Content uploaded by Franklin Mirer
Author content
All content in this area was uploaded by Franklin Mirer on Sep 22, 2017
Content may be subject to copyright.
Occupational Injury Disparities in the
US Hotel Industry
†
Susan Buchanan, MD,MPH,
1
*Pamela Vossenas, MPH,
2
Niklas Krause, MD,PhD,
3
Joan Moriarty, MS,
4
Eric Frumin, MA,
4
Jo Anna M. Shimek, MS,
5
Franklin Mirer, PhD,CIH,
6
Peter Orris, MD,MPH,
7
and Laura Punnett, ScD
8
Background Hotel employees have higher rates of occupational injury and sustain more
severe injuries than most other service workers.
Method OSHA log incidents from five unionized hotel companies for a three-year period
were analyzed to estimate injury rates by job, company, and demographic characteristics.
Room cleaning work, known to be physically hazardous, was of particular concern.
Results A total of 2,865 injuries were reported during 55,327 worker-years of observa-
tion. The overall injury rate was 5.2 injuries per 100 worker-years. The rate was highest for
housekeepers (7.9), Hispanic housekeepers (10.6), and about double in three companies
versus two others. Acute trauma rates were highest in kitchen workers (4.0/100) and
housekeepers (3.9/100); housekeepers also had the highest rate of musculoskeletal
disorders (3.2/100). Age, being female or Hispanic, job title, and company were all
independently associated with injury risk.
Conclusion Sex- and ethnicity-based disparities in injury rates were only partially due to
the type of job held and the company in which the work was performed. Am. J. Ind. Med.
2009. 2009 Wiley-Liss, Inc.
KEY WORDS: occupational injury; hotel workers; housekeepers; musculoskeletal
disorders; health disparities
BACKGROUND
Health disparities between the sexes and between racial/
ethnic groups have been documented for a wide spectrum of
diseases [Satcher and Higginbotham, 2008] but research on
disparities in the rates of injuries and diseases occurring in the
workplace is still emerging. Recent studies have shown that
Hispanic workers have the highest rate of fatal and non-fatal
OSHA-reported injuries in the US, followed by black non-
Hispanic workers [Richardson et al., 2003; USBLS, 2007a].
Among agricultural and hospital workers, a disproportionate
burden of occupational injury is carried by women, African
Americans, and Latinos [McGwin et al., 2000; Simpson and
Severson, 2000; McCurdy et al., 2003]. Elevated risks among
these groups are partially explained by disproportionate
employment in high-risk industries and occupations, but
there may also be disparities within the same industry or
job classification, perhaps resulting from sex, racial, or ethnic
discrimination and other factors.
Accepted 22 May2009
DOI10.1002/ajim.20724.Published online in Wiley InterScience
(www.interscience.wiley.com)
Contract grantsponsor: UNITEHERE.
*Correspondenceto:SusanBuchanan, MD,MPH,835 S.Wolcott,MC-684,Chicago,IL60612.
E-mail: sbucha3@uic.edu
1
Division of Environmental and Occupational Health Sciences, University of Illinois at
Chicago S chool of Public Health, Chicago, Illinois
2
Occupational S afety and Health Program,UNITE HERE,New York,New York
3
Division of Occupational a nd Environmental Medicine, University of California San Fran-
cisco, San Franciso, Cali fornia
4
WorkersUnited/SEIU, New York, NewYork
5
Division of Environmental and Occupational Health Sciences, University of Illinois at
Chicago S chool of Public Health, Chicago, Illinois
6
Environmental and Occupational Health Sciences, Urban Public Health Program,Hunter
College School of Health Sciences, New York,New York
7
Departmentof Occupation al and Environmental Medicine,University of Illinois at Chicago
Medical Center,Chi cago, Illinois
8
Department of Work Environment, University of Massachusetts Lowell, Lowell, Massa-
chusetts
†
Workcon ducted while Joan Moriar ty and Eric Frumin were at UNITEHERE.
AMERICAN JOURNAL OF INDUSTRIAL MEDICINE (2009)
2009 Wiley-Liss,Inc.
Within the US hospitality industry, hotels, and motels
employ 1.8 million workers [USBLS, 2007b]. In the United
States, hotel workers are nearly 40% more likely to be injured
on the job than all other service sector workers. Hotel
workers also sustain more severe injuries resulting in
more days off work, more job transfers, and more medically
restricted work compared to other employees in the
hospitality industry [USBLS, 2005].
Approximately 25% of hotel workers are employed in
housekeeping departments [USBLS, 2007b]. Housekeepers
constitute the single largest occupational group in the
hotel industry and include room cleaners (maids or room
attendants) and housemen. Many room attendants are immi-
grant or minority women, with a majority being either Asian,
Latin American, or African American [Wial and Rickert,
2002]. Thus, they belong to several groups that have been
repeatedly identified as having excessive occupational
risks: women [Stellman, 1999; NIOSH, 2002; Kauppinen
et al., 2003; Messing, 2004; Treaster and Burr, 2004],
immigrants [Improving Health and Safety Conditions for
California’s Immigrant Workers, 2002], ethnic/racial minori-
ties [Frumkin et al., 1999], and low-wage workers [Frumkin
and Pransky, 1999]. However, very little is known about
occupational injuries among hotel housekeepers; the US
Bureau of Labor Statistics (BLS) does not provide rates of
occupational injury and illness for single occupations. Among
Las Vegas hotel room cleaners, the prevalence of self-reported
pain associated with work was 75% during the previous year
[Scherzer et al., 2005]; 63% had had severe or very severe low
back pain just in the prior month [Krause et al., 2005].
In 1996, the first National Institute for Occupational
Safety and Health (NIOSH) research agenda (“NORA”)
called for innovative occupational health research to deter-
mine the extent and severity of disease and injury among
special worker populations [NIOSH, 1996]. Ten years later,
the revised NORA research agenda targeted the service
sector, which accounts for 80% of the US workforce.
Hotel workers have been repeatedly identified as an under
-researched population with significant problems such as
musculoskeletal injuries; even less is known about dish-
washers, cooks, and other food service workers.
This study analyzes the rates of OSHA-reported injury
within the hotel industry for four leading hotel job categories
(hotel housekeepers, cooks/kitchen workers, stewards/
dishwashers, and banquet servers), and examines disparities
in injury risk by race/ethnicity and sex.
METHODS
Study Population
Institutional Review Board approval was obtained from
the University of Illinois at Chicago under the “exempt”
classification. The study population consisted of non-
supervisory hotel workers employed for a minimum of
2 weeks in at least 1 year during the study period of
2003–2005, at full-service hotels operated by the five
largest hotel companies in the United States. For this study,
full-service hotels are defined as properties with at least 100
guest rooms and with a minimum of 10,000 square feet of
conference space. These criteria were intended to increase
the likelihood that job classifications and workplace expo-
sures to ergonomic and safety hazards would be similar.
Luxury chains were excluded because the design and pace of
work varies significantly at these properties.
The five companies operate several hotel chains that
together make up over 70% of the full-service hotel rooms
nationwide, with each company establishing its own
standards of service. According to information found on
the companies’public websites in February 2007, these
companies operate 964 hotel properties in the US that meet
the study’s definition of full-service hotels. UNITE HERE,
the largest hospitality workers union in North America,
represents workers at many of these hotels.
Hotel Sampling
Upon request from the union, 71 of the hotels with
collectively bargained contracts provided data, which could
be utilized for this study. The two largest companies repre-
sented an unbalanced proportion of the sample, so a random
number generator [Research Randomizer, 1997–2008] was
used to select 12 hotels from each of these two. All hotels
from the three other companies were included in the data
analysis. This produced a sample of 50 hotels with sufficient
data from 2003 to 2004 and 45 from 2005 (Table I). Study
hotels were dispersed across the country with concentrations
in large urban areas including New York City, Chicago, San
Francisco, Los Angeles, and Honolulu.
Job Classifications
Job titles are numerous within hotel departments andvary
from employer to employer. The authors in collaboration with
TAB L E I. HotelCompany Distributionsof US Full-Service HotelsandHotelsin
the Study Sample
Company
Full-servicehotels Study sample
No.% No.%
Company1 334 35 12 24
Company 2 95 10 12 24
Company 3 10 1 5 10
Company 4 319 33 9 18
Company 5 206 21 12 24
Totals 964 100 50 100
2 Buchanan et al.
experienced union field staff familiar with the specific job
titles, grouped the jobs that share similar tasks and exposures
to workplace hazards (e.g., “dishwasher”and “pot washer,”
“housekeeping attendant”and “room attendant”). Five key job
categories were created—housekeepers, banquet servers,
stewards/dishwashers, cooks/kitchen workers, and “other.”
Housekeepers perform guest room cleaning including
making beds, vacuuming floors, cleaning shower walls and
bathroom fixtures, dusting furniture, and pushing carts.
Banquet servers provide food service such as carrying plated
food from the kitchens to the customers, dispensing drinks,
and supplying food to cafeteria and buffet services. Stewards
retrieve, sort, load/lift, unload, and return dishes, glasses,
pots, utensils and silverware, and provide these items by
pushing carts to cafeteria and buffet lines. In addition,
stewards maintain cleanliness in food preparation areas.
Cooks lift, weigh, measure, mix, cut and grind food ingre-
dients; they cook these ingredients and compose salads and
other food for serving [USBLS Occupational Outlook Hand-
book, 2008–2009]. All remaining jobs were categorized as
“other.”Jobs classified as “other”were those that did not
share similar job tasks or exposures with the other four key
job categories. These included lobby attendant, cashier, door
person, host/hostess, among others.
Database Creation
Employee rosters and OSHA 300 log data were provided
to the union by the five hotel companies for the period
2003–2005. The employee rosters provided employee name,
department, job title, date of birth, date of hire, termination
date, sex, and race/ethnicity. Race/ethnicity was defined by
the employer based on employee self-report as one of
the following five mutually exclusive categories: American
Indian, Asian, Black, Hispanic, and White.
The OSHA 300 logs included employee name, depart-
ment name or location where injury event occurred, job title,
date of injury, injury description, days away from work, and
days on restricted duty. These data were matched to the
employee rosters using employee name and date of birth. The
final dataset included a single record for each employee. Up
to three injury or illness incidents during the 3-year study
period were abstracted for each individual. Employee names
were removed from all datasets before data analysis began. A
record number was assigned to each injury incident and was
subsequently used in all data analyses.
Injury Coding
Nature of injury data was constructed from the injury
description section of OSHA log entries and were grouped by
the authors into four categories: musculoskeletal disorders
(MSDs), acute trauma injuries, other, and not classifiable.
MSDs were coded according to the US BLS definition: “an
injury or disorder of the muscles, nerves, tendons, joints,
cartilage, or spinal discs. MSDs do not include disorders
caused by slips, trips, falls, motor vehicle accidents, or
similar accidents”[USBLS, 2007c]. Back pain or pain at
other body locations and strain or sprain injuries were coded
as MSDs unless the entry referenced stairs or ladders, or the
employer-reported description of the injury referenced a slip
or fall. “Acute trauma”cases included contusions, fractures,
lacerations, heat burns, and sprain or strain injuries with
evidence of an injury mechanism that involves acute contact
with outside objects (e.g., hit by, struck against) that were not
otherwise categorized as an MSD. “Other”incidents includ-
ed chemical exposures, foreign bodies in the eye, and all
other cases. “Not classifiable”injuries had insufficient infor-
mation to determine the nature of injury.
Statistical Analysis
All data were analyzed using SAS (SAS v. 9.1, 2007.
SAS Institute, Cary, NC) and Excel (Microsoft Office 2003,
Seattle, Washington). Injury rates and risk ratios were calcu-
lated to compare the injury experience of hotel workers by
sex, race/ethnicity, and job title for the entire study popula-
tion and by company. The denominator for all calculations
was calculated from the number of workers who met the
inclusion criterion of employment for a minimum of 2 weeks
during each year of study. As individual employees may be
counted in more than one study year, the denominators
represent total worker-years of observation. The available
data did not provide information on part-time/full-time
status. The race and ethnicity characterization was left blank
on the employee rosters for <1% of the sample. Therefore,
this race/ethnicity “not classified”group was excluded from
all data analyses.
Age was computed by subtracting birth date from the last
day of the year being analyzed (e.g., in 2003, Age ¼12/31/
2003 birth date) divided by 365.25. Only employees aged
18–70 years were included in the analysis. A job tenure
variable was similarly created by subtracting termination
date from hiring date.
Risk ratios were calculated using the following referent
groups: males, whites, and “other”job title. For analyses by
hotel company, Company 1 was chosen as the referent group
on the basis of the level of union presence at its hotels,
thereby a measure of labor and management’s negotiation of
working conditions.
Because we had injury count data and repeated measures
(multiple years per subject), we performed multivariable
Poisson regression modeling (Loomis et al. 2005) with
generalized estimating equations (GEE) using SAS Proc
Genmod with a Poisson distribution, unstructured correla-
tions and log link to estimate relative risk. Regression
models included age (18–27 years, 28–37 years, 48–57 years,
58–70 years), sex, race/ethnicity, job title, job tenure (0–10
Occupational Injury in Hotel Workers 3
years, 11–20 years, 21–30 years, 31–40 years, 41–52 years),
and hotel company as independent variables. In addition,
cross tabulation and regression modeling were perform-
ed within the subset of female housekeepers. Similar
analyses were not conducted within other subsets of other
job classifications; female housekeepers were a particularly
large subset.
RESULTS
There were a total of 55,327 worker-years of observation
in the sample. Fifty-six percent of the sample was male and
44% female (Table II). By job title, 21% of the employees
were housekeepers, 11% were banquet servers, 6% were
stewards/dishwashers, 8% were cooks/kitchen workers, and
54% had other jobs. Most of the workers were non-white
(Black, Asian, Hispanic), comprising 80% of the sample.
American Indians and male housekeepers were very few in
number. Hispanics comprised the largest proportion of three
job titles: housekeepers, stewards, and cooks. The mean age
of the study population was 44.5 years (SD 13.5). The mean
job tenure was 9.61 years (SD 8.8).
There were 2,865 injuries recorded on the OSHA
300 logs in 2003–2005 (Table III), for an injury rate of
5.2 injuries per 100 worker-years. Acute trauma accounted
for 52% of the injuries, 39% were musculoskeletal injuries,
and 9% were “other”or “not classifiable.”Women workers
had a higher overall injury rate (6.3) than men (4.3).
Housekeepers had the highest overall injury rate and the
highest rate of MSDs, at 7.9 and 3.2 per 100 workers,
respectively. Acute trauma rates were highest in cooks/
kitchen workers and housekeepers. Banquet servers had the
lowest injury rates. Excluding the six injuries among
American Indians, among housekeepers (Table IV), Hispanic
workers had the highest overall injury rate at 10.6, the highest
rate of MSDs (4.4), and the highest rate of acute traumas
(4.9). Among cooks (not shown), Asians had the highest rate:
8.4% for all injuries, with 7.9% among males and 10.1%
among females.
In each job title of interest (housekeepers, etc.), injuries
of the upper extremity were the most common, followed by
back injuries and lower extremity injuries. By nature of
injury, over 40% of MSDs involved the back, 22% distal
upper extremities, and 13% the shoulder. In contrast, 44% of
acute traumatic incidents were to the upper extremity,
especially the hand.
Women workers overall and Asian and Hispanic men
were about 1.5 times more likely to have been injured than
their referent groups (Table V). Female American Indians
fared the worst, although the number of injuries were so few
that the confidence intervals are relatively wide. Hispanic
women had almost double the risk of injury than their white
female counterparts. Within job categories, non-white
female cooks/kitchen workers fared poorly compared to their
white counterparts as did non-white male banquet servers.
Female housekeepers had about three times the risk of injury
than male housekeepers, and Hispanic housekeepers were
70% more likely to be injured than white female
housekeepers.
When analyzed by hotel company, the overall injury
rates differed markedly by company, with companies 2, 3,
and 4 in particular having almost twice the rate of Company
1 (Table VI). Company 2 had the highest rate of injury for
housekeepers (10.4). This overall effect was consistent in
analysis by injury type, with the lowest rates for both MSDs
and acute trauma injuries in Company 1. These same patterns
by company were also evident for key demographic groups
within the four key jobs. Of the 15 job/race/sex groups with
sufficient cases for comparison, Companies 2 and 3 had the
highest injury rates for five of them and Company 4 had
almost as many. Company 1 had only one such group, and
Company 5 had none.
TAB L E I I. Demographic Breakdown of Hotel Workers* Employed 2003–2005 in 50 Unionized Full-Service Hotels(n ¼55,327)
Total Housekeeper Banquetserver Steward/dishwasher Cook/kitchenworker Other jobs
No.%No.%No.%No.%No.%No.%
Male 31,135 56.4 269 2.3 3,406 66.8 2,948 85.1 3,269 72.0 20,280 69.2
Female 24,048 43.6 11,320 97.7 1,693 33.2 518 14.9 1,271 28.0 9,008 30.8
White 11,187 20.3 982 8.4 2,137 36.8 286 8.1 882 19.3 6,898 23.3
Asian 13,352 24.2 3,109 26.7 909 15.6 594 16.9 1,202 26.3 7,538 25.4
Black 12,252 22.2 3,439 29.5 712 12.3 962 27.3 872 19.0 6,267 21.1
Hispanic 18,392 33.3 4,118 35.3 2,047 35.3 1,678 47.7 1,622 35.4 8,927 30.1
American
Indian
14 4 <112<132<17<110<183<1
Total (%)
a
55,327 100.0 11,660 21.1 5,837 10.5 3,527 6.4 4,588 8.3 29,713 53.7
*
Total person-years observed, not total employees.
a
Total excludes race “not specified”(<1%of total).
4 Buchanan et al.
The regression analyses of all hotel workers (Table VII)
confirmed the higher injury risk for housekeepers and His-
panic workers, and the lower risk in Company 1, after
adjusting for demographic characteristics. Comparison of
univariable and multivariable models showed that some of
the apparent excess risk in Black, Hispanic, and Asian
workers was reduced after adjustment for job title and hotel
company. This was consistent with the fact that Blacks
were most likely (30%), and Whites least likely (8%), to be
employed as housekeepers rather than in other jobs, and that
Company 1 had fewer Black and Asian employees. Job
tenure had a slight inverted-U effect (risk was highest for
21–30 years of seniority and then decreased) but it was
dropped from the multivariable models because the coeffi-
cient was very small, the confidence intervals wide, and the
type 3 (GEE) score statistics indicated that the variable did
not contribute any explanatory power. Among female house-
keepers, the predictors of injury were quite similar to those
for all hotel workers, with increased risk for being Hispanic
or employment at Companies 2, 3, and 4.
DISCUSSION
Several studies have shown that cleaning tasks in various
industries demand a high level of physical effort, including
high aerobic strain and repetitive movements [Hagner and
Hagberg, 1989]; high static muscular loads [Milburn and
Barrett, 1999]; high frequency of unsatisfactory postures
such as stooping and crouching [Woods et al., 1999]; and
subjective experience of strenuous work [Sogaard et al.,
1996; Seifert and Messing, 2006]. In hotel workers specifi-
cally, guest room cleaning work is marked by time pressure,
low job control, low wages, increasing use of contingent
employees without job security, and few opportunities for
career advancement [Parker, 1999; Lee and Krause, 2002;
Wial and Rickert, 2002; Bernhardt et al., 2003; Krause et al.,
2005]. The present study is one of the first to quantify the
incidence, rates, and risk of injury among hotel workers.
We found that women were more often injured than men
and that housekeepers in general suffered the highest injury
rate among the four job titles of interest. Moreover, our
results show an alarming injury rate among housekeepers
in general and Hispanic housekeepers in particular. While
close to half of the total workers here are women, they were
heavily grouped in the housekeeping category, a set of jobs
with very high physical demands. This study strengthens the
evidence that job gender stereotyping within the American
economy remains a potent defining factor for the workforce
and potentially a substantial risk factor for injury [Mergler,
1995; Messing et al., 1998, 2003; Punnett and Herbert, 2000].
Socioeconomic status (SES) in general, and income
inequality, education, and job-specific occupational hazards
in particular, have all been proposed as possible explanations
for racial/ethnic as well as gender health disparities. There is
TABLE III. Injury Incidencean d Rates*for the Hotel Worker Study Population,by Sex and JobTitle, 2003–2005
Total Male Fema le Hou sekee per Banqu et ser ver Ste ward/dis hwasher Cook /kitche n worker Oth er jobs
a
Inj no. R ate Inj no. Rate In j no. Rate Inj no . Rate Inj no. Rat e Inj no. Rate Inj n o. Rate Inj no. Rat e
MSDs 1,117 2.02 525 1.68 592 2.46 368 3.16 63 1.08 70 1.99 80 1.74 536 1.82
Acute
trauma
1,497 2.71 709 2.28 788 3.28 456 3.91 94 1.62 116 3.30 182 3.98 649 2.19
Otherinjuries 251 0.45 110 0.35 141 0.59 93 0.80 7 0.12 24 0.68 12 0.26 115 3.88
Total injuries 2,865 5.19 1,344 4.32 1,521 6.32 917 7.87 164 2.82 210 5.97 274 5.99 1,300 4.92
*
Injury rate is number of cases per 100 person-years.
a
Injuries that were “not classifiable”were collapsed into “other”jobs.
Occupational Injury in Hotel Workers 5
consistent epidemiologic evidence that low status jobs
are associated with a high burden of disease, injury, and
disability [Robinson, 1989; Krause et al., 1997, 2001; Amick
et al., 1998; Borg and Kristensen, 2000; Pransky et al., 2000;
Berkman and Kawachi, 2002; d’Errico et al., 2007]. This
burden falls disproportionately on workers who are multiply
disadvantaged in society and who have been under-repre-
sented and under-served in occupational health research.
Female immigrant cleaners are a typical example of a
minority population at the low end of the well-established
SES gradient.
As yet, there has been no evaluation of the causes of
differential injury rates by race/ethnicity within job title in
this industry. One must question whether discrimination in
the treatment of such workers—in the form of dispropor-
tionate assignment to high-risk jobs, refusal to fix unsafe
conditions, or workers’disempowerment—resulting in
unwillingness to speak up about such conditions, is at fault.
As Murray [2003] noted, previous studies have observed
informal systems of work assignments to non-white workers
resulting in greater exposures to the hazards therein. More-
over, US BLS has already found that disproportionate em-
ployment of Hispanics in specific jobs is not associated with
increased risk of injury after controlling for such employ-
ment patterns [Richardson et al., 2003]. In essence, race/
ethnicity itself is not an indicator of increased risk.
The injury rate for the workers in this sample was
5.19 per 100 workers. For 2004, the US BLS reported a rate
of 5.8 per 100 FTEs in hotel workers and 4.2 per 100 FTEs in
the service sector overall. The lower overall injury rate
reported in our sample may be due to the inability to identify
the proportion of part time workers in this sample or that
unionized employees work under conditions defined by
collective bargaining agreements, which are intended to
improve workplace safety. The study sample included only
unionized workers, whereas the majority of US hotel
employees do not belong to unions. Since unions function
as the bargaining agent between the employer and the
employee, it is likely that non-unionized hotels, in which
workers do not have a formal means to gain better working
conditions, would have even higher injury rates than those
reported in this study. Further, it is possible that hotels not
providing data were those at which workplace safety is less of
a priority and which have higher injury rates than those
reported here.
These results also need to be seen in the context of the
tendency of many workers not to report their injuries, espe-
cially if they are non-unionized, immigrants, or otherwise
politically vulnerable [Azaroff et al., 2002, 2004; Brown
et al., 2002; Scherzer et al., 2005]. Non-reporting of injuries
may be due to language barriers, fear of retaliation, or lack of
understanding of legal rights under Workers Compensation
laws and OSHA standards. Although our data represent
unionized workers who reported their injuries, the results
may still represent an under-estimation of the true injury risk.
Other possible limitations to this study include quality
of the data, coding, and job grouping errors. Injury data
obtained from OSHA 300 logs may have contained inaccu-
racies. The individual responsible for completing these logs
varies by workplace and is not always well trained in
correct recording procedures. There may well be systematic
differential approaches to OSHA 300 log completion by
different hotel companies. Nevertheless, we saw no evidence
of frequent recording errors or systemic bias in recording
through regular quality control checks as well as consulta-
tions with experts on the coding and grouping criteria.
Although the high rate of acute injuries in housekeepers
may suggest coding errors, the OSHA logs frequently
included event/exposure data such as contact with
furniture, tripping over sheets, slips in bathtubs, etc. Further-
more, coding error is possible since some acute injuries in
housekeeping may have been MSDs. However, the patterns
of injury we found are also seen in US BLS data.
The hotels in this study sample were included based on
number of rooms and size of meeting space in order to ensure
similarity in job task burden among workers in the sample.
Working conditions in full-service hotels are determined and
standardized in major part by corporate-level policies such as
TAB L E I V. Injury Incidence and Rates* for Housekeepers by Race/Ethnicit y, 2003–2005
All injuries MSDs Acute trauma Other/not classifiable
Inj no. Rate Inj no. Rate Inj no. Rate Inj no. Rate
Asian 228 7.33 102 3.28 106 3.41 20 0.64
Black 189 5.50 58 1.69 113 3.29 18 0.52
Hispanic 435 10.56 183 4.44 203 4.93 49 1.19
White 62 6.31 24 2.44 32 3.26 6 0.61
American Indian 6 50.00 1 8.33 5 41.67 None
Total
a
920 7.89 368 3.16 459 3.94 93 0.80
*
Injury rate is number of cases per100 person-years.
a
Total excludes race “not specified”(<1%of total).
6 Buchanan et al.
TAB L E V I. Injury Incidence Rate*, and Rate Ratio for the Hotel Worker Study Population,by JobTitle and Hotel Company, 2003–2005
Job titles
Company1
a
Company 2 Company 3 Company 4 Company 5
# Inj Rate # Inj Rate RR (95 % CI) # Inj Rat e RR (95% CI) # Inj Rate RR (95% CI ) # Inj Rate RR (95 % CI)
Housekeeper 211 5.47 276 10.36 1.93 (1.59–2.34) 86 9.67 1.78 (1. 37–2.32) 211 9.44 1.74 (1.41–2.13) 135 6.18 1.13 (0.89 –1.43)
Banquet Server 5 n.a. 56 3.69 n.a. 14 3.97 n.a. 69 4.33 n.a. 20 4.25 n.a.
Steward/
dishwasher
51 4.63 60 7.15 1.55 (1.04–2 . 31 ) 3 2 11.19 2.48 (1.48 –4.14) 4 5 9.15 1.99 (1.29–3.08) 22 2 .60 0.56 (0.34–.93)
Cook/kitchen
worker
47 3.90 88 7.48 1.94 (1.35 –2.79) 26 12.32 3.29 (2.01–5.40) 59 6.54 1.68 (1.15 –2.46) 56 4.94 1.27 (0.86–1.8 9 )
Other workers 258 2.72 317 5.72 2.10 (1.77–2.50) 140 6.23 2.31 (1.8 4 –2.89) 354 5.54 2.04 (1.72–2.42) 232 3.72 1.3 7 (1.13–1.6 5 )
All jobs 572 3.26 797 6.79 2.10(1.87–2.36) 298 7.48 2.33 (1.99–2.72) 738 6.36 1.95 (1.74–2.20) 465 4.28 1.31(1. 15 –1.49)
n.a., insufficient data.
*
Injury rate is the number of injuries per100 person-years.
a
Company 1is the referent group for all other companies.
Statistically significant results are shown in bold.
7
TAB L E V. Injury Rate Ratios* for the Hotel Worker StudyPopulation by JobTitle, Sex, and Race/Ethnicit y, 2003–2005
Males Females
All females American Indian Asian Black Hispanic American Indian Asian Black Hispanic
Job t itle RR(9 5% CI) RR (95% CI) RR (95% CI) RR (95% CI) RR (95% CI) RR (95% CI) RR (95% CI) RR (95% CI) RR (95% CI)
All hotel workers 1.46 (1.35–1. 57 ) 0 .4 1 (0 .0 6 –2.87) 1.52 (1.28–1.8 2) 1.07 (0. 87–1.3 2 ) 1.54 (1.30–1. 8 2 ) 2.19 (1.08–4.46) 1.39 (1.15–1.6 7 ) 1. 14 ( 0 . 9 4 –1.3 8 ) 1.91 (1. 6 –2.27)
Housekeepers 3.19 (1.53–6.64) n.a. n.a. n.a. n.a. 4.00 (1.65–9.67) 1.19 ( 0.87–1.62) 0.87 (0.63–1.2 0 ) 1.70 (1.26–2.29)
Banquet servers 1.38 (1.00–1.89) n.a. 1.65 (n.a.) 1.87 (n.a.) 2.02 (n.a.) n.a. 0.66 (n.a.) 1.20 (n.a.) 1.14(n.a.)
Stewards/
dishwasher
1. 4 2 ( 1. 0 0 –1.97) n.a. 1.29 (n.a.) 1.46 (n.a.) 1.78 (n.a.) n.a. n.a. 0.42(n.a.) 0.45 (n.a.)
Cook/kitchen worker 1. 34 (1.0 4–1.72) n.a. 1.42 (n.a.) 0.51(n.a.) 0.89 (n.a.) n.a. 2.77 (n.a.) 2.20 (n.a.) 1.94 (n.a.)
Other workers 1.05 (0.93–1.19 ) 0. 7 5 ( 0 .11 –5.21) 1.39 (1.12–1. 7 3 ) 0. 9 5 ( 0 . 7 4 –1.2 2 ) 1.48 (1.21–1. 8 1 ) 1. 8 8 ( 0 . 7 0 –5.09) 1.11 (0.82–1.5 0 ) 1. 0 0 ( 0 . 7 3 –1.3 7 ) 1. 44 (1.08–1. 9 3 )
n.a., insufficient data.
*
Referent groups:Males are referent group for females; white males are referent groupfor American Indian, Asian, Black, and Hispanic males; white females are the referent groupfor American Indian, Asian, Black, and Hispanic females.
Statistically significant results are shown in bold.
job task lists and the use of branded products such as luxury
beds. Hotels with fewer than 100 rooms would be less likely
to have standardized room quotas, which might affect work-
load pressure and therefore injury risk among housekeepers.
Thus, we believe that the inter- and intra-hotel variations in
work tasks among job title groups are likely to be minimal in
our sample of properties.
There were substantial and consistent differences in
injury rates among the five companies. These differences
persisted for all injuries, for injuries by job title, and by
demographic groups. As this study sought to standardize job
tasks between companies, this differential suggests the
influence of management policies and practices, meaning
that workplace intervention has a significant ability to modify
the risks identified in this study. These marked differences
between companies demonstrate the potential for sharp
improvement by individual companies in injury rates. They
also underscore the need for companies with high rates to
investigate whether discriminatory workplace practices
contribute to these disparities—in order to remedy the dis-
crimination and reduce the injury risk accordingly.
CONCLUSION
Injury rates for hotel workers are higher than those in the
service sector as a whole. Characteristics that increased the
injury risk among the workers in our study included female
sex, Hispanic ethnicity, housekeeper job title, and hotel
company. Hispanic banquet servers had the highest risk
amongst men, and American Indian housekeepers had the
highest risk among women. Hispanic female housekeepers
suffered more injuries than other female room cleaners.
Immediate action is needed with respect to the control of
hazards to housekeepers, especially those stressing the upper
extremities, and to food service workers with respect to
acute trauma. The ethnic, gender, and employer differentials
deserve further exploration to adequately understand the
interaction of social forces with ergonomic and safety
hazards in the workplace. Large differences of injury rates
between employers indicate a substantial potential for injury
prevention in the hotel sector.
ACKNOWLEDGMENTS
Financial support for the data analysis was provided by
UNITE HERE. The following individuals contributed to this
study: Christopher E. Mason, PhD, employee hiring list data
preparation; Emily Perry, BA, research on full service hotel
sector; and John M. Halpin, MD, MPH, housekeeper injury
analysis of earlier version of dataset. We are grateful to
Rebecca Gore, PhD for statistical advice.
TAB L E V I I. Regression Modelsof Injuries Per Year* to US Unionized Hotel workers, 2003–2005: RiskRatios and 95% Confidence Intervals
Unadjustedmodels
(all ho tel wor kers)
Multivariable model
(all ho tel wor kers)
Multivariable model
(all ho tel wor kers)
Multivariable model
(female housekeepers)
Odds ratio 95% CI Odds ratio 95% CI Odds ratio 95% CI Odds ratio 95% CI
Age 1.07 1.04–1.0 9 1. 0 8 1. 0 5–1. 11 1. 0 9 1. 0 6 –1.12 1. 10 1.0 3 –1.18
Job tenure 1.08 1.04–1. 12
Female 1.46 1.35–1. 5 8 1. 2 4 1.12 –1. 3 7 1. 2 1 1. 0 9 –1.3 4
American Indian 1.35 0.67–2.72 1.33 0.68–2 . 61 1.15 0 .6 0 –2.22 2.54 1.05–6.13
Asian 1.46 1.29–1.6 7 1. 2 5 1. 10 –1.4 2 1. 11 0 . 9 7 –1.26 0.97 0.71–1. 3 3
Black 1.15 1.00–1.32 0.97 0.84–1.11 0.85 0.74–0.98 0.75 0.54–1. 0 3
Hispanic 1.70 1.50–1.9 2 1. 5 0 1. 3 3 –1. 7 0 1. 4 2 1. 2 6 –1.6 1 1. 5 0 1.11 –2.02
Housekeeper 1.80 1.65–1. 97 1. 5 0 1. 3 4 –1.6 8 1. 5 2 1. 3 6 –1.7 0
Banquet server 0.64 0.54–0.77 0.60 0.50–0.72 0.56 0.47–0.67
Steward/
dishwasher
1. 3 7 1.1 7 –1.6 1 1. 3 0 1. 11 –1. 5 3 1. 3 1 1. 12 –1.5 4
Cook/kitchen
worker
1. 3 8 1. 2 0 –1. 5 8 1. 3 4 1. 17 –1. 5 4 1. 3 1 1.15 –1. 51
Company 2 2.10 1.87–2.36 2.17 1.94–2.44 1.94 1.59–2.35
Company 3 2.33 1.99–2.72 2.41 2.07–2.81 1.84 1.41–2.39
Company 4 1.95 1.74–2.20 2.06 1.83–2.32 1.74 1.41–2.14
Company 5 1.31 1.15–1. 50 1. 3 7 1. 2 0 –1.5 6 1. 19 0 .9 4 –1. 5 0
Male is the referent groupfor female; White is the referentgroup for Black, Hispanic, Asian, and American Indian; “Other jobs”is the referent groupfor housekeeper, banquet server,
steward, and cook/kitchen worker; Company 1 is the referent group.
Up to three injuries per year per employee; denominators ¼55,311person-years of observation for all hotel workers an d 11,375 person-years for female housekeepers.
8 Buchanan et al.
REFERENCES
Amick BC III, Kawachi I, Coakley EH, Lerner D, Levine S, Colditz GA.
1998. Relationship of job strain and iso-strain to health status in a cohort
of women in the United States. Scand J Work Environ Health 24(1):
54–61.
Azaroff L, Levenstein C, Wegman DH. 2002. Occupational injury and
illness surveillance: Conceptual filters explain underreporting. Am
J Public Health 92:1421–1429.
Azaroff LS, Lax MB, Levenstein C, Wegman DH. 2004. Wounding
the messenger: The new economy makes occupational health
indicators too good to be true. Int J Health Serv 2004. 34:271–
303.
Berkman SF, Kawachi I. editors. 2002. Social epidemiology.New York:
Oxford University Press.
Bernhardt A, Dresser L, Hatton E. 2003. The coffee pot wars: Unions
and firm restructuring in the hotel industry. In: Appelbaum E, Bernhardt
A, Murnane RJ, editors. Low-wage America: How employers are
reshaping opportunity in the workplace. New York: Russell Sage
Foundation, p 33–76.
Borg V, Kristensen TS. 2000. Social class and self-rated health: Can the
gradient be explained by differences in life style or work environment?
Soc Sci Med 51(7):1019–1030.
Brown MP, Domenzain A, Villoria-Siegert N. 2002. California’s im-
migrant workers speak up about health and safety in the workplace. Los
Angeles, California: University of California Los Angeles Labor
Occupational Safety and Health Program Health and Safety Policy
Brief.
d’Errico A, Punnett L, Cifuentes M, Boyer J, Tessler J, Gore R, Scollin P,
Slatin C. 2007. Hospital injury rates in relation to socioeconomic status
and working conditions. Occup Environ Med 64(5):325–333.
Frumkin H, Pransky G. 1999. Special populations in occupational
health. Occup Med 14(3):479–484.
Frumkin H, Walker ED, Friedman-Jimenez G. 1999. Minority workers
and communities. Occup Med 14(3):495–517.
Hagner IM, Hagberg M. 1989. Evaluation of two floor-mopping work
methods by measurement of load. Ergonomics 32(4):401–408.
Improving Health and Safety Conditions for California’s Immigrant
Workers. Report and Recommendations of the California Working
Immigrant Safety and Health Coalition. November 2002. Berkeley,
California.
Kauppinen K, Kumpulainen R, Copsey S. 2003. Gender issues in safety
and health at work—A review. Finland: European Agency for Safety
and Health at Work.
Krause N, Lynch J, Kaplan GA, Cohen RD, Goldberg DE, Salonen JT.
1997. Predictors of disability retirement. Scand J Work Environ Health
23:403–413.
Krause N, Frank JW, Dasinger LK, Sullivan TJ, Sinclair SJ. 2001.
Determinants of duration of disability and return-to-work after work-
related injury and illness: Challenges for future research. Am J Ind Med
40(4):464–484.
Krause N, Scherzer T, Rugulies R. 2005. Physical workload, work
intensification, and prevalence of pain in low wage workers: Results
from a participatory research project with hotel room cleaners in Las
Vegas. Am J Ind Med 48:326–337.
Lee PT,Krause N. 2002. The impact of a worker health study on working
conditions. J Public Health Policy 23(3):268–285.
Loomis D, Richardson DB, Elliott L. 2005. Poisson regression analysis
of ungrouped data. Occup Environ Med 62:325–329.
McCurdy SA, Samuels SJ, Carroll DJ, Beaumont JJ, Morrin LA. 2003.
Agricultural injury in California migrant Hispanic farm workers. Am
J Ind Med 44(3):225–235.
McGwin G, Jr., Enochs R, Roseman JM. 2000. Increased risk of
agricultural injury among African-American farm workers from Ala-
bama and Mississippi. Am J Epidemiol 152(7):640–650.
Mergler D. 1995. Adjusting for gender differences in occupational
health studies. In: Messing K, Neis B, Dumais L, editors. Invisible:
Issues in women’s occupational health/La sant
e des travailleuses.
Charlottetown, P.E.I., Canada: Gynergy Books. p 236–251.
Messing K. 2004. Physical exposures in work commonly done by
women. Can J Appl Physiol 29(5):639–656.
Messing K, Tissot F, Saurel-Cubizolles M, Kaminski M, Bourgine M.
1998. Sex as a variable can be a surrogate for some working conditions.
J Occup Environ Med 40:250–260.
Messing K, Punnett L, Bond MA, Alexanderson K, Pyle JL, Stock SR,
Wegman DH, Zahm S, Stock SR, de Grosbois S. 2003. Be the fairest of
them all: Challenges and recommendations for the treatment of gender
in occupational health research. Am J Ind Med 43:618–629.
Milburn PD, Barrett RS. 1999. Lumbosacral loads in bedmaking. Appl
Ergon 30(3):263–273.
Murray LR. 2003. Sick and tired of being sick and tired: Scientific
evidence, methods, and research implications for racial and ethnic
disparities in occupational health. Am J Public Health 93:221–226.
National Institute for Occupational Safety and Health. 1996. National
Occupational Research Agenda: Special Populations at Risk. Cincin-
nati, OH: DHHS (NIOSH) Publication No. 96-115.
National Institute for Occupational Safety and Health. 2002. The
changing organization of work and the safety and health of working
people. Cincinnati, OH: DHHS (NIOSH) Publication No. 2002-
116.
Parker E. 1999. Job quality in the hospitality industry: Findings from the
San Francisco housekeeping study. Report to the Rockefeller
Foundation. University of Wisconsin-Madison and University of
California-Berkeley: Madison, Wisconsin, and Berkeley, California.
p 23.
Pransky G, Benjamin K, Hill-Fotouhi C, Fletcher KE, Himmelstein J,
Karz J. 2000. Work-related outcomes in occupational low back pain: A
multidimensional analysis. Am J Ind Med 37(4):400–409.
Punnett L, Herbert R. 2000. Work-related musculoskeletal disorders: Is
there a gender differential, and if so, what does it mean? In: Goldman
MB, Hatch MC, editors. Women and health. San Diego, CA: Academic
Press. p 474–492.
Research Randomizer. Social Psychology Network. Copyright
Ó1997–2008 by Geoffrey C. Urbaniak and Scott Plous. http://
www.randomizer.org/form.htm. Accessed April 10, 2007.
Richardson S, Ruser J, Suarez P. 2003. Hispanic workers in the United
States: An analysis of employment distributions, fatal occupational
injuries, and non-fatal occupational injuries and illnesses in Safety is
Seguridad. National Research Council of the National Academies,
Washington, DC.
Robinson JC. 1989. Exposure to occupational hazards among Hispa-
nics, blacks and non-Hispanic whites in California. Am J Public Health
79(5):629–630.
Satcher D, Higginbotham EJ. 2008. The public health approach to
eliminating disparities in health. Am J Public Health 98(9 Suppl):
S8–S11.
Scherzer T, Rugulies R, Krause N. 2005. Work-related pain, injury, and
barriers to workers’compensation among Las Vegas hotel room clea-
ners. Am J Public Health 95(3):478–488.
Occupational Injury in Hotel Workers 9
Seifert AM, Messing K. 2006. Cleaning up after globalization: An
ergonomic analysis of work activity of hotel cleaners. Antipode
38:557–578.
Simpson CL, Severson RK. 2000. Risk of injury in African American
hospital workers. J Occup Environ Med 42(10):1035–1040.
Sogaard K, Fallentin N, Nielsen J. 1996. Work load during floor
cleaning. The effect of cleaning methods and work technique. Eur
J Appl Physiol Occup Physiol 73(1–2):73–81.
Stellman JM. 1999. Women workers: The social construction of a
special population. Occup Med 14(3):559–580.
Treaster DE, Burr D. 2004. Gender differences in prevalence of
upper extremity musculoskeletal disorders. Ergonomics 47(5):495–
526.
Wial H, Rickert J. 2002. U.S. Hotels and their workers: Room for
improvement, in the state of U.S. Industries. AFL-CIO Working for
America Institute.
Woods V, Buckle P, Haisman M. 1999. Musculoskeletal Health of
Cleaners. Robens Centre for Health Ergonomics, European
Institute for Health and Medical Sciences, University of Surrey, Eng-
land: p 1–12.
10 Buchanan et al.
