PROCEEDINGS of the HUMAN FACTORS and ERGONOMICS SOCIETY 57th ANNUAL MEETING - 2013
981
Quantification of the Physical Demands for Servers in Restaurants
1
Angela C Wills1, Kermit G Davis1, Susan E. Kotowski2
Low Back Biomechanics and Workplace Stress Laboratory, University of Cincinnati
2
Allied Health Sciences, University of Cincinnati
Serving staff are an integral part of restaurant establishments. Although over 2.2 million
individuals are employed as wait staff in the US and experience a considerable number of
musculoskeletal disorders each year, little is known about their work-related risk factors
for MSDs. Using observational methods, a physical activity monitor, and surveys, this
study quantified the physical demands experienced by servers over a shift. Twenty wait
staff were evaluated for postures used during serving, tray weight, pain symptoms, and
perceived workload. Results showed increases (upwards of 60%) in pain throughout their
shift, substabtial time spent standing or walking (~75%), and potentially risky arm/wrist
postures during serving, even though a majority of servers indicated lighter than normal
workloads during the observation period. While further research is needed, this study
provides a first glimpse at the physical demands experienced by wait staff.
Copyright 2013 by Human Factors and Ergonomics Society, Inc. All rights reserved. DOI 10.1177/1541931213571219
INTRODUCTION
Individuals working as waitresses and waiters in
the United States comprise a large portion of the
workforce, yet very little is known about the risk
factors that these workers routinely encounter. In
2012, the total number of waitresses and waiters
employed in the United States was 2,289,010 (BLS,
2011). Although little is known about risk factors
for this population, injuries are prevalent, with
waiters and waitresses reporting more than 1,190
musculoskeletal disorders (MSDs), representing an
incidence rate of 8.6 per 10,000 FTEs (BLS, 2011).
However, given the transient nature of the industry,
many of the servers may be pre-disposed to future
MSDs as they transfer into other professions.
Many of the traditional ergonomic risk factors
are thought to be commonly found for wait staff
such as lifting and transferring large, heavy, and
awkward loads, reaching for items, repetitious
tasks, and standing for long periods of time on their
feet oftentimes on concrete or non-pliant surfaces.
However, research is limited in actual
documentation of these factors and quantifying the
physical demands that servers undertake when
serving food to customers. The few studies that
have evaluated servers have concentrated on
specific tasks and the related musculoskeletal pain
(Dempsey and Filiaggi, 2006, Jones et al., 2003).
Chyuan and associates (2004) reported the demands
of serving staff in hotel restaurants.
The objective of the study was to quantify the
physical demands for servers for an entire shift. The
quantification included tray weight, frequency of
trays carried, and postures adopted when
transferring trays through direct observation;
amount of sitting and walking via functional
activity monitor, workload perception via survey,
and current body region symptom through selfreport survey.
METHODS
Study Overview: The research team observed
and measured the ergonomic stressors for wait
servers at three different restaurant locations in the
Midwest metropolitan area. The participants were
observed performing their normal job duties during
an entire shift.
Study Site: All three of the restaurant locations
specialized in their own style of cuisine. They were
all full-service restaurants, opened 7 days a week.
Restaurants 1 and 2 were privately owned and
restaurant 3 was a franchise.
Subjects: A total of 20 wait staff (5 males and
15 females) was recruited from three restaurants in
a Midwest metropolitan area. Inclusion criteria for
Downloaded from pro.sagepub.com at PENNSYLVANIA STATE UNIV on February 19, 2016
PROCEEDINGS of the HUMAN FACTORS and ERGONOMICS SOCIETY 57th ANNUAL MEETING - 2013
participation were the following: at least 2 or more
years of total experience serving, serving at current
location for more than 6 months, no major surgeries
within the last 2 years, not currently pregnant
(females), and being older than the age of 18 (which
was for convenience for consent but was not an
issue given all three restaurants served alcohol and
required servers to be 18 years).
Observations: Frequency of tray carrying and
postures of specific body joints were recorded on a
data worksheet. The worksheet had checklist entries
that recorded the number of times a tray was
transferred, what was being carried, and posture
categories for the neck, shoulder, hand and wrist,
elbows, upper back, lower back, hip, knee and
lower leg and foot. The posture ranges were based
on RULA categories (McAtamney and Corlett, 1993).
The tray weight was measured by having the server
step on a scale located in a convenient position near
the kitchen. The weight of the server and tray was
recorded and then converted into tray weight later
by subtracting the weight of the server.
Physical Activity Monitor: The amount of time
sitting, standing and walking was captured by a
physical activity logging device known as the
ActivPAL™ (PAL Technologies Limited, United
Kingdom). The ActivPal™ device was placed on
upper thigh of the server prior to the beginning of
each observed shift. The Activity device has been
found to provide accurate estimates of sitting,
standing, and walking (Aminian and Hinkson,
2012).
Workload Perception Survey: The Workload
Perception Survey assessed the whether the server
felt the observed shift was typical of the work
demands for their restaurant for the given day and
shift. The survey was composed of 7 questions that
ask about typical shift relative to the current one
being observed. The survey was developed
specifically for this study.
Current Symptom Survey: Participants
completed a Current Symptom Survey at the start
and end of the observed shifts. The survey assessed
current discomfort in various body regions on a
numeric analog scale of 0 to 10; 0 being no pain
982
experienced, 3 mild pain, 5 moderate pain, and 10
severe pain. The 9 body regions were neck,
shoulders, elbows, hands/wrists, upper back, low
back, hip, knees and lower legs, and feet.
Experimental Procedures: Upon arriving at
each of the restaurant locations, the logistics of the
study was explained verbally to the participants.
After confirmation of understanding the study, the
consent process was completed which included
reading and signing of the consent form by the
participant along with the completion of a
verification script by the PI. After the consent
process was completed, each participant started the
study by filling out the Current Symptom Survey.
Next, the ActivPAL™ was attached to upper thigh
of the participants and was worn for the entire
duration of the study and the waitress or waiter then
started their shift. At the end of the servers observed
shift, the ActivPal™ was removed from the
participants’ upper thigh. Once the ActivPal™ is
placed in a horizontal position it will stop recording
and logging the physical activity of the participant
and the data is then able to be transferred onto a
computer or laptop. Finally, the participants
complete the second Current Symptom Surveys as
well as the Workload Perception Survey.
Given the observational nature of the study, the
statistics were limited to descriptive summaries
including frequencies, averages and standard
deviations.
RESULTS AND DISCUSSION
Figure 1 shows the discomfort in the different
body regions at the start and end of the shift. The
largest increase in pain for any of the body regions
occurred in the upper back of 0.80 (55%). The other
regions with increased pain at the end of the shift
was the neck (increased by 0.65 or 45%), lower
back (increased by 0.60 or 50%), hand/wrist
(increased by 0.45 or 35%), and leg/foot (increased
by 0.40 or 60%). Two of the body regions actually
had a decrease in pain between the end and start of
the shift: shoulders had an average decrease of 0.10
(45%) and knees had a decrease of 0.05 (30%).
Overall, the demands of the job appeared to impact
the discomfort of the servers.
Downloaded from pro.sagepub.com at PENNSYLVANIA STATE UNIV on February 19, 2016
PROCEEDINGS of the HUMAN FACTORS and ERGONOMICS SOCIETY 57th ANNUAL MEETING - 2013
Based on the perceptions of the servers, only
40% reported that the observed shift was similar to
a typical day with the remaining 60% indicating the
observed shift was lighter than usual. This may
indicate that our observations represent a very
conservative estimate of the physical demands on
servers. The number of patrons typically served
during a normal shift was reported to be 1 to 3
people by 5%, 7 to 9 people by 25%, 10-12 people
by 30%, and 12 or more people by 40%.
The total amount of time the servers were
observed was 68.9 hours (or about 3.4 hours per
server). The total amount of weight per hour for
lunch and dinner service was 16.3 kg and 16.6 kg,
respectively. The number of trays was 2.5 per hour
and 2.9 per hour for lunch and dinner, respectively.
There were 93 trays (69.9%) carried in front of the
servers, 23 trays (17.3%) on the shoulder, 10 (7.5%)
trays using the hand flat techniques, and 7 (5%) of
the trays lifted with finger tip method. Based on the
perceptions that the shifts were generally light, the
amount of weight and number of trays carried may
be significantly under-represented. However, the
tray carrying technique likely represents the general
trends of the industry as technique appeared to be
server dependent,
The breakdown of the time spent sitting,
walking, and standing was 24.5%, 30.9%, and
44.6%. The total number of steps per hour per
server for all three restaurants was 609.8 steps.
Even with relatively slow observation periods, the
amount of standing and walking was substantial,
indicating a large potential for muscle fatigue for
the servers.
There were several considerations that need to
be considered when interpreting the results. The
study population was a sample of convenience in
that only one shift was monitored, a limited number
of servers were observed (N = 20), limited to 3
locations, and shift demands were less than
expected. These may have biased the results, in that,
the physical demands on servers may actually be
983
much greater than observed. The limited
observation time and number of location may have
the broadest ramifications as the observed physical
demands may vary tremendously. However, the
current study seems to conservatively estimate the
physical demands of wait staff.
CONCLUSION
With the transient workforce, servers are often
over-looked as an at risk population. Overall, this
study is the first step into understanding of physical
demands on servers during their shifts. There has
been little to no research done on this population.
While the current study did not find a substantial
overall burden, the loads handled and the postures
adopted could be a substantial risk for servers
working at very busy restaurants or during peak
hours.
REFERENCES
Aminia, S., Hinckson, E.A. (2012). Examining the
validity of the ActivPAL monitor in measuring
posture and ambulatory movement in children.
International Journal of Behavioral Nutrition and
Physical Activity 9, 119. Accessed March 21, 2013
http://www.biomedcentral.com/content/pdf/14795868-9-119.pdf
Bureau of Labor and Statistics, 2011,
http://www.bls.gov/news.release/osh2.t18.htm
Bureau of Labor and Statistics, 2012,
http://www.bls.gov/ooh/food-preparation-andserving/waiters-and-waitresses.htm
Chyuan, J.-Y.A., Du, C.-L., Yeh, W.Y. and Li, C.-Y.,
2004, Musculoskeletal disorders in hotel restaurant
workers. Occupational Medicine, 54, 55 - 57.
Dempsey, P.G., and Filiaggi, A.J., 2006, Cross-Sectional
Investigation of Task Demands and Musculoskeletal
Discomfort among Restaurant Wait Staff.
Ergonomics, 49, 93-106.
Jones, T., Strickfaden, M., Kumar, S. (2005) Physical
demands analysis of occupational tasks in
neighborhood pubs. Applied Ergonomics, 36 (5),
535-545.
McAtamney L and Corlett EN. RULA: a survey method
for the investigation of work-related upper limb
disorders. Applied Ergonomics 1993; 24:91-9.
Downloaded from pro.sagepub.com at PENNSYLVANIA STATE UNIV on February 19, 2016
Percetage of Servers with Discomfort (%)
PROCEEDINGS of the HUMAN FACTORS and ERGONOMICS SOCIETY 57th ANNUAL MEETING - 2013
70
60
50
40
30
20
10
0
Body Region
Before
After
Figure 1: Percentage of servers that had discomfort at the start and end of the shift being observed.
Downloaded from pro.sagepub.com at PENNSYLVANIA STATE UNIV on February 19, 2016
984