MILITARY MEDICINE, 172, 6:592, 2007
Effects of Heavy Load Carriage during Constant-Speed, Simulated,
Road Marching
Guarantor: Matthew D. Beekley, PhD
Contributors: Matthew D. Beekley, PhD; MAJ Jonathan Alt, ORSA USA; MAJ Clyde M. Buckley, EOD USA;
Michael Duffey, MS; Todd A. Crowder, PhD
Load carriage is a key element in dismounted military opera-
tions. Load carriage requirements in the field regularly exceed
50% of lean body mass (LBM) and have only rarely been stud-
ied. Therefore, our purpose was to determine the metabolic
and motivational effects of heavy loads (30 –70% LBM) during
constant-rate “road” marching on a treadmill. Ten healthy
male Army officers carried loads of 30%, 50%, and 70% LBM in
an all-purpose, lightweight, individual, carrying equipment
pack for 30 minutes, at a speed of 6 km/h. Oxygen consump-
tion (V̇O2), ventilation, heart rate (HR), respiratory exchange
ratio, rating of perceived exertion (RPE), and Self-Motivation
Inventory scores were recorded at each trial. Significant in-
creases were observed for V̇O2, ventilation, and HR between the
trials. RPE significantly increased for the 70% LBM trial, com-
pared with the 30% and 50% trials. No significant differences
were seen in respiratory exchange ratio or Self-Motivation In-
ventory scores. Increasingly heavy loads carried in a rucksack
resulted in increased V̇O2, RPE, and HR; therefore, increasing
the load that a soldier is required to carry may negatively affect
road march performance.
Introduction
oldiers are sometimes required to carry equipment and sup-
S plies in load carriage systems. In recent years, the load
carried has increased to a considerable percentage of an indi-
vidual’s body mass (BM), presumably because of the need to
carry more equipment for increased protection or firepower.1,2 It
is well known that this increase in the load carried may increase
the risk of injury or reduce the military or physical performance
of the individual carrying the load.1,2
In general, there are only a few published studies examining
heavier carried loads, and many of those were conducted by
using military personnel as subjects (cited in Ref. 3). This is
likely attributable to lower fitness or motivation levels of the
general subject population. Studying carriage of heavier loads is
important, because it is likely that soldiers will actually be
carrying heavier loads in the field.
The metabolic response to carried lighter carried loads has
been examined. For example, Quesada et al.4 examined the
metabolic responses of 12 male Army recruits during a simu-
lated road march on a treadmill by using three load carriage
conditions: 0% body weight (BW) backpack load, 15% BW load,
and 30% BW load. Significant differences were observed be-
tween loads in oxygen consumption (V̇O2), ventilation (VE), and
heart rate (HR) throughout the marches, indicating a linear
Department of Physical Education, U.S. Military Academy, West Point, NY 10996.
Presented at the American College of Sports Medicine Annual Meeting, June 2–5,
2004, Indianapolis, IN.
This manuscript was received for review in August 2005. The revised manuscript
was accepted for publication in May 2006.
Reprint & Copyright © by Association of Military Surgeons of U.S., 2007.
Military Medicine, Vol. 172, June 2007 592
response in metabolism to the differing loads. However, meta-
bolic responses to heavier loads (⬎50% BW) were not examined.
In the study by Quesada et al.,4 the load was standardized
Downloaded from https://academic.oup.com/milmed/article/172/6/592/4578047 by guest on 23 February 2024
according to BM. If the load is not standardized according to BM,
then it becomes hard to compare data for subjects of varying
sizes. However, it should be pointed out that lean BM (LBM) (i.e.,
muscle and bone mass) represents the metabolically active tis-
sues that “carry” the load. Therefore, we chose to examine met-
abolic responses to heavier loads of 30%, 50%, and 70% of LBM
in a fit military population during simulated road marching.
Methods
Army officers were solicited from the U.S. Military Academy
faculty and 10 male officers agreed to participate in the study.
Subjects were considered fit for full duty and deployment, were
not being monitored by their physicians for any disease, and
had no history of musculoskeletal injury or recent pain com-
plaints. Subjects self-reported their current 2-mile run times
and ages. Subjects with 2-mile run times of ⬎14 minutes were
screened out of the study. All procedures were approved by the
institutional review board at the U.S. Military Academy, Keller
Army Community Hospital.
Subjects reported to the laboratory for a screening trial.
Height, weight, body composition (determined via Bod Pod), and
treadmill V̇O2max were ascertained at this trial. Body composition
was always determined before treadmill V̇O2max. The Bod Pod
was calibrated and used according to the manufacturer’s in-
structions (Life Measurement, Concord, California). The Bod
Pod determines whole-body density from BM and body volume
(determined by air displacement), which is then converted to
body fat percentage via the equation described by Siri.5 After
determination of body fat percentage, LBM was calculated [total
BM ⫺ (percentage of body fat ⫻ total BM) ⫽ LBM]. Treadmill
V̇O2max was determined by using AEI model S-3A oxygen and
model CD-3A carbon dioxide analyzers (Applied Electrochemis-
try, Inc., Naperville, Illinois), with a UVM turbine transducer
(Vacumed, Ventura, California) to measure VE. Turbofit soft-
ware (Vacumed) was used for V̇O2 calculations. Subjects ran for
2 minutes at 10 km/h, 0% grade, for 2 minutes at 14 km/h, 0%
grade, and for 2 minutes at 18 km/h, 0% grade, with the grade
increasing 2% every 2 minutes thereafter until fatigue. Criteria
for V̇O2max were three of four of the following: a plateau in V̇O2
with increasing work, a respiratory exchange ratio (RER) greater
than 1.1, achievement of an age-predicted maximal HR (220 ⫺
age, in years), or a rating of perceived exertion (RPE) of 19 or 20.6
Subjects were required to rest for at least 2 days after the
screening trial before completing three separate simulated road
march trials. “Rest” meant subjects were asked to refrain from
all personal training activities, as well as group physical train-
Heavy Load Carriage 593

ing. Road march trials were separated by at least 48 hours (with

rest between trials) and were completed in random order.
For a road march trial, subjects reported to the laboratory
with no caffeine or food in the previous 4 hours. The same time
of day was used for all three trials for each individual subject.
Standard-issue physical training shorts, T-shirt, black wool
socks, and combat boots were used as the dress for all simu-
lated road march trials. Subjects marched on a Kistler Gaitway
treadmill (Kistler Instruments, Amherst, New York), whose
speed was calibrated according to the manufacturer’s instruc- Fig. 1. Mean ⫾ SD V̇O2 (squares) and V̇E (diamonds) after 15 minutes of
tions. Subjects were asked to march for 30 minutes at 6 km/h, treadmill foot marching (for each workload) at 6 km/h (n ⫽ 10). ⴱ, Significantly
carrying 30%, 50%, or 70% of LBM. Each subject’s personal different from 30% LBM load for both V̇O2 and V̇E (p ⬍ 0.05). †Significantly different
from 50% LBM load for both V̇O2 and V̇E (p ⬍ 0.05).
all-purpose, lightweight, individual, carrying equipment (ALICE)

Downloaded from https://academic.oup.com/milmed/article/172/6/592/4578047 by guest on 23 February 2024

backpack (without modifications) was used, filled to the appro-
priate weight for each subject and trial with sandbags. Subjects
breathed through a Hans-Rudolph two-way rebreathing valve
connected to the turbine and gas analyzers. V̇O2, V̇E, RER, HR
(measured with a Polar HR monitor), and RPE (scale of 6–20)
were determined every 5 minutes of the trial.
The Self-Motivation Inventory (SMO) described by Dishman et
al.,7 which consists of 40 items with 5-point, Likert-format re-
sponses, was administered immediately after completion of the
trial. The SMO is suggested to measure the tendency to persist
Fig. 2. Mean ⫾ SD HR after 15 minutes of treadmill foot marching (for each
in vigorous physical activity regardless of extrinsic reinforce- workload) at 6 km/h (n ⫽ 10). ⴱ, Significantly different from 30% LBM load (p ⬍ 0.05).
ment.7 It is usually used to measure exercise adherence without †Significantly different from 50% LBM load (p ⬍ 0.05).
regard to external factors. We wanted to determine whether the
subjects’ motivation, as measured with the SMO, changed on
the basis of the load carried for each trial.
The intraclass correlations (which indicate the reliability of
the measures8) from our laboratory, based on pilot data for five
subjects (data not shown), were as follows: V̇O2, 0.93; V̇E, 0.94.
The same testers were present to assess subjects for all trials.
Statistical analyses were conducted by using a repeated-mea-
sures analysis of variance. Statistical significance was set at p ⱕ
0.05. Four subjects were unable to complete the full 30-minute
trial at 70% LBM. For this reason, data taken at the midpoint of Fig. 3. Mean ⫾ SD RER after 15 minutes of treadmill foot marching (for each
each trial (15 minutes), which all subjects were able to complete, workload) at 6 km/h (n ⫽ 10). There were no significant differences between loads
were used for analysis. (p ⬎ 0.05).

Results
Subject characteristics and mean loads are shown in Table I.
V̇O2, V̇E, and HR were significantly different between the three
load trials (Figs. 1 and 2). RER did not significantly change
during the trials, although it trended upward (Fig. 3). RPE was
significantly higher in the 70% LBM trial, compared with the
other two trials (Fig. 4). SMO scores did not significantly change
in the trials (Fig. 4).
To create Figures 5 and 6, we used data from the study by
Quesada et al.4 and combined them with our results. Their
mean subject BW was very similar (78.6 kg) to ours (79.6 kg).
Their subjects marched on a treadmill at the same speed as in
our study but used 0%, 15%, or 30% BM for the load in an
ALICE backpack. Technically, we used LBM instead of BM to
calculate pack weight; however, we wished only to indicate the
trend in the data, and the difference in pack weights using BM,
compared with LBM, should be only ⬃10% for this group. How-
ever, responses for a greater range of carried loads could be
examined by combining these two data sets. Data points used
were taken from the subject responses after 20 minutes of sim-
ulated road marching, as reported by Quesada et al.4
Four subjects were unable to complete the 70% LBM trial and
quit of their own volition. We examined subject characteristics
for clues regarding why they failed. We analyzed these charac-
teristics with a t test for unequal sample sizes. Two of the

TABLE I
SUBJECT CHARACTERISTICS AND LOADS CARRIED (n ⫽ 10)

V̇O2max Load (kg)

Height (m) Weight (kg) Body Fat (%) Age (years) (mL/kg per Minute) 30% LBM 50% LBM 70% LBM
1.79 ⫾ 0.07 79.6 ⫾ 7.7 11.6 ⫾ 5.5 32.4 ⫾ 1.3 52.6 ⫾ 6.2 20.6 ⫾ 1.9 34.3 ⫾ 3.2 48.0 ⫾ 1.9

Military Medicine, Vol. 172, June 2007

594
Fig. 4. Mean ⫾ SD RPE (squares) and SMO score (diamonds) after 15 minutes of
treadmill foot marching (for each workload) at 6 km/h (n ⫽ 10). ⴱ, Significantly
different from 30% LBM load for RPE (p ⬍ 0.05). There were no significant differ-
ences for SMO scores (p ⬎ 0.05).
Fig. 5. Mean percentage of V̇O2max after ⬃15 to 20 minutes of treadmill foot
marching (for each workload) at 6 km/h. Data are from the study by Quesada et al.4
(diamonds) and the current study (squares). SDs were omitted for clarity.
Fig. 6. Mean HR after ⬃15 to 20 minutes of treadmill foot marching (for each
workload) at 6 km/h. Data are from the study by Quesada et al.4 (diamonds) and the
current study (squares). SDs were omitted for clarity.
subjects who failed were significantly shorter than the others
(1.67 ⫾ 0.03 m vs. 1.82 ⫾ 0.06 m; p ⬍ 0.05), and the other two
subjects who failed had significantly higher percentages of body
fat than the others (20 ⫾ 4.4% vs. 12 ⫾ 5.1%; p ⬍ 0.05). During
the 70% LBM trial, V̇O2, V̇E, and HR at 15 minutes were signif-
icantly greater for the two shorter subjects than for the rest of
the subjects (mean V̇O2, 39.7 ⫾ 5.9 mL/kg per minute vs.
28.4 ⫾ 5.1 mL/kg per minute; mean V̇E, 86.4 ⫾ 12.3 L/min vs.
62.3 ⫾ 11.7 L/min; HR, 195 ⫾ 18 beats per minute vs. 153 ⫾ 17
beats per minute; p ⬍ 0.05 for all). There were no differences in
V̇O2, V̇E, and HR for the two “fatter” subjects, compared with the
rest of the subjects, in any trial.
Discussion
It is well known that there is a systematic increase in the
energy cost of load carriage with increases in load mass, BM,
Military Medicine, Vol. 172, June 2007
Heavy Load Carriage
and velocity (reviewed in Ref. 1). Therefore, it is not surprising
that we saw systematic increases in V̇O2, HR, and V̇E with
increased loads at the same velocity of road marching. We ex-
tended previous research by using loads that were heavier than
previously used.4 Using combined data, Figures 5 and 6 suggest
that HR and relative energy costs (percentage of V̇O2max) are
fairly linear from no load up to a load of 70% of LBM for fit male
subjects during road marches of ⬃15- to 20-minute duration at
6 km/h with 0% grade.
These increases in metabolic cost with heavier carried loads
do not come without penalty. For instance, increased metabolic
load affects how fast soldiers can move, inhibits movement over
obstacles, affects how fatigued soldiers are upon arrival, in-
Downloaded from https://academic.oup.com/milmed/article/172/6/592/4578047 by guest on 23 February 2024
creases food (calorie) needs, and increases the risk of injuries
such as blisters, knee pain, low back injuries, and stress frac-
tures (reviewed in Ref. 3). Two extensive reviews have made
suggestions on how to improve soldier mobility with load car-
riage.1,3
Current reports from Afghanistan suggested that the average
rifleman’s fighting load was 29 kg (63 lb), the average approach
march load was 44 kg (96 lb), and the average emergency ap-
proach march load was 58 kg (127 lb) (C. Dean, personal com-
munication). These loads were carried for most of 1 day. Obvi-
ously, these loads are not standardized for BM, which could
further increase metabolic cost for smaller individuals. Clearly,
the load carried by today’s soldier continues to increase; there-
fore, problems encountered with load carriage will continue to
plague the military for the foreseeable future.
Two of the subjects who could not complete our study (at 70%
LBM load) were significantly shorter in stature than the other
subjects, although body fat and fitness level (measured as
V̇O2max) were not significantly different from those for the rest of
the subjects (data not shown). Increased V̇O2, V̇E, and HR at 15
minutes were also seen for these subjects, indicating the shorter
subjects might have adopted a different strategy for transport-
ing the load. Further study, including kinematic analysis of the
subjects, may reveal the difference that subject height has in
carriage of heavy loads.
The other two subjects who could not complete our study (at
70% LBM load) had significantly higher percentages of body fat,
compared with other subjects, although BM and fitness level
(measured as V̇O2max) were not significantly different from those
for the rest of the subjects (data not shown). It has been sug-
gested that individuals with more fat-free mass (e.g., more mus-
cle mass) can perform load carriage more rapidly.1 Along these
lines, our data suggest that individuals with higher percentages
of body fat (and presumably less muscle mass) may reach fa-
tigue faster during simulated road marching with heavy loads.
The effect of body fatness on load carriage warrants further
investigation.
The 50% LBM load was not perceived (based on mean RPE
response) as being harder work than the 30% LBM load (Fig. 4),
although clearly the metabolic cost was greater (Figs. 1 and 2).
Interestingly, with simulated road marching at the same speed
and grade, with the same backpack (ALICE), Quesada et al.4
found no difference in RPE responses between no load (0% BM)
and 30% BM load, although the metabolic cost was greater.
Quesada et al.4 suggested two possibilities, that is, (1) there may
be a load threshold below which subjects do not notice differ-
Heavy Load Carriage
ences in sensation of effort or (2) military subjects underesti-
mated their RPEs if they thought these responses reflected their
physical or mental prowess. Our results suggest these same
hypotheses, although we tend to favor the latter. In that respect,
these similar results suggest that researchers should be careful
using RPE, or overinterpreting RPE results, for military sub-
jects.
The SMO described by Dishman et al.7 is suggested to mea-
sure the tendency to persist in vigorous physical activity regard-
less of extrinsic reinforcement. It is usually used to measure
exercise adherence without regard to external factors. We
wanted to determine whether the subjects’ motivation, as mea-
sured with the SMO score, changed on the basis of the load
carried for each trial. The results of the SMO score suggest that
the subjects’ self-motivation did not change as a result of the
recent completion of the trial, regardless of load (Fig. 4).
The fact that we used sandbags in an ALICE backpack might
have played a role in our results. Attention was paid to distrib-
uting the load equally along the height of the pack, to prevent
migration of the load to the bottom of the pack, because higher
energy costs have been found with a load placed lower on the
back, compared with one placed higher on the back.9 Despite
our efforts, sandbags placed in an empty ALICE backpack
tended to migrate to the lowest central area of the pack, which
might have driven up energy costs and possibly induced fatigue
earlier in our subjects. This may be attributable to subjects
having to bend forward more to maintain their center of mass
over their feet. Researchers should also consider that much of a
soldier’s load in the field is carried on or about the body and not
in a backpack (C. Dean, personal communication). This allows
materials to be readily available and “at hand” during combat
situations. Researchers should carefully consider placement
and type of load in further research using backpack load car-
riage, including trying to simulate in a laboratory environment
how and what soldiers actually carry in the field.
Our subjects were generally older (mean age, 32.4 years) than
the subjects in the study by Quesada et al.4 (mean age, 22.4
years) and were also generally older than basic training recruits.
Age tends to cause a decrease in LBM and an increase in body
fat, which would likely negatively affect load carriage, although
the V̇O2max of our subjects was approximately the same as that
reported by Quesada et al.,4 as well as that of basic Army train-
ing recruits (52.6 mL/kg per minute [current study], 58.3
mL/kg per minute,4 and 50.6 mL/kg per minute,10 respectively),
indicating similar aerobic fitness levels. Researchers should
consider using younger subjects to closely mimic the average
age of soldiers in the field.
595
Finally, researchers should consider testing for a longer du-
ration. It has been shown that, as subjects experience muscular
fatigue, the biomechanics of load carriage change, which could
increase the risk of injury.4 It has also been suggested that
oxygen uptake can increase over time, indicating increased me-
tabolism.3 Most importantly, soldiers in the field cannot quit at
predetermined times and may be carrying loads for extended
periods of time.
In conclusion, our results, coupled with previous research,
suggest that HR and relative energy cost are fairly linear be-
tween 0% LBM load and 70% LBM load in ALICE backpacks
during marching with no grade at 6 km/h for 20 minutes. The
relationships between height and body fatness and load carriage
Downloaded from https://academic.oup.com/milmed/article/172/6/592/4578047 by guest on 23 February 2024
need to be investigated further. Increasing loads that soldiers
are required to carry may negatively affect road march perfor-
mance. Increasing loads, with possible attendant decrements in
performance, will continue to be carried by our military.
Acknowledgment
This research was supported in part by Program Executive Officer
Soldier.
References
1. Knapik JJ, Harman EA, Reynolds KL: Load carriage using packs: a review of
physiological, biomechanical and medical aspects. Appl Ergon 1996; 27: 207–16.
2. Dubik JM, Fullerton TD: Soldier overloading in Grenada. Milit Rev 1987; 67:
38 – 47.
3. Knapik JJ, Reynolds KL, Harman EA: Soldier load carriage: historical, physiolog-
ical, biomechanical, and medical aspects. Milit Med 2004; 169: 45–56.
4. Quesada PM, Mengelkoch LJ, Hale RC, Simon SR: Biomechanical and metabolic
effects of varying backpack loading on simulated marching. Ergonomics 2000;
43: 293–309.
5. Siri WE: Body composition from fluid spaces and density: analysis of methods. In:
Techniques for Measuring Body Composition. Edited by Brozek J, Henschel A.
Washington, DC, National Academy of Sciences National Research Council, 1961.
6. McCardle WD, Katch FI, Katch VL: Exercise Physiology, Ed 5. Philadelphia, PA,
Lippincott Williams & Wilkins, 2001.
7. Dishman RK, Ickes W, Morgan WP: Self-motivation and adherence to habitual
physical activity. J Appl Soc Psychol 1980; 10: 115–32.
8. Vincent WJ: Statistics in Kinesiology, Ed 3. Champaign, IL, Human Kinetics,
2005.
9. Obusek JP, Harman EA, Frykman PN, Palmer CJ, Bills RK: The relationship of
backpack center of mass to the metabolic cost of load carriage. Med Sci Sports
Exerc 1997; 29: S205.
10. Sharp MA, Patton JF, Knapik JJ, et al: Comparison of the physical fitness of men
and women entering the U.S. Army: 1978 –1998. Med Sci Sports Exerc 2002; 34:
356 – 63.
Military Medicine, Vol. 172, June 2007