47)
1.00
2.37 (1.42e3.97)
1.53 (1.21e1.93)
1.00
3.68 (1.68e8.08)
1.32 (0.83e2.11)
1.00
Abbreviation: HS ¼high School.
a
Includes hot and other teas.
b
Includes cola-type beverages and other sodas.
Fig. 1. Daily Average Consumption of Caffeinated Substances among Air Force
Personnel who Consume Caffeine.
J.J. Knapik et al. / Food and Chemical Toxicology 105 (2017) 377e386382
with the few exceptions noted above findings were similar across
the four services.
There are several previous population-based estimates of
caffeine consumption in Americans. Data from NHANES
(Drewnowski and Rehm, 2016; Fulgoni et al., 2015) was collected
from a 24-h dietary recalls. One study (Fulgoni et al., 2015) using
2001 to 2010 NHANES data found that 89% of men and 89% women
consumed caffeine on any given day with an average consumption
by consumers of 211 and 161 mg/day for 19 year old men and
women, respectively. The other NHANES study (Drewnowski and
Rehm, 2016) using 2011e2012 data found that male and female
caffeine consumers aged 20 years consumed an average of 196
and 151 mg/day, respectively. Caffeine data from the US Depart-
ment of Agriculture (USDA) Continuing Survey of Food Intakes by
Individuals (1994e1996 and 1998) (Frary et al., 2005) was based on
2 days of dietary intake data that included beverages and foods but
not energy drinks or medications. Their data (Frary et al., 2005)
indicated that 89% of adult men and 91% of adult women (18e34
years of age) consumed caffeinated substances with an average of
199 and 166 mg/day for male and female caffeine consumers,
respectively. Another study provided corrected values for the Frary
et al. (2005) survey based on an updated USDA nutrient database
and reported caffeine consumptions of 193 and 149 mg/day for
men and women 20 years of age, respectively (Ahuja et al., 2006).
Data from the Kantar Worldwide Beverage Consumption Panel
(involving US consumers only) was obtained from an on-line, 7-day
beverage consumption record and indicated that about 90% of
individuals 18 years of age consumed caffeinated beverages with
an average caffeine consumption equal to about 200 mg/day among
caffeine consumers (males and females were not separated)
(Mitchell et al., 2014). Prevalence values in these population-based
were slightly higher than the 84% observed in the present study
(1 week), but the average consumption of 224and 180 mg/day for
Air Force men and women, respectively, was higher. The differences
in the methods used in these studies noted above and those of the
present investigation must be considered when interpreting these
differences.
The estimated average daily caffeine consumption for Air Force
personnel was well below the levels of 400 mg/day for men and
200e300 mg/day for women of reproductive age that may be
associated with adverse effects (Nawrot et al., 2003; USDA, 2015).
Nonetheless, the present study found about 12% of men and 14%
(300 mg/day) to 31% (200 mg/day) of women exceeded these rec-
ommended amounts. Caffeinated products consumed by these
higher caffeine consumers differed from that of the entire group
with high male consumers ingesting a greater proportion of
caffeine from energy drinks and women from coffee and energy
drinks. These consumers may represent a unique subgroup that
consume energy drinks for reasons that are not clear from the data
collected here. Also, there are genetic differences that may lead to
higher caffeine consumption. A genetic polymorphism allows some
individuals to metabolize (N
3
-demethylation) caffeine in the liver
more rapidly than others, and another polymorphism may be
associated with higher caffeine tolerance and consumption
(Cornelis et al., 2007; Huang et al., 2005; Sachse et al., 1999).
Energy drink consumption prevalence (1 time/week) and daily
amounts of caffeine from energy drinks was 27% in the present
study and has varied in the military studies from 21% to 39%, as
shown in Table 6. Another Air Force investigation at a single loca-
tion (McCord Air Force Base, WA) reported 31% of active duty air-
men had consumed an energy drink (Schmidt et al., 2008), while a
sample of SMs consisting of many professional military medical
Table 5
Comparison of study sample to entire air force population.
Variable Strata DMED Present study Difference in proportions,
DMED-present study (%)
n Proportion (%) n Proportion (%)
Gender Men
Women
267,187
63,770
80.7
19.3
1113
365
75.3
24.7
5.4
5.4
Age 18-24 yr
25-29 yr
30-39 yr
40 yr
108,893
82,948
97,105
42,008
32.9
25.1
29.3
12.7
521
424
383
170
34.8
28.3
25.6
11.3
1.9
3.2
3.7
1.4
Marital Status Single
Married
135,572
195,382
41.0
59.0
690
806
46.1
53.9
5.1
5.1
Race White
a
Black
Other
242,425
47,356
41,173
73.3
14.3
12.4
1180
169
138
79.4
11.3
9.2
6.1
3.0
3.2
Abbreviation: DMED ¼Defense Medical Epidemiological Database.
a
The DMED did not have category for “Hispanics”, but conversations with a DMED administrator indicated that these were included in the “White”category.
Table 6
Comparison of prevalence and amount of caffeine use in the military services.
Prevalence (%) Daily consumption (mg/day)
a
Army
b
Navy
c
Marine corps
c
Air force Army
b
Navy
c
Marine corps
c
Air force
Any Caffeine Product 82 88 86 84 347 217 232 212
Coffee 56 67 64 45 155 136 125 78
Hot Tea 11 21 14 12 6 10 7 7
Other Tea 39 25 22 29 14 15 21 22
Cola 54 42 38 51 32 17 15 25
Other Soda 44 14 15 19 30 4 6 9
Energy Drink 39 21 32 27 97 29 43 63
Gums or Medications 8 9 8 5 10 7 14 8
a
Caffeine consumers only.
b
From Reference (Lieberman et al., 2012).
c
From Reference (Knapik et al., 2016).
J.J. Knapik et al. / Food and Chemical Toxicology 105 (2017) 377e386 383
personnel and military college students reported a 38% consump-
tion prevalence for energy drinks (Stephens et al., 2014). Studies of
energy drink consumption among US college students found that
39% consumed an energy drink in the past week (Skewes et al.,
2013) and 36% within the past two weeks (Marczinski, 2011).
4.2. Characteristics associated with caffeine consumption
In agreement with the present study, others (Drewnowski and
Rehm, 2016; Frary et al., 2005; Fulgoni et al., 2015; Knapik et al.,
2016; Lieberman et al., 2012) have reported that men consumed
larger amounts of caffeine than women. Nonetheless, this study
and others (Frary et al., 2005; Knapik et al., 2016) found that when
caffeine consumption was determined on a per kg body weight
basis, men and women consumed similar amounts. Acute caffeine
consumption modestly affects moods such as vigor and fatigue as
well as hemodynamic measures (e.g. blood pressure, cardiac
output) in both men and women (Amendola et al., 1998; Farag et al.,
2010; Hartley et al., 2004), although cardiovascular effects are more
likely to be observed at higher doses. It may be that both men and
women consume caffeinated products to provide similar behavioral
and/or cardiovascular effects.
Investigations involving representative civilian (Drewnowski
and Rehm, 2016; Frary et al., 2005; Fulgoni et al., 2015; Mitchell
et al., 2014) and military (Knapik et al., 2016) samples reported
overall caffeine consumption increased with age. In this study, coffee
consumption accounted for most of the caffeine ingestedin most age
groups (except for the 18e24 year olds), but younger (<40 years)
individuals consumed over twice as much caffeine from energy
drinks compared to older (40 years) individuals (67 vs. 33 mg/day)
and were over twice as likely to consume energy drinks (29 vs. 12%).
Energy drinks are a relatively new source of caffeine introduced into
the American market in 1997 (Heckman et al., 2010). Advertising of
these drinks is targeted to teenagers and individuals in the 18e34
year age group (Lal, 2007). This advertising may have influenced
energy drink consumption in the younger age groups in the present
study. It is possible that energy drink consumption may continue to
increase in the future since the US energy drink market's earning
growth is projected to be 52% from 2014 to 2019 (Bailey, 2015). It
should be noted that NHANES data examining trends in caffeine
consumption from 2001 to 2010 found little change in total con-
sumption among 19e22 year olds because increased caffeine con-
sumption from energy drinks was largely offset by decreases in
caffeine consumption from soda (Branum et al., 2014).
Several investigations have reported that, compared to those of
white race/ethnicity, those of black race/ethnicity had a lower
prevalence of caffeine consumption and a lower total amount
among consumers (Drewnowski and Rehm, 2016; Knapik et al.,
2016; Lieberman et al., 2012). Race/ethnic differences have been
reported in general dietary intake (Hiza et al., 2013; Wang and
Chen, 2011) that do not appear to be explained by nutritional
knowledge and briefs (Wang and Chen, 2011). Education level and
income differences may explain some of the variance (Bahr, 2007;
Raffensperger et al., 2010; Wang and Chen, 2011). In the present
study, the racial/ethnic differences in caffeine consumption
remained after controlling for education and other factors like to-
bacco use and exercise volume, similar to studies in Navy and
Marine Corps personnel (Knapik et al., 2016). The potential reasons
for the race/ethnic differences in caffeine prevalence and con-
sumption are likely complex and may differ in the military
compared to the general population.
Investigations using randomized, population-based samples
have shown that less sleep is associated with higher caffeine con-
sumption (Jacobson et al., 2012; Kant and Graubard, 2014; Knapik
et al., 2016; Toblin et al., 2012). The present study found in both
univariate and multivariate analyses that less sleep was associated
with caffeine consumption, especially for energy drinks and
caffeinated gums/medications. This is in general agreement with
studies of Navy and Marine personnel (Knapik et al., 2016) and the
relationship between less sleep and energy drink consumption has
also been found in other military investigations (Jacobson et al.,
2012; Toblin et al., 2012). Air Force and other military personnel
may be using caffeine-containing products to maintain alertness
and enhance cognitive performance degraded by lack of sleep.
Caffeine has been shown improve cognitive performance of Air
Force pilots flying simulated overnight missions and to enhance
performance in many other types of military operations (Doan
et al., 2006; Kaminori et al., 2015; Lieberman et al., 2002).
Caffeine, due to its ability to block central adenosine receptors,
increases alertness (Drake et al., 2013; Drapeau et al., 2006; Huang
et al., 2014; Ribiro and Sebastiao, 2010). When ingested shortly
before sleep periods in sufficient doses, it can reduce time spent
sleeping (Drake et al., 2013; Drapeau et al., 2006; Landolt et al.,
1995). As consequence, individuals often choose to ingest caffeine
to increase alertness and avoid sleep and caffeine is sold as an over-
the-counter drug for this purpose. Compared to civilian pop-
ulations, military personnel sleep less per night and frequently
have mandatory early morning activities such as group physical
training (Troxel et al., 2015). Furthermore, military operations and
training can occur at any time of the day and of ten entail substantial
loss of sleep. As a consequence the US Department of Defense has
developed, tested, and provided caffeine-containing products, such
as caffeine-containing gum, to sustain alertness and cognitive
performance when sleep is disrupted (Kaminori et al., 2015).
Tobacco use was associated with caffeine intake, especially for
coffee, colas, and energy drinks in both univariate and multivariate
analyses and tobacco users consumed more caffeine, especially
from coffee and sodas. Similar associations have repeatedly been
found in both military (Lieberman et al., 2012) and civilian pop-
ulations (Friis et al., 2014; Hewlett and Smith, 2006; Swanson et al.,
1994; Treur et al., 2016a). Smoking accelerates caffeine metabolism
(Brown et al., 1988; Parsons and Nelms, 1978) possibly requiring
smokers to consume more caffeine to achieve stimulatory effects.
Also, both caffeine and smoking increase dopaminergic activity in
different brain regions and the two substances may be used
concurrently to modulate stimulation (Tanda and Goldberg, 2000).
Studies of monozygotic and dizygotic twins suggest that both ge-
netic and environmental factors may be involved in the relation-
ship between tobacco and caffeine use (Hettema et al., 1999;
Kendler et al., 2008; Swan et al., 1997; Treur et al., 2016b) with
genetic influences increasing as individuals age (Kendler et al.,
2008). A recent study (Treur et al., 2016b) used a variety of tech-
niques to estimate the contribution from genetic and environ-
mental factors to caffeine consumption and found that genetic
correlations ranged from 0.44 to 0.47 while environmental corre-
lations ranged from 0.00 to 0.28 suggesting that the genetic
contribution was larger.
Air Force personnel showed an inverse relationship such that
the prevalence of overall caffeine consumption decreased as the
volume of aerobic exercise increased in both univariate and
multivariate analysis. This was primarily due to lower intake of
coffee and sodas in the more aerobically active Air Force personnel.
Previous studies of Army (Lieberman et al., 2012), Navy and Marine
Corps personnel (Knapik et al., 2016) have not found a relationship
between caffeine consumption and weekly aerobic exercise dura-
tion. Another difference across studies was the association between
resistance training and caffeine consumption. The Army investi-
gation (Lieberman et al., 2012) found little association between
resistance training and caffeine consumption in agreement in the
present study while the Navy and Marine Corps investigation
J.J. Knapik et al. / Food and Chemical Toxicology 105 (2017) 377e386384
(Knapik et al., 2016) found less caffeine consumption among those
who performed more resistance training. In this latter study, the
resistance training question was restructured (the only question
that differed in the military investigations) so that not only fre-
quency, but also duration of training could be assessed and the total
duration of weekly training could be calculated, similar to that of
the aerobic activity question. The greater ability to quantify the
volume of resistance exercise in the Navy/Marine Corps study may
more accurately describe the association; however, this will need to
be explored in future investigations.
4.3. Limitations
This study has several limitations, most of which relate to esti-
mates of daily caffeine consumption. First, all data were self-
reported and the usual shortcomings associated with this method
apply, including recall bias, social desirability, errors in self-
observation, and inadequate recall (Furnham, 1985; Podsakoff
et al., 2003). These biases could account for errors in reporting
serving sizes and how many times per week caffeinated products
were used, and thus errors in estimating daily caffeine consump-
tion. Second, caffeine data were obtained primarily from beverages
and gums/medications and did not include caffeine from food
sources. However, a previous study (Fulgoni et al., 2015) that did
include caffeine in both foods and beverages found that beverages
accounted for 98% of caffeine consumption, so it is likely that
omitting foods had a minimal effect on the caffeine intake esti-
mates. Third, the caffeine levels were obtained from company
websites, labels, and a database of caffeinated products. Although
company websites and food labels appear relatively accurate
(Attipoe et al., 2016), estimates of caffeine in coffee and tea may be
less so. The geographic location of coffee beans or tea leaves are
harvested and how they are processed and brewed can affect their
caffeine content (Barone and Roberts, 1996; Chin et al., 2008;
Hecimovic et al., 2011). Finally, there were a large number of sta-
tistical tests examining relationships between caffeine consump-
tion and amounts, and the demographic, lifestyle, and military
characteristics of Air Force personnel. The more effects investigated
the greater the chance of making a Type 1 error where the null
hypothesis will be incorrectly accepted. However, it is important to
show these relationships and the probability levels for adequate
comparisons with other investigations.
5. Conclusions
Among Air Force personnel, 84% reported consuming caffein-
ated products 1 time/week with male and female consumers
ingesting (mean ±standard error) 224 ±11 and 180 ±12 mg/day,
respectively. The most commonly consumed caffeinated products
(% users) were sodas (56%), coffee (45%), teas (36%), and energy
drinks (27%), although coffee provided the most caffeine. The
prevalence of energy drink consumption and caffeine ingested
from energy drinks was about twice as high among those <40 years
of age compared to those 40 years of age. Characteristics inde-
pendently associated with regular caffeine consumption included
older age, ethnicity other than Black, tobacco use, less aerobic
training and less sleep. The prevalence of caffeine consumption by
Air Force personnel (84%) was slightly lower but similar to that
reported for the civilian US population based on NHANES data
(89%) (Fulgoni et al., 2015); total consumption was higher in Air
Force personnel (212 mg/day) than that estimated from NHANES
data (173e186 mg/day) (Drewnowski and Rehm, 2016; Fulgoni
et al., 2015). Characteristics associated with caffeine use in Air
Force SMs were generally similar to those observed in civilian
investigations.
Disclaimers
The views, opinions, and findings in this report are those of the
authors and should not be construed as an official Department of
Defense policy, or decision, unless so designated by other official
documentation. Citations of commercial organizations and trade
names in this report do not constitute an official Department of the
Army or Air Force endorsement or approval of the products or
services of these organizations. The investigators have adhered to
the policies for protection of human subjects as prescribed in DOD
Instruction 3216.02 and the project was conducted in adherence
with the provisions of 32 CFR Part 219. Approved for public release;
distribution is unlimited. US Government Work (17 USC 105). Not
copyrighted in the US.
Acknowledgements
We would like to thank Dr Angie Cost at the Defense Health
Agency for providing us the Air Force population data. This research
was supported in part by an appointment to the Knowledge Pres-
ervation Program at the US Army Research Institute of Environ-
mental Medicine (USARIEM) administered by the Oak Ridge
Institute for Science and Education through an interagency agree-
ment between the US Department of Energy, and USARIEM. Other
support was provided by the DoD Center Alliance for Nutrition and
Dietary Supplement Research.
Transparency document
Transparency document related to this article can be found
online at http://dx.doi.org/10.1016/j.fct.2017.04.050.
References
Ahuja, J.K.C., Goldman, J.D., Perloff, B.P., 2006. The effect of improved composition
data on intake estimates in the United States of America. J. Food Consum. Anal.
19, S 7eS13.
Amendola, C.A., Gabrieli, J.D.E., Lieberman, H.R., 1998. Caffeine's effect on perfor-
mance and mood are independent of age and gender. Nutr. Neurosci. 1,
269e280.
Attipoe, S., Leggit, J., Deuster, P.A., 2016. Caffeine content in popular energy drinks
and energy shots. Milit. Med. 181, 1016e1020.
Bahr, P.R., 2007. Race and nutrition: an investigation of Black-White differences in
health-related nutrition behaviors. Sociol. Health Illn. 29 (6), 831e856.
Bailey, S., 2015. Energy drinks continue to thrive despite controversies. Mintel
Forecast. Strong Growth U. S. Energy Drink Mark.
Barone, J.J., Roberts, H.R., 1996. Caffeine consumption. Food Chem. Toxicol. 34 (1),
119 e129.
Branum, A.M., Rossen, L.M., Schoendorf, K.C., 2014. Trends in caffeine intake among
US children and adolescents. Pediatrics 133 (3), 386e393.
Brown, C.R., Jacob, P., Wilson, M., Benowitz, N.L., 1988. Changes in rate and pattern
of caffeine metabolism after cigarette abstinence. Clin. Phamacol. Ther. 43 (5),
488e491.
Caffeine Informer, 2015. Caffeine Content of Drinks.
Chin, J.M., Merves, M.L., Goldberg, B.A., Sampson-Cone, A., Cone, E.J., 2008. Caffeine
content of brewed teas. J. Anal. Toxicol. 32, 702e704.
Cornelis, M.C., Al-Sohemy, A., Campos, H., 2007. Genetic polymorphism of the
adenosine A2A receptor is associated with habitual caffeine consumption. Am. J.
Clin. Nutr. 86, 240e244.
Doan, B.K., Hickey, P.A., Lieberman, H.R., Fisher, J.R., 2006. Caffeinated tube food
effect on pilot performance during a 9-hour, simulated U-2 mission. Aviat.
Space Environ. Med. 77, 1034e1040.
Drake, C., Roehrs, T., Shambroom, J., Roth, T., 2013. Caffeine effects on sleep taken 0,
3, or 6 hours before going to bed. J. Clin. Sleep Med. 9 (11), 1195e1200.
Drapeau, C., Hamel-Herbrt, I., Robillard, R., Selmaoui, B., Filipini, D., Carrier, J., 2006.
Challenging sleep in aging: the effect of 200 mg of caffeine during the evening
in young and middle-aged moderate caffeine consumers. J. Sleep Res. 15,
133 e141.
Drewnowski, A., Rehm, C.D., 2016. Sources of caffeine in diets of US children and
adults: trends by beverage type and purchase location. Nutrients 8, 154.
Farag, N.H., Whitsett, T.L., McKey, B.S., Wilson, M.F., Vincent, A.S., Everson-
Rose, S.A.l., Lovallo, W.R., 2010. Caffeine and blood pressure response: sex, age
and hormonal status. J. Women's Health 19 (6), 1171e1176.
Frary, C.D., Johnson, R.K., Wang, M.Q., 2005. Food sources and intakes of caffeine in
J.J. Knapik et al. / Food and Chemical Toxicology 105 (2017) 377e386 385
the diets of persons in the United States. J. Am. Diet. Assoc. 105, 110e113.
Friis, K., Lyng, J.I., Lasgaard, M., Larsen, F.B., 2014. Energy drink consumption and the
relation to socio-demographic factors and health behaviors among young
adults in Denmark. A population-based study. Eur. J. Public Health 24 (5),
840e844.
Fulgoni, V.L., Keast, D.R., Lieberman, H.R., 2015. Trends in intake and sources of
caffeine in the diet of US adults: 2001-2010. Am. J. Clin. Nutr. 101, 1081e1087.
Furnham, A., 1985. Response bias, social desirability and dissimulation. Personal.
Individ. Differ. 7 (3), 385e400.
Hartley, T.R., Lovallo, W.R., Whitsett, T.L., 2004. Cardiovascular effects of caffeine in
men and women. Am. J. Cardiol. 93, 1022e1026.
Hecimovic, I., Belscak-Cvitanovic, A., Horzic, D., Komes, D., 2011. Comparative study
of polyphenols and caffeine in different coffee varieties affected by the degree
of roasting. Food Chem. 129, 991e100 0.
Heckman, M.A., Sherry, K., GonzalezDeMejia, E., 2010. Energy drinks: assessment of
their market size, consumer demographics, ingredient profile, functionality,
and regulations in the United States. Compr. Rev. Food Sci. Food Saf. 9, 303e317.
Hettema, J.M., Corey, L.A., Kendler, K.S., 1999. A multivariate genetic analysis of the
use of tobacco, alcohol and caffeine in a population based sample of male and
female twins. Drug Alcohol Depend. 57, 69e78.
Hewlett, P., Smith, A., 2006. Correlates of daily caffeine consumption. Appetite 46,
97e99.
Hiza, H.A.B., Casavale, K.O., Guenther, P.M., Davis, C.A., 2013. Diet quality of Amer-
icans differ by age, sex, race/ethnicity, income and educational level. J. Acad.
Nutr. Diet. 113 (2), 297e306.
Huang, Z.L., Qu, W.M., Eguchi, N., Chen, J.F., Schwarzschild, M.A., Fredholm, B.B.,
Urade, Y., Hayaishi, O., 2005. Adenosine A
2A
but not A
1
receptors mediate the
arousal effect of caffeine. Nat. Neurosci. 8 (7), 858e859.
Huang, Z.L., Zhang, Z., Qu, W.M., 2014. Roles of adenosine and its receptors in sleep-
wake regulation. Int. Rev. Neurobiol. 119, 349e371.
Jacobson, I.G., Horton, J.L., Smith, B., Wells, T.S., Boyko, E.J., H.R, L., Ryan, M.A.K.,
Smith, T.C., 2012. Bodybuilding, energy, and weight loss supplements are
associated with deployment and physical activity in U.S. military personnel.
Ann. Epidemiol. 22 (5), 318e330.
Kaminori, G.H., McLellan, T.M., Tate, C.M., Voss, D.M., Niro, P., Liberman, H.R., 2015.
Caffeine improves reaction time, vigilance and logical reasoning during
extended periods with restricted opportunities for sleep. Psychopharmacology
232, 2013e2042.
Kant, A.K., Graubard, B.I., 2014. Association between self-reported sleep duration
with eating behaviors of American adults: NHANES 2005-2010. Am. J. Clin. Nutr.
100, 938e947.
Kendler, K.S., Schmitt, E., Aggen, S.H., Prescott, C.A., 2008. Genetic and environ-
mental influences on alcohol, caffeine, cannabis and nicotine use from early
adolescence to middle adulthood. Arch. General Psychiatry 65 (6), 674e682.
Knapik, J.J., Trone, D.W., McGraw, S., Steelman, R.A., Austin, K.G., Lieberman, H.R.,
2016. Caffeine Use Among Act. Duty Navy Mar. Corps Pers. Nutr. 6, 620.
Knight, C.A., Knight, I., Mitchell, D.C., Zepp, J.E., 2004. Beverage caffeine intake in the
US consumers and subpopulations of interest: estimates from the Share of
Intake Panel survey. Food Chem. Toxicol. 42 (12), 1923e1930.
Lal, G.G., 2007. Getting specific with functional beverages. Food Technol. 61 (12),
25e31.
Landolt, H.P., Werth, E., Borbeely, A.A., Dijk, D.J., 1995. Caffeine intake (200 mg) in
the morning affects human sleep and EEG power spectra at night. Brain Res.
675, 67e74.
Lieberman, H.R., Stavinoha, T., McGraw, S., White, A., Hadden, L., Marriott, B.P., 2012.
Caffeine use among active duty US Army soldiers. J. Acad. Nutr. Diet. 112 (6),
902e912.
Lieberman, H.R., Tharion, W.J., Shukitt-Hale, B., Speckman, K.L., Tulley, R., 2002.
Effects of caffeine, sleep loss, and stress on cognitive performance and mood
during US Navy SEAL training. Psychopharmacology 164, 250e261.
Marczinski, C.A., 2011. Alcohol mixed with energy drinks: consumption patterns
and motivations for use in U.S. college students. Int. J. Environ. Res. Public
Health 8, 3232e3245.
Mitchell, D.C., Knight, C.A., Hokenberry, J., Teplansky, R., Hartman, T.J., 2014.
Beverage caffeine intakes in the U.S. Food Chem. Toxicol. 63, 136e142 .
Nawrot, P., Jordan, S., Eastwood, J., Rotstein, J., Hugenholtz, A., Feeley, M., 2003.
Effects of caffeine on human health. Food Addit. Contam. 20 (1), 1e30.
Parsons, W.D., Nelms, A.H., 1978. Effects of smoking on caffeine clearance. Clin.
Pharmacol. Ther. 24 (1), 40e45.
Podsakoff, P.M., MacKenzie, S.B., Lee, J.Y., Podsakoff, N.P., 2003. Common method
biases in behavioral research: a critical review of the literature and recom-
mended remedies. J. Appl. Psychol. 88 (5), 879e903.
Raffensperger, S., Kuczmarski, M.F., Hotchkiss, L., Cotugna, N., Evans, M.K.,
Zonderman, A.B., 2010. The effect of race and predictors of socioeconomic status
on diet quality in the in the Healthy Aging in Neighborhoods of Diversity across
the Life Span (HANDLS) study sample. J. Natl. Med. Assoc. 102 (10), 923e930.
Ribiro, J.A., Sebastiao, A.M., 2010. Caffeine and adenosine. J. Alzheimer's Dis. 20,
S3eS15.
Sachse, C., Brockmoller, J., Bauer, S., Roots, I., 1999. Functional significance of a C to A
polymorphism in intron 1 of the cytochrome P450 CYP1A2gene tested with
caffeine. Br. J. Clin. Pharmacol. 47, 445e449.
Schmidt, R.M., McIntire, L.K., Caldwell, J.A., Hallman, C., 2008. Prevalence of Energy-
drink and Supplement Usage in a Sample of Air Force Personnel. Air Force
Research Laboratory, Wright-Patterson Air Force Base.
Skewes, M.C., Decou, C.R., Gonzalez, V.M., 2013. Energy drink use, problem drinking
and drinking motives in a diverse sample of Alaska college students. Int. J.
Circumpolar Health 72, 21204.
Stephens, M.B., Attipoe, A.S., Jones, D., Ledford, C.J.W., Deuster, P.A., 2014. Energy
drink and energy shot use in the military. Nutr. Rev. 72 (Suppl. 1), 72e77.
Swan, G.E., Carmelli, D., Cardon, L.R., 1997. Heavy consumption of cigarettes, alcohol
and coffee in male twins. J. Stud. Alcohol 58 (2), 182e190 .
Swanson, J.A., Lee, J.W., Hopp, J.W., 1994. Caffeine and nicotine: a review of their
joint use and possible interactive effects in tobacco withdrawal. Addict. Behav.
19 (3), 229e256.
Tanda, G., Goldberg, S.R., 2000. Alternations of the behavioral effects of nicotine by
chronic caffeine exposure. Pharmacol. Biochem. Behav. 66 (1), 47e64.
Toblin, R.L., Clarke-Walper, K., Kok, B.C., Sipos, M.L., Thomas, J.L., 2012. Energy drink
consumption and its association with sleep problems among U.S. service
members on a combat deploymenteAfghanistan, 2010. MMWR 61 (44),
895e898.
Treur, J.L., Taylor, A.E., Ware, J.J., McMahon, G., Hottenga, J.J., Baselmans, B.M.L.,
Willemsen, G., Boomsma, D.I., Munafo, M.R., Vink, J.M., 2016a. Association be-
tween smoking and caffeine consumption in two European cohorts. Addiction
111, 10 5 9e10 68 .
Treur, J.L., Taylor, A.E., Ware, J.J., Nivard, M.G., Neale, M.C., McMahon, G.,
Hottenga, J.J., Baselmans, B.M.L., Boomsma, D.I., Munafo, M.R., Vink, J.M., 2016b.
Smoking and caffeine consumption: a genetic analysis of their association.
Addict. Biol. (in press).
Troxel, W.M., Shih, R.A., Pedersen, E., Geyer, L., Fisher, M.P., Griffin, B.A., Haas, A.C.,
Kurz, J.R., Steinberg, P.S., 2015. Sleep in the Military: Promoting Healthy Sleep
Among US Service Members. Rand Corporation, Santa Monica CA.
USDA, 2015. Scientific Report of the 2015 Dietary Guidelines. Advisory Committee.
Wang, Y., Chen, X., 2011. How much of racial/ethnic disparities in dietary intake,
exercise, and weight status can be explained by nutrition- and health-related
psychosocial factors and socioeconomic status among US adults. J. Am. Diet.
Assoc. 111, 1904e1991.
J.J. Knapik et al. / Food and Chemical Toxicology 105 (2017) 377e386386
