Using the OVAKO working posture analysis system in cleaning occupations

Abstract

BACKGROUND:

Cleaning workers experience severe musculoskeletal symptoms.

OBJECTIVE:

The objective of this paper was to examine musculoskeletal symptoms in cleaners of different heights to evaluate the effects of height on working postures in the work environment (schools).

METHODS:

We used a three-stage method including using the Nordic Musculoskeletal Questionnaire (NMQ) to evaluate musculoskeletal symptoms, a task analysis to confirm typical cleaning tasks, and the OVAKO Working Posture Assessment System (OWAS) for posture analysis. Multinomial logistic regression was performed to evaluate the adjusted effects of individual characteristics on painful body regions, using individuals without any pain as the reference category.

RESULTS:

This study found that the prevalence of musculoskeletal symptoms is very high for cleaners, especially in the shoulders, elbows, and lower back. Odds ratios for the accumulation of two or more risk factors were higher among men and were inversely associated with national economic indicators. The relatively high prevalence of musculoskeletal symptoms may stem from the multiple operations involved in cleaning tasks, such as trash collecting, floor mopping, toilet cleaning, and mirror polishing. Workers of different heights had differential work loadings for different tasks.

CONCLUSIONS:

This paper proposes recommendations for job adaptations and occupational safety training. Cleaners of different heights execute the typical tasks via different postures, and awkward postures often result in musculoskeletal symptoms. Cleaners should be provided with specific tools and training regarding working postures on the basis of height. These findings can be used as a reference for related operation designs and task improvements to ensure correct tool usage and safer working postures during cleaning.

Keywords

Cleaning job musculoskeletal symptoms NMQ task analysis working posture

1 Introduction

Cleaning is an important occupation; it is physically demanding and labor-intensive [1] and is undertaken by millions worldwide, resulting in a high cardiovascular load [2, 3]. Musculoskeletal disorders and symptoms are common in working populations [4 –6], predominantly in the lower back [7], neck, and upper limbs [8]. The United States National Institute for Occupational Safety and Health has reported that using awkward postures at work is strongly related to musculoskeletal injuries [9], and this is also true for cleaners [10]. Previous studies have indicated that cleaners are at high risk of developing health problems, and particularly of developing musculoskeletal problems affecting the back, neck, shoulders, elbows, and hands [11, 12]. Moreover, relationships have been demonstrated between poorly designed workplaces, poor working postures, tools, and diseases of the musculoskeletal system [13, 14].

New tools and better work techniques have entered the cleaning industry over the last few decades. Despite this, De Vito et al. [15] reported an increased prevalence of disorders of the elbow, wrist/hand, and cervical spine among cleaners who used good work tools. Further, the cleaning workforce is comprised of a high proportion of elderly women with low social status, who are generally poorly educated and lack social support. In Taiwan, the situation is similar: there are approximately half a million people working as cleaners, and most are aging women with relatively low incomes who work predominantly in public buildings such as schools, hospitals, offices, and retail stores [16, 17].

Standardization is needed when analyzing and recording musculoskeletal symptoms. The Nordic Musculoskeletal Questionnaire (NMQ) was thus developed to standardize the analysis of musculoskeletal symptoms [18, 19].

There are physical, psychosocial, and individual risk factors for work-related musculoskeletal disorders [20 –22]. The majority of prior research suggests that the individual risk factors are: age, gender, and related working experience [23]. Interestingly, females have a significantly higher prevalence than males of developing many types of upper extremity musculoskeletal disorders, even after controlling for the data source and confounders such as age or work factors [24]. However,On the basis of previous studies little is known about how personal anthropometric differences, such as height variability, affect musculoskeletal disorders.

Previous research has suggested that some low-cost improvements can help reduce work-related risks and prevent stress at work based on improvements associated with participatory action-oriented programs in different work settings [25]. However, such improved devices have the same issue— they do not consider individual cleaner anthropometric dimensions such as weight, height, and body mass index [3]. Height is especially important, as it can significantly affect the working postures required for upper and lower cleaning task positions, and can lead to musculoskeletal symptoms[26, 27].

Following from the above, this study aims to evaluate musculoskeletal symptoms across a sample of cleaners of different heights, and thereby understand more about the associated effects of cleaner height on working postures. The results may be helpful in reducing potential musculoskeletal symptoms during cleaning. Based on a three-stage study of musculoskeletal ill health drawing heavily on the participation of the cleaning workforce sample, this paper intends to:

Assess the association between cleaning workers and musculoskeletal symptoms using the NMQ;

Assess working posture differences across the sampled cleaners using the OWAS;

Provide guidelines for preventing musculoskeletal symptoms for cleaning workers of different heights;

Propose guidelines for the design of a cleaning occupational safety education program.

2 Methodology

A three-stage method was used in this study. First, we surveyed participants about their musculoskeletal symptoms using the NMQ. Second, we used task analysis to filter typical cleaning tasks based on the relevant working posture changes. Third, we divided our sample into three height categories and analyzed each using OWAS to identify posture changes during typical cleaning tasks.

2.1 Stage 1: Interview and questionnaire

To obtain reliable information, face-to-face interviews were conducted with all participants by three well-trained interviewers between September and December 2016. In total, 115 cleaners were recruited, and completed the questionnaire twice within one week with informed consents. All volunteer participants were female and the mean age±SD was 57.42±3.94 years (range 53–62). We divided the sample into three height groups (short/average/tall). The NMQ was translated into Chinese according to the guidelines for the process of cross-cultural adaptation of self-report measures [18]. The NMQ includes a drawing of the human body depicting the ten body regions considered while answering the questionnaire. We examined the reliability of the Chinese version among Chinese cleaners. Overall internal reliability (Cronbach’s alpha = 0.89) and test-retest reliability (0.87) were high.

2.2 Stage 2: Task analysis

Task analysis included analyzing videotaped recordings of the cleaners working in a selected space for the purpose of the test. The task started with cleaning a toilet, and ended with trash collection; all tasks were part of the daily cleaning job at a school. The procedures associated with cleaning in the school environment were analyzed and divided into 10 elementary tasks. This study focused on tasks that required body postures changes to complete; tasks that involved mainly table or platform cleaning were disregarded. Of the 10 tasks, target tasks were selected for postural observations that were computed as OWAS scores. Figure 1 shows a typical school-cleaning task.

Fig. 1

Three typical school-cleaning tasks (polish the mirror, mop the floor, and clean the trash).

2.3 Stage 3: OWAS analysis

The OWAS method (Appendix A-1) has been safely and successfully applied in varied sectors and working environments [28, 29]. The OWAS analysis provided information on how cleaners of different heights performed their job. In total, 30 cleaning workers were recruited and divided into three height groups; mean heights were 148 cm (short height group), 153 cm (average height group), and 158 cm (tall height group), with 10 participants in each group.

3 Results

3.1 NMQ questionnaire results

Table 1 depicts the data obtained from the NMQ musculoskeletal symptoms at any of the four body sites for the sample of cleaners over a 12 month period and in the previous week. The results show that the majority of workers experienced pain or discomfort in nine regions of the body, especially during the 12 month period. The most common musculoskeletal symptoms in the 12 month period were in the shoulder (63.9%), elbow (60.9%), lower back (57.9%), and neck (54.7%). In relation to the previous week, the most common symptoms included limited function in a shoulder (55.5%), followed by the lower back (53.9%), elbow (52.4%), wrist/hand (48.4%), and knee (46.0%). Participants who received medical treatment for their injuries most commonly noted trouble with a shoulder (46.8%), elbow (46.0%), lower back (41.2%), or wrist/hand (36.5%). The neck, shoulders, elbows, and lower back were associated with a higher prevalence of musculoskeletal symptoms than other regions of the body. Table 2 shows that the lower back had the highest odds ratio (OR = 9.06, 95% CI = 3.14–26.7) in the short height group, followed by the shoulder (OR = 8.62, 95% CI = 2.88–25.8), and neck (OR = 3.76, 95% CI = 2.15–12.3), after adjusting for some demographic variables. In the tall height group, the elbow had the highest odds ratio (OR = 17.72, 95% CI = 3.84–53.21), followed by the wrist/hand (OR = 11.31, 95% CI = 2.72–40.36). The lower back had the highest odds ratio of the three height groups. The musculoskeletal symptom odds ratios revealed a positive relationship between musculoskeletal symptoms and difference of cleaner height.

Table 1
Musculoskeletal disorders’ percentage of complains in all regions (n = 115)

Questions for answers

Body region of the pain Have you at any time during the last 12 months had trouble (such as aches, pain, discomfort, numbness)? (%) Have you had trouble at any time during the last 7 days? (%) Have you at any time during the last 12 months had medical treatments for the troubles? (%)

Yes No Yes No Yes No

Neck 54.7 45.3 39.6 60.1 15.8 84.2

Shoulder 63.4 36.6 55.5 44.5 46.8 53.2

Elbow 60.9 39.1 52.4 47.6 46.0 54.0

Wrist/hand 53.9 46.1 48.4 51.6 36.5 63.5

Upper back 39.1 60.9 28. 71.5 23.8 76.2

Low back 57.3 42.7 53.9 46.1 41.2 58.8

Hip/thigh 31.3 68.7 19.0 81.0 19.0 81.0

Knee 52.3 47.7 46.0 54.0 28.5 71.5

Ankle/feet 15.6 84.4 15.0 85.0 9.5 90.5

	Questions for answers
Neck	54.7	45.3	39.6	60.1	15.8	84.2
Shoulder	63.4	36.6	55.5	44.5	46.8	53.2
Elbow	60.9	39.1	52.4	47.6	46.0	54.0
Wrist/hand	53.9	46.1	48.4	51.6	36.5	63.5
Upper back	39.1	60.9	28.	71.5	23.8	76.2
Low back	57.3	42.7	53.9	46.1	41.2	58.8
Hip/thigh	31.3	68.7	19.0	81.0	19.0	81.0
Knee	52.3	47.7	46.0	54.0	28.5	71.5
Ankle/feet	15.6	84.4	15.0	85.0	9.5	90.5

Table 2

Odds ratio adjusted for the three height groups for MSDs during last 12 months, according to analysis of multiple logistic regression

Variable^a	n (%)^b	Short height group	Normal height group	Tall height group
		150.3±2.1 cm (n = 39)	155.2±2.8 cm (n = 38)	159.6±1.9 cm (n = 38)
		OR (95% CI)^c	OR (95% CI)	OR (95% CI)
Neck	63(54.7%)	3.76(2.15–12.3) ^*d	3.72(2.01–6.84)	1.03(0.30–3.42)
Shoulder	73(63.4%)	8.62(2.88–25.83) ^***	2.51(1.41–5.30) ^*	3.26(0.61–14.51)
Elbow	70(60.9%)	1.26(0.33–4.85)	1.57(0.60–2.8)	17.72(3.84–53.21)^***
Wrist/hand	62(53.9%)	2.65(0.85–8.31)	1.91(0.83–7.06)	11.31(2.72–40.36)^***
Upper back	45(39.1%)	1.85(0.73–6.85)	1.42(0.48–3.59)	2.46(0.74–8.3)
Low back	66(57.3%)	9.06(3.14–26.7)^***	2.81(1.71–7.2)^*	5.40(1.95–16.61)^***
Hip/thigh	36(31.3%)	3.36(0.92–15.5)	4.35(0.82–16.5)	5.21(0.61–37.51)
Knee	60(52.3%)	3.14(0.88–8.53)	3.08(0.75–8.53)	8.56(1.17–38.61)^***
Ankle/feet	18(15.6%)	1.98(0.78–5.98)	1.59(0.62–4.08)	3.39(0.81–16.5)

^aAll variables are controlled for others of the same level and high levels. The reference category of the outcome is independence. The category of the predictive variables that received Odds Ratios of 1.00 are reference categories. Reference categories: Percentage without MSDs. ^bPercentage with MSDs during last 12 months. ^cOdds Ratios (Confidence interval 95% ). ^d, *means p < 0.05, ^***means p < 0.001.

3.2 Task analysis of school cleaning jobs

Figure 2 shows the percentage of maximum OWAS scores and OWAS categories for 10 elementary tasks involved in a school cleaning job. Figure 3 shows the typical tasks (polishing mirrors, mopping floors, cleaning toilets, and picking up trash) that received a high percent of OWAS code for action categories level AC4. As noted above, all tasks were videotaped at a school-cleaning workspace and later analyzed in the laboratory. The observation time was 30–60 minutes for the ten tasks, during which 60 still videotape frames were sampled at intervals of 10s from the representative contents of the task for later analysis. There was no interference during any of the videotaping.

Fig. 2

Percent of maximum OWAS score and OWAS category for 10 elementary tasks of school cleaning job.

Fig. 3

Typical tasks of school cleaning job selected by higher OWAS level code (trash cleaning, floor mopping, toilet cleaning and mirror polishing).

3.3 OWAS analysis for three height groups

The OWAS analysis results for the three height groups are shown in Table 3. The short height group had the highest percent for the sub-task “bend to pick up trash” (category AC4) at 52%, and the lowest percent for the sub-task “handle trash” (category AC2) at 22%. The average and tall height groups ranged between 29–40% for all sub-tasks. With respect to the floor mopping task, the tall height group reported the highest percent for the sub-task “bend to mop the corner” (category AC4) at 49%, and the lowest percent for the sub-task “clean the floor” (category AC1) at 25%. The short and average height groups ranged between 30–38% for all sub-tasks.

Table 3
OWAS analysis-percentage of posture code in the typical tasks

No Task OWAS code (category) Percent of frames for three height group (%) All ANOVA p-value

Short (n = 10) Average (n = 10) Tall (n = 10)

Trash cleaning

1 Bend to clean trash 4234 (4) 52 41 34 42.3 14.51^ <0.01

2 Lift the trash 2244 (3) 26 31 34 30.3 8.67 0.21

3 Move trash 2122 (2) 23 29 30 27.3 7.56 0.18

Floor mopping

1 Floor clean 1121 (1) 37 30 25 30.6 7.25 0.18

2 Mop the corner 2131 (3) 33 32 26 30.3 8.34 0.27

3 Bend to mop the corner 2151 (4) 30 38 49 39.0 13.68^ <0.01

Toilet cleaning

1 Clean toilet(top) 2131 (2) 36 33 31 33.3 3.24 0.17

2 Bent to clean toilet(bottom) 2161 (4) 30 31 40 33.6 11.85^* 0.03

3 Brush toilet 2141 (3) 34 36 29 33.0 6.67 0.19

Mirror polishing

1 Clean the mirror(top) 1351 (4) 49 36 28 37.6 16.81^ <0.01

2 Clean the mirror(middle) 1121 (1) 21 26 22 23.0 2.81 0.14

3 Clean the mirror(bottom) 4151 (4) 30 38 50 39.3 18.41^ <0.01

No	Task	OWAS code (category)	Percent of frames for three height group (%)	All	ANOVA	p-value
Trash cleaning
1	Bend to clean trash	4234 (4)	52	41	34	42.3	14.51^**	<0.01
2	Lift the trash	2244 (3)	26	31	34	30.3	8.67	0.21
3	Move trash	2122 (2)	23	29	30	27.3	7.56	0.18
Floor mopping
1	Floor clean	1121 (1)	37	30	25	30.6	7.25	0.18
2	Mop the corner	2131 (3)	33	32	26	30.3	8.34	0.27
3	Bend to mop the corner	2151 (4)	30	38	49	39.0	13.68^**	<0.01
Toilet cleaning
1	Clean toilet(top)	2131 (2)	36	33	31	33.3	3.24	0.17
2	Bent to clean toilet(bottom)	2161 (4)	30	31	40	33.6	11.85^*	0.03
3	Brush toilet	2141 (3)	34	36	29	33.0	6.67	0.19
Mirror polishing
1	Clean the mirror(top)	1351 (4)	49	36	28	37.6	16.81^**	<0.01
2	Clean the mirror(middle)	1121 (1)	21	26	22	23.0	2.81	0.14
3	Clean the mirror(bottom)	4151 (4)	30	38	50	39.3	18.41^**	<0.01

Notes: ^*means the p-value is significant when <0.05. ^**means the p-value is significant when <0.01.

For toilet cleaning, the tall group had the highest percent for the sub-task “bend to clean the toilet (bottom)” (category AC4) at 40%, and the lowest for the sub-task “brush the toilet” (category AC3) at 29%. The short and average height groups ranged between 30–36% for all sub-tasks. For the last task, mirror polishing, the short height group had the highest percent for the sub-task “clean the mirror (top)” (category AC4) at 49%, and the lowest percent for the sub-task “clean the mirror (middle)” (category AC1) at 21%. In contrast, the tall group had the highest percent for the sub-task “clean the mirror (bottom)” (category AC4) at 50%, and the lowest percent for the sub-task “clean the mirror (middle)” (category AC1) at 22%. Finally, the average height group ranged between 26–38% for all sub-tasks.

The results of the one-way ANOVA for the frame percent in the OWAS analysis by the three height groups is depicted in Table 3. The findings show that there were significant differences in the OWAS analysis across the three height groups in terms of bending to clean trash, bending to mop the corner, cleaning the mirror (top), cleaning the mirror (bottom; P < 0.001), and bending to clean the toilet (down; P < 0.005).

4 Discussion

The findings of this study show that 63.9% of cleaners in the last year and 55.5% in the previous week reported at least one musculoskeletal symptom due to their job requirements. From the literature review, musculoskeletal symptoms are the most common work-related injury, comprising 40.4% to 82.3% of the total number of work-related symptoms in different parts of the body [30]. Awkward, extreme, or repetitive working postures have been identified as the main risk factors for musculoskeletal symptoms across various industries [9, 31].

The prevalence of musculoskeletal symptoms in our sample was higher than in the above studies, and we also identified the different body parts associated with the musculoskeletal symptoms. The shoulder, elbow, and lower back were the regions that were most commonly cited for musculoskeletal symptoms in both the previous week and year. Musculoskeletal symptoms in the “lower back” may be caused by bending tasks, such as bending to pick up trash, mopping the corner, and cleaning the mirror (bottom). Lower back pain (LBP) has been found to be associated with forceful movements [32], as well as working in a bent forward position [33]. Thus, evaluation of lower lumbar segmental mobility should be considered in routine clinical assessments, as this type of evaluation provides information on disability over time [6].

Professional cleaning is one of the most common occupations worldwide. Substandard work places and worker inattention to caution instructions, as well as a lack of national and effective preventive strategies or programs may be responsible for the high rate of issues [1, 2]. The majority of cleaners in Taiwan are older women [34], and many are poorly educated with little social support [15]. These factors are likely to be associated with the high rate of musculoskeletal symptoms in cleaners. As such, educational instruction programs should be redesigned to better fit the needs of cleaners. The results of this study can be used to help establish safety guidelines for activities requiring bending, pushing, pulling, and carrying, such as load capacity limits to protect the lower back against occupational disorders [35].

Several studies on the cleaning occupation describe the typical physical demands of this type of work [1 , 36–38]. Prolonged static and repetitive muscle activity causes muscle fatigue and may lead to musculoskeletal disorders [2]. Other researchers have found that the tools and equipment used in cleaning require users to engage in both dynamic and static muscular activity [37]. However, height differences across cleaners may create differences to the static and dynamic changes involved in working postures.

In related research, a comparison of pushing and pulling forces measured using a high inertia cart with those measured on a treadmill shows that pushing and pulling forces that use a high inertia cart are greater for males, while they are about the same for females. At present, it is not clear whether pushing or pulling should be favored. Similarly, it is not clear what handle heights are optimal for pushing and pulling [39].

One possible explanation is that most of the existing research has investigated the workload of office cleaners, rather than hotel-room cleaners whose workload is characterized by more compulsory work positions and movements [34, 40]. As such, analysis should be conducted on the impact of physical load in combination with other work environment risks (status, size, and task space) that can worsen the influence of the physical workload on school cleaners.

5 Conclusions

From this study, it can be concluded that height may affect how cleaning tasks are performed and create different work loadings. The results of this three-stage study indicate that the safe execution of cleaning tasks depends not only on educational instructions on equipment usage, but also on the work environment (e.g., toilet, trash, mirror, or floor), task analysis (e.g., trash cleaning, floor mopping, toilet cleaning, or mirror polishing), and personal anthropometric differences (e.g., height). Cleaners of different heights must use different postures to complete the required tasks, and inappropriate postures will result in problems, especially musculoskeletal symptoms. Therefore, height may be a critical factor affecting muscular symptoms in the cleaning industry.

Cleaners should be educated on correct working postures and provided with specific tools according to their height and may fulfill specific tasks according to the task demands. For the employer, educational instructions should warn about the propensity for musculoskeletal symptoms relative to height of cleaners. The present results can be used as a reference for related operations design and improvements to cleaning jobs.

6 Limitations and future research

The present study has some limitations: first, due to the self-reporting nature of the NMQ, the work experience and educational level of respondents may have affected questionnaire completion. Second, the data cannot provide specific evidence regarding the changes that need to be made to overcome postural stresses at work.

We conclude that musculoskeletal symptoms in cleaning jobs occur at a high rate, and recommend that additional studies investigate means to accurately assess musculoskeletal symptom risk factors. Programs must focus on reducing physical exposure to musculoskeletal symptom risk factors of these body regions.

Conflict of interest

None to report.

Footnotes

Appendix

A-1 Definition of posture codes and action categories in Ovako Working Posture Analysis System (OWAS) (Karhu et al., 1977)

Posture			OWAS action category &action requires
Trunk	Arm	Leg	Force (kg)
1 = straight/upright	1 = both arms below shoulder height	1 = sitting	1=<10	AC1: No action required
2 = bent forward	2 = one arms above shoulder height	2 = standing on both legs straight	2 = 10–20	AC2: Action required in the near future
3 = straight &twisted	3 = both arms above shoulder height	3 = standing on one legs straight	3=>20	AC3: Action required as soon as possible
	4 = bent &twisted	4 = standing on both legs bent		AC4: Action required immediately
		5 = standing on one bent leg
		6 = kneeling on one or both legs
		7 = walking

Acknowledgment

We would like to thank Uni-edit () for editing and proofreading this manuscript.

References

Hagner

, Hagberg

. Evaluation of two floor-mopping work methods by measurement of load. Ergon. 1989;32(4):401–8.

Kumar

, Chaikumarn

, Kumar

. Physiological, subjective and postural loads in passenger train wagon cleaning using a conventional and redesigned cleaning tool. Int J Ind Ergon. 2005;35(10):931–8.

Kotowski

, Davis

. Influence of weight loss on musculoskeletal pain: Potential short-term relevance. Work. 2010;36(3):295–304.

Davis

, Sobeih

, Succop

, Jetter

, Kotowski

, Bhattacharya

. Impact of obesity on the postural balance of firefighters. Occup Ergon. 2009;8(2, 3):115–23.

Ojukwu

, Anyanwu

, Nwabueze

, Anekwu

, Chukwu

. Prevalence and associated factors of work related musculoskeletal disorders among commercial milling machine operators in South-Eastern Nigerian markets. Work. 2017;58(4):473–80.

Lundberg

, Gerdle

. Musculoskeletal signs in female homecare personnel: A longitudinal epidemiological study. Work. 2017;58(2):135–47.

Troup

JDG

, Edwards

. Manual handling:Areviewpaper. Health and Safety Executive. London. Her Majesty’s Stationary Office. 1985.

Oxenburgh

, Rowe

, Douglas

. Repetition strain injury in keyboard operators. J Occup Health and Safety. 1985;1(12):106–12.

Bernard

, Putz-Anderson

. Musculoskeletal disorders and workplace factors; a critical review of epidemiologic evidence for work-related musculoskeletal disorders of the neck, upper extremity, and low back. 1997.

10.

Veiersted

. Sustained muscle tension as a risk factor for trapezius myalgia. Int J Ind Ergon. 1994;14(4):333–9.

11.

Huth

, Kruger

, Zorzi

. Gesundheitsforderung im krankenhausbetrieb, funktionsbereich reinigung. Abschlubericht, Fachhochschule Hamburg. 1996;5.

12.

Kumar

, Kumar

. Musculoskeletal risk factors in cleaning occupation—A literature review. Int J Ind Ergon. 2008;38(2):158–70.

13.

Louhevaara

, Hakola

, Ollila

. Physical work and strain involved in manual sorting of postal parcels. Ergon. 1990;33(9):1115–30.

14.

Moriguchi

, Carnaz

, Veiersted

, Hanvold

, Hæg

, Hansson

G-Å

, et al. Occupational posture exposure among construction electricians. Appl Ergon. 2013;44(1):86–92.

15.

De Vito

, Molteni

, Camerino

, Bordini

, Molinari

, Capodaglio

. Aging and work: Health aspects in cleaners. Med Lav. 2000;91(4):387–402.

16.

IOSH (Institute of Occupational Safety and Health). 2007. Safety and health issues in female hospital cleaners in Chinese. Taipei, Council of Labor Affaris, Executive Yuan, Taiwan.Available at. http://www.iosh.gov.tw/Publish.aspx?cnid1/416&P1/4172 access on Sep. 20, 2016.

17.

CLA (Council of Labor Affairs). 2010. Labor Statistics Survey: Survey on Earning by Occupation in Chinese. Taiwan: Taipei. Available at. http://statdb.cla.gov.tw/html/svy99/995016.xls. access on Sep. 20, 2016.

18.

Kuorinka

, Jonsson

, Kilbom

, Vinterberg

, Biering-Sørensen

, Andersson

, et al. Standardised Nordic questionnaires for the analysis of musculoskeletal symptoms. Appl Ergon. 1987;18(3):233–7.

19.

Dickinson

, Campion

, Foster

, Newman

, O’rourke

, Thomas

. Questionnaire development: An examination of the Nordic Musculoskeletal questionnaire. Appl Ergon. 1992;23(3):197–201.

20.

Buckle

, Devereux

. The nature of work-related neck and upper limb musculoskeletal disorders. Appl Ergon. 2002;33(3):207–17.

21.

Waldenström

, Theorell

, Ahlberg

, Josephson

, Nise

, Waldenström

, et al. Assessment of psychological and social current working conditions in epidemiological studies: Experiences from the MUSIC-Norrtälje study. Scand J Public Health. 2002;30(2):94–102.

22.

Bongers

, Kremer

, Laak

. Are psychosocial factors, risk factors for symptoms and signs of the shoulder, elbow, or hand/wrist?: A review of the epidemiological literature. Am J Ind Med. 2002;41(5):315–42.

23.

Garden

, Moore

, Jorm

. The current and future health status of the New South Wales population. New South Wales Department of Health, Sydney. 2005.

24.

Treaster

, Burr

. Gender differences in prevalence of upper extremity musculoskeletal disorders. Ergon. 2004;47(5):495–526.

25.

Kogi

, Yoshikawa

, Kawakami

, Lee

, Yoshikawa

. Low-cost improvements for reducing multifaceted work-related risks and preventing stress at work. Ergon. 2016;6(147):2.

26.

Satyanarayana

, Naidu

, Chatterjee

, Rao

. Body size and work output. Am J Clin Nutr. 1977;30(3):322–5.

27.

Biewener

. Bone strength in small mammals and bipedal birds: Do safety factors change with body size? J Exp Biol. 1982;98(1):289–301.

28.

Freund

, Toivonen

, Takala

. Grip forces of the fingertips. Clin Biomech. 2002;17(7):515–20.

29.

Santos

, Sarriegi

, Serrano

, Torres

. Using ergonomic software in non-repetitive manufacturing processes: A case study. Int J Ind Ergon. 2007;37(3):267–75.

30.

Wong

, Grundfast

, Levi

. Assessing work-related musculoskeletal symptoms among otolaryngology residents. Am J Otolaryngol. 2017;38(2):213–7.

31.

Silverstein

, Fine

, Armstrong

. Occupational factors and carpal tunnel syndrome. Am J Ind Med. 1987;11(3):343–58.

32.

Punnett

, Fine

, Keyserling

, Herrin

, Chaffin

. Back disorders and nonneutral trunk postures of automobile assembly workers. Scand J Work Environ Health. 1991;337–46.

33.

Vingård

, Alfredsson

, Hagberg

, Kilbom

, Theorell

, Waldenström

, et al. To what extent do current and past physical and psychosocial occupational factors explain care-seeking for low back pain in a working population? Results from the musculoskeletal intervention center-norrtälje study. Spine. 2000;25(4):493–500.

34.

Hsieh

Y-CJ

, Apostolopoulos

, Sönmez

. The world at work: Hotel cleaners. Occup Environ Med. 2013;70(5):360–4.

35.

Shoaf

, Genaidy

, Karwowski

, Waters

, Christensen

. Comprehensive manual handling limits for lowering, pushing, pulling and carrying activities. Ergon. 1997;40(11):1183–200.

36.

Hopsu

, Leppänen

, Ranta

, Louhevaara

, editors. Perceived work ability and individual characteristics as predictors for early exit from working life in professional cleaners. International congress series; 2005.

37.

Hopsu

, Louhevaara

, Korhonen

, Miettinen

, Huikko

, Lehtonen

, et al. Feasibility and effects of the intervention for developing work organisation on stress and strain in professional cleaning. Rakennushallitus, Report. 1994;3:1994.

38.

Søgaard

, Fallentin

, Nielsen

. Work load during floor cleaning. The effect of cleaning methods and work technique. Eur J Appl Physiol. 1996;73(1-2):73–81.

39.

Garg

, Waters

, Kapellusch

, Karwowski

. Psychophysical basis for maximum pushing and pulling forces: A review and recommendations. Int J Ind Ergon. 2014;44(2):281–91.

40.

Stephen

. Anthropometry, ergonomics and the design of work. CRC Press, Guildfor. 2016.

	Questions for answers
Body region of the pain	Have you at any time during the last 12 months had trouble (such as aches, pain, discomfort, numbness)? (%)		Have you had trouble at any time during the last 7 days? (%)		Have you at any time during the last 12 months had medical treatments for the troubles? (%)
	Yes	No	Yes	No	Yes	No
Neck	54.7	45.3	39.6	60.1	15.8	84.2
Shoulder	63.4	36.6	55.5	44.5	46.8	53.2
Elbow	60.9	39.1	52.4	47.6	46.0	54.0
Wrist/hand	53.9	46.1	48.4	51.6	36.5	63.5
Upper back	39.1	60.9	28.	71.5	23.8	76.2
Low back	57.3	42.7	53.9	46.1	41.2	58.8
Hip/thigh	31.3	68.7	19.0	81.0	19.0	81.0
Knee	52.3	47.7	46.0	54.0	28.5	71.5
Ankle/feet	15.6	84.4	15.0	85.0	9.5	90.5