1 INCIDENCE RATE AND RISK FACTORS ASSOCIATED WITH LOW BACK PAIN

Download Keywords: incidence, low back pain, physical fitness, office workers, risk factor ... mass index (BMI) based on the obesity measure for Tha...

0 downloads 419 Views 211KB Size
Proceedings 19th Triennial Congress of the IEA, Melbourne 9-14 August 2015

Incidence rate and risk factors associated with low back pain among university office workers in Thailand a,*

b

Sunisa Chaiklieng , Pornnapa Suggaravetsiri , Jenny Stewart

c

a

Department of Environmental Health Science, Faculty of Public Health, Khon Kaen University, Khon Kaen, b Thailand. Department of Epidemiology, Faulty of Public Health, Khon Kaen University, Khon Kaen, c Thailand. Department of Rehabilitation, Faculty of Heath and Environmental Science, AUT, New Zealand. Persistent low back pain (LBP) is common in society and is clearly an occupation-related problem. The nature of office work and the office environment have been associated with increased risks of LBP. This 12-month prospective cohort study investigated the incidence rate of low back pain (LBP) and the risk factors associated with LBP among university-based office workers. The participants were 159 office workers in Khon Kaen University. Thailand met the inclusion criteria, Data were collected by use of a structured interview questionnaire and by measurements of physical fitness, workstation dimensions and lighting intensity. The LBP incidence was estimated from a 12-month follow-up. The data were analysed using descriptive statistics, t-tests and multiple logistic regression. The results showed that the incidence of LBP during the 12-month follow-up was 83.0%. Physical fitness tests showed that the back, leg and hand grip strengths were significantly lower in LBP cases than in non-cases. The multiple logistic regression analysis indicated that the risk factors associated 2 with LBP were BMI ≤25.0 kg/m (ORadj= 3.49, 95%CI: 1.27-9.55), poor back pain preventive behaviour (ORadj = 3.44, 95%CI: 1.08-10.98) and inappropriate workstation width (ORadj = 5.72, 95%CI: 1.44-22.70). A family member with LBP and worker with high stress had significantly increased risk of LBP from univariate analysis. Most workstations (91.8%) had a lighting intensity lower than the standard requirement (600 lux). Practitioner Summary: The results provide an indication of the nature of hazards affecting LBP in office workers in the academic workplace. In order to prevent LBP in this group of office workers, it is proposed that ergonomics education and the better design of workstations, improved lighting, and a greater emphasis on the physical fitness of staff are needed. Keywords: incidence, low back pain, physical fitness, office workers, risk factor

1. Introduction Persistent musculoskeletal disorders (MSDs) are common in society. Most people (70-85%) have low back pain (LBP) at some time in their life [1]. In general, the annual prevalence of LBP ranges from 0.8% to 82.5% [2]. But prevalence of neck or shoulder pain in office workers is reported to be much higher than in the general population. In a study of university academics in Hong Kong, the one-year prevalence of neck pain was found to be 46.7% [3]. A study of medical secretaries in Sweden reported that 63%, 62% and 51% had experienced neck, shoulder and lower back pain, respectively, over the past year [4]. A survey of school personnel in Nagoya, Japan, found that the prevalence of LBP among female office staff was 21.4% [5]. In Thailand, the six-month prevalence of repetitive strain injury among school teachers was found to be as high as 73.7% [6]. The various possible risk factors for MSDs in office workers have included individual factors (age and gender) [3], work environment factors (workstations, lifting, repetitive work), physical factors (prolonged sitting, awkward posture) [7,8], and psychosocial factors (stress, negative emotional states) [9]. A high prevalence of MSDs (69.0%) and LBP (66.9%) has previously been reported in a group of Khon Kaen University office workers [10], but it is still unclear how far LBP was caused by the physical nature of office work in the academic workplace or by other factors related to LBP. This study therefore aimed to update the incidence rate of LBP and to provide a clearer identification of the factors associated with LBP among university office workers.

 

1  

Proceedings 19th Triennial Congress of the IEA, Melbourne 9-14 August 2015

2. Materials and methods 2.1

Recruitment of participants

This study was designed as 12-month follow-up study among office workers in Khon Kaen University of Thailand. The sample size was calculated on the basis of a one-year follow-up study to estimate LBP incidence among Khon Kaen University office workers population. An estimation of sample size under the null hypothesis with one-sided test of incidence rate was used to calculate this [11]. The six-month incidence rate of LBP among university office workers was reported to be 35% [12]. To test the value of the LBP incidence rate (λ0= 0.35) at the 5% level of significance with a power of 90%, the anticipated incidence rate (incidence rate >35%) was 45% (λa=0.45). The minimum sample size requirement was therefore 133 office workers. To allow for the withdrawal and dropout of participants during the follow-up, the required sample size was increased by 20% to a total of 158 office workers for the cohort group. All units in Khon Kaen University that provided name lists of office workers were included in computer program. After this process of simple random sampling for the screening group, participants were first invited and then interviewed before acceptance to this study. They were considered eligible for inclusion in the study if they were a full-time university employee, worked with computer for at least 4 hours per day, had at least one-year work experience at the current position, and were willing to participate. Participants were excluded if they had current low back pain at severe level (established by interview), a history of an episode of care of low back pain in the past three months or any specific medical condition affecting thoracic or lumbar spine (such as rheumatoid arthritis, degenerative disc disease, infection, tumors), a prolonged absence from work anticipated within the next 12 months or were pregnant. The final sample size for LBP follow-up was 159 office workers. In this study, low back pain was defined as pain experienced in the spine area specifically from the lumbar vertebrae to the buttocks or gluteal folds, or between the lumbosacral vertebral prominences [13]. This study obtained ethical approval from Khon Kaen University ethics committee, Thailand, No.HE522091. All participants gave informed consent prior to entering the study. 2.2

Data collection

  Baseline measures and screening data were collected by face-to-face interviews with structured questions based on Chaiklieng et al. [10]’s questionnaire, which was divided into four parts. Part 1 enquired about demographic characteristics i.e. age, gender, work experience, second job or part time job. Part 2 enquired about health status, history of trauma, congenital diseases and chronic diseases (opened question), body 2 mass index (BMI) based on the obesity measure for Thai people (BMI >25 kg/m ) [14]. Part 3 was concerned with the assessment of psychosocial stress. The instrument for this was based on the standardised test developed by the Thai Department of Mental Health [15] and included 20 questions. Total scores on these questions were used to classify subjects by the level of stress experienced (high to normal). Part 4 enquired about work characteristics and behaviour including, for example, daily duration of sitting at work, frequency of breaks from sitting at work, back pain preventive behaviour, and the suitability of work stations. Preventive behaviour was assessed by the total score in response to 20 questions regarding, for example, postures, movements, lifting, and exercise, and these total scores were classified as indicative of poor, fair or good levels of preventive behaviour. The suitability of workstations was assessed in terms of their dimensions (the width, height and depth of the seat and table, and the width and breadth of the workspace), and lighting intensity. Each workstation characteristic was classified as ‘appropriate’ or ‘inappropriate’. Part 5 enquired about MSDs (including LBP) experienced in the three months prior to the date of data collection [16]. The severity of MSDs and low back pain were classified as mild, moderate, or severe with good reliability test before using by Cronbach’s alpha at 0.92. The International Committee for the Standardization of Physical Fitness Tests (ICSPFT) was used as the basis for assessing the physical fitness of the subjects. Grip dynamometry was used to measure handgrip strength, a back-leg dynamometer was used to measure leg and back strength, and a back flexion-extension dynamometer was used to measure back-muscle flexibility. The lighting intensity at computer workstations was measured with a LUTRON LX-105 light meter, and the readings were compared with the Thai standard for computer work in an office workplace [17].

 

2  

Proceedings 19th Triennial Congress of the IEA, Melbourne 9-14 August 2015

Subjects of cohort group were followed-up for 12 months, during which time a modification of the follow-up procedure of Hush et al. [18] was adopted: every two weeks each subject was contacted and asked the primary question, "Have you experienced any low back pain lasting more than 24 hours during the past fortnight (Y/N)?" If the subject reported any onset of low back pain, additional information was requested regarding date of onset, treatment sought, whether the pain affected their work performance and/or daily activities, and whether it resulted in any loss of work. A positive response to any of the following questions confirmed that the episode of LBP was work-related: 1) Did your low back pain start at work? 2) Did your low back pain result from an injury or event at work? 3) Have you submitted a worker's compensation claim for this low back pain? 4) Did your health care provider specify that this is a workrelated injury? When a subject reporting an episode of LBP, he/she continued to be followed up for a further period of one month to establish the severity of the LBP: it was deemed to be more severe if a health professional had been consulted and more severe still if the pain resulted in a period of work absence. If the LBP was not work-related, the subject was withdrawn from the study, and, if appropriate, the subject was advised to consult a health care practitioner was provided. The incidence rate of LBP was calculated for the 3-, 6-, 9-, and 12-month follow-up periods. An accumulated incidence rate was evaluated at 12 months.

    2.3

Data analysis Data were recorded by Epi-info for Windows (Texas, USA, 2007) using a method of double data entry and the analysis was performed using STATA version 10.1. Descriptive statistics were used to describe the worker and workstation characteristics i.e. the percentage, mean and standard deviation (SD). The LBP incidence was estimated for every three month of the 12-month follow-up. The accumulated incidence rate of LBP = (Total number of new cases of LBP at 3- or 6- or 9- or 12-month period x 100) /159. No subjects needed to be withdrawn during follow-up. The associations between LBP and specified study factors were analysed by t-tests, and univariate and multivariate logistic regression. Factors with a pvalue < 0.20 in the univariate analysis and potential confounding factors (age, gender, work experience) were included in the multiple logistic regression analysis. Statistical significance was set p-value<0.05. 3. 3.1

Results Personal factors and health status

Most (76.7%) of the 159 university office workers were female, and the mean age was 33.8 ± 9.9 years (min= 23, max=59). The mean length of work experience was 12.6 ± 10.1 years (min = 1, max = 39). Most (37.7%) had 1-5 years work experience, 17.0% had 16-20 years, and 14.5% had 21-25 years work experience. Regarding educational background, 79.9% had a bachelor’s degree, and 20.1% had completed a master’s degree. According to the Thai standards for assessing nutritional status [16], 20.1% 2 2 of the workers were classified as obese (BMI>25 kg/m ), 15.7% were overweight (23.0-25.0 kg/m ), and 2 54.7% were in the normal range (18.5-23.0 kg/m ). In terms of psychosocial stress, 62.3% of workers were classified as experiencing a normal level of stress, and 21.4% were in the high stress category. Most workers (79.2%) did not take regular exercise (at least 30 minutes, and 3 times a week). Back pain preventive behaviour was assessed as ‘fair’ in 61.3%, ‘poor’ in 21.4%, and ‘good’ in 11.3%. Back pain experienced by a family member was reported by 40.3%. 3.2

Work environmental factors

A large majority (91.8%) of the subjects’ workstations were found to have light intensities lower than the minimum standard requirement (600 lux) (min= 110 lux, max=823 lux). While 95% of their office chairs had a backrest, and 43.4% were up-down flexible and 75.5% were revolvable, the height and width of the seats were ‘inappropriate’ in 66.7% and 72.3% of cases, respectively. Furthermore, when their table dimensions were measured, 66.0% and 63.5% were assessed as ‘inappropriate’ for height and width, respectively. Workspace width and depth were inappropriate in 66.7% and 75.5% of workstations, respectively. 3.3

 

Incidence of low back pain

3  

Proceedings 19th Triennial Congress of the IEA, Melbourne 9-14 August 2015

Data from the fortnightly follow-up interviews (Table 1) show that the incidence rate of LBP at 3-months was 52.8% (84 cases). The accumulated incidence of LBP at the end of the 12-month follow-up was 83.0% (n=132 cases). All cases were work-related LBP. The main reported cause of LBP was prolonged work at a computer in an awkward posture. There were 62 cases of LBP (46.9%) involving symptoms which the subjects reported to have affected the quality of their work performance and normal life. In eight cases (6.1%) the subjects reported receiving treatment from a health professional and taking medication. Three cases (2.3%) resulted in the subjects taking sick leave because of their symptoms. When the 12-month follow-up period was divided into four three month sub-periods, there appeared a trend towards an increased rate of LBP incidence following the first sub-period. Table 1. Incidence rate of LBP among university office workers for different periods of the follow-up (N=159).

Follow-up period

Accumulated incidence rates (%)

Accumulated numbers Cases

Non- cases

New cases of LBP at in each subperiod n (%)

84

75

52.8

84 (52.8)

6 months (2 sub-period)

nd

103

56

64.8

19 (25.3)

rd

119

40

74.8

16 (28.5)

132

27

83.0

13 (32.5)

st

3 months (1 sub-period)

9 months (3 sub-period) th

12 months (4 sub-period)

The strength of back, leg and hand grip were all significantly lower in LBP cases than in the nonLBP cases (see Table 2). Table 2. Baseline physical fitness of LBP cases compared with non-cases LBP (N=159).

Low back pain

Variables (mean±SD)

p-value

Cases (n=132)

Non-cases (n=27)

Back strength (kg)

47.4±2.3

63.0±5.5

0.003*

Leg strength (kg)

88.2±3.8

116.0±10.8

0.002*

Hand grip strength (kg)

29.7±0.8

34.7±2.3

0.006*

Back muscle flexibility (cm)

5.5±1.2

4.2±1.5

0.311

*significant by t-test at p-value< 0.05 The factors significantly associated with LBP in the univariate analysis (see Table 3) were poor 2 back pain preventive behaviour, family members with back pain, high stress, a BMI≤ 25 kg/m , and use of a 2 workstation with inappropriate width. Worker who had BMI ≤ 25 kg/m had 2.94 times higher risk to the LBP 2 development compared to BMI >25 kg/m (95%CI=1.18-7.27). Also, if there was back pain in members of the family, workers had 3.55 times higher risk to the development of LBP compared to no back pain in members of the family (95%CI=1.27-9.96). Workers who had the high stress had 3.01 times higher risk to the development of LBP compared to no stress (95%CI=1.01-9.43). Workers who had poor back pain prevention behaviour had 3.85 times higher risk to the development of LBP compared to those of good

 

4  

Proceedings 19th Triennial Congress of the IEA, Melbourne 9-14 August 2015

behaviour (95%CI=1.33-11.10). Workers with inappropriate workstation width had the higher risk to LBP development compared to the appropriate workstation width (95%CI=1.32-7.21). Table 3. Univariate analysis of potential risk factors for LBP among University office workers (N = 159).

LBP cases

Non-cases

OR

95%CI

n (%)

n (%)

27 (73.0) 105 (86.1)

10 (27.0) 17 (13.9)

2.29

0.94-5.56

0.068

Age (years) > 40 ≤ 40

59 (78.7) 73 (86.9)

16 (21.3) 11 (13.1)

1.79

0.78-4.17

0.171

Work experience (years) > 15 ≤ 15

53 (77.9) 79 (86.8)

15 (22.1) 12 (13.2)

1.86

0.81-4.29

0.140

Chronic disease Yes No

59 (89.4) 73 (78.5)

7 (10.6) 20 (20.5)

2.30

0.90-5.83

0.077

110 (86.6) 22 (68.8)

17 (13.4) 10 (31.2)

2.94

1.18-7.27

0.020*

59 (92.2) 73 (76.8)

5 (7.8) 22 (23.2)

3.55

1.27-9.96

0.016*

High stress Yes No

46 (92.0) 86 (78.9)

4 (8.0) 23 (21.1)

3.10

1.01-9.43

0.049*

Back pain preventive behavior Poor Good

121(85.8) 11(61.1)

20(14.2) 7(38.9)

3.85

Workstation width Inappropriate Appropriate

94(88.7) 38(71.1)

12(11.3) 15(28.3)

3.09

1.32-7.21

Workstation depth Inappropriate Appropriate

103 (85.8) 29 (74.4)

17 (14.2) 10 (25.6)

2.09

0.86-5.05

Factors

Gender Male Female

p-value

2

BMI (kg/m ) ≤ 25 >25 Member of family had back pain Yes No

1.33-11.10

0.013*

0.009*

0.102

* significant at p-value< 0.05

 

5  

Proceedings 19th Triennial Congress of the IEA, Melbourne 9-14 August 2015

2

The multivariate analysis the significant risk factors for LBP were BMI ≤ 25.0 kg/m , poor back pain prevention behaviour and poor workstation design (inappropriate workstation width) (Table 4).

 

6  

Proceedings 19th Triennial Congress of the IEA, Melbourne 9-14 August 2015

Table 4. Multivariate analysis of potential risk factors for LBP among University office workers (LBP cases = 132).

Factors

LBP case (%)

OR

ORadj (95%CI)

27 (73.0) 105 (86.1)

2.29

2.00 (0.73-5.55)

0.179

Age (years) > 40 ≤ 40

59 (78.7) 73 (86.9)

1.79

0.99 (0.22-4.45)

0.986

Work experience (years) > 15 ≤ 15

53 (77.9) 79 (86.8)

1.86

1.32 (0.32-5.46)

0.701

Chronic disease Yes No

59 (89.4) 73 (78.5)

2.30

2.19 (0.74-6.55)

0.157

110 (86.6) 22 (68.8)

2.94

3.49 (1.27-9.55)

0.015*

2.74 (0.84-8.97)

0.096

Gender Male Female

p-value

2

BMI (kg/m ) ≤ 25 >25 Member of family had back pain No Yes

59 (92.2) 73 (76.8)

High stress Yes No

46 (92.0) 86 (78.9)

3.10

2.54 (0.68-9.50)

0.165

121 (85.8) 11 (61.1)

3.85

3.44 (1.08-10.98)

0.037*

Workstation width Inappropriate Appropriate

94 (88.7) 38 (71.1)

3.09

5.72 (1.44-22.70)

0.013*

Workstation depth Inappropriate Appropriate

103 (85.8) 29 (74.4)

2.09

0.43 (0.10-1.83)

0.257

Back pain prevention behaviour Poor Good

3.55

* significant at p-value<0.05

4.

Discussions

The finding of a high accumulated incidence of LBP in Thai university office staff over a 12-month follow-up period (83%) supports the high incidence found in the authors’ previous study [9]. The use of a prospective

 

7  

Proceedings 19th Triennial Congress of the IEA, Melbourne 9-14 August 2015

cohort with frequent regular follow-up contacts with the subjects over the entire follow-up period in the present study minimised the risk of recall bias and adds further strength to this finding of a high incidence of LBP in this occupational group. Cases of MSD have been reported among academic staff in the University of Hong Kong [3], university office workers in Malaysia [12], and school personnel in Japan [5]. However, the six-month incidence rate found in this present study (64.8%) was higher than the incidence reported for lower back disorders in the Malaysian study control group (38.2%). The higher incidence in the present study may be explained by the difference in eligibility criteria. In the Malaysian study [12], subjects were selected if they worked on computer for at least three hours per day. In our study, the criterion was at least four hours per day, and this difference may have been crucial in the determination of risk for LBP. Also confirmed were the major causes of LBP among university office workers, prolonged sitting in the same posture and awkward posture [10]. These factors might explain why office workers have a higher risk developing LBP than people in general [1, 3-5]. In some cases during follow-up, subjects reported taking sick leave due to the occurrence of LBP. Chronic LBP can result in long term treatment costs [19]. It can also be a cause of an increased sick leave among Thai employees [20], and this can have indirect additional negative impacts on academic activities and management. The nature of office work with computers involves sitting in a fixed position in front of screen or typing documents and performing numerical tasks for at least four hours a day. In the present study, as in other studies [8, 10], inappropriate workstation design appeared to have a potential role in provoking LBP. 2 Moreover, it was noticeable that office workers with a BMI ≤ 25 kg/m were more at risk for the 2 development LBP than those with a BMI >25 kg/m . As previously reported by the authors [10], it seems likely that the non-optimized physical dimensions of the workstations of individual workers may have contributed to the development of LBP. This would be particularly the case for those who were small and spent a prolonged period of time working with a computer. LPB is a chronic disorder in that the symptoms can often reoccur [1]. Even after completing successful treatment, repeated episodes of LBP seem likely when someone is re-exposed to the same risk factors or fails to improve their preventive behavior. Even if individual preventive behaviour improves, it is unlikely to be sufficient. We therefore propose that a process of ergonomic education is required to ensure that workstations are routinely designed so that can be optimally matched to an office worker’s individual physical needs. Other aspects of the work environment are also important such as the finding, for example, that the lighting intensity at 91.8% of the workstations was lower than the minimum standard requirement (600 lux). Insufficient illumination normally plays an important role in creating unsafe workplaces. In the use of computers by office workers, prolonged looking at screens can lead to eye strain and, in the long term, can cause MSDs when staff adopt unwise postures so that they can see their work more clearly [6]. Lighting intensity should therefore be adjusted to meet the standard for protection of the Thai labour force. Another important finding to emerge from the present study was the relationship between poor muscle strength and an increased risk of LBP. This suggests that staff should be made more aware of the need to take regular exercise and that tests of physical fitness should be used in the health risk assessment and health monitoring of office staff. In conclusion, the apparent trend of an increasing incidence rate of LBP at longer period of the follow-up might be a warning of about adverse health effect among office workers due to longer exposure to office work hazards. However, the statistical significance of this trend is recommended for analysis in the future studies. The results provide an indication of the nature of office work hazards affecting LBP in office workers in the academic workplace. In order to prevent LBP in this group of office workers, it is proposed that ergonomics education and the better design of workstations, improved lighting, and a greater emphasis on the physical fitness of staff are needed. Acknowledgements We acknowledge research-funding support from Khon Kaen University, Khon Kaen, Thailand. The authors declare that they have no competing interests.

References 1.

 

Andersson GBJ.Epidemiological features of chronic low-back pain. Lancet 1999; 354: 581-5.

8  

Proceedings 19th Triennial Congress of the IEA, Melbourne 9-14 August 2015

2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

 

Hoy D, Brooks P, Blyth F, Buchbinder R. Epidemiology of low back pain. Best Practice & Research Clinical Rheumatology. 2010; 24(6): 769-81. Chiu TTW, WY K, Lee MH, Sum MH, Wan MP, Wong CY, et al. A study on the prevalence of risk factors for neck pain among university academic in Hong Kong. Journal of Occupation Rehabilitation 2002; 12: 77-91. Kamwendo K, Moritz U. Neck and shoulder disorders in medical secretaries. Part I. Pain prevalence and risk factors. Scandinavian Journal of Rehabilitation Medicine 1991; 23: 127-33. Tsuboi H, Takeuchi K, Watanabe M, Hori R, Kobayashi F. Psychosocial factors related to musculoskeletal pain among school personnel in Nagoya, Japan. Japan Industrial Health 2001; 40: 266-71. Chaiklieng S, Suggaravetsiri P. Risk factors for repetitive strain injuries among school teachers in Thailand. WORK: A Journal of Prevention, Assessment & Rehabilitation 2012; 42: 2510-5. Fredriksson K, Alfredsson L, Ahlberg G, Josephson M, Kilbom A, Jelm EW, et al. Work environment and neck and shoulder pain: the influence of exposure time. Results from a population based case-control study. Occupational and Environmental Medicine 2002; 59: 182-8. Spyropoulos P, Papathanasiou G, Georgoudis G, Chronopoulos E, Koutis H, Koumoutsou F. Prevalence of musculoskeletal pain in Greek public office workers. Pain Physician 2007; 10: 651-60. Linton SJ. A review of psychosocial risk factors in back and neck pain. Spine 2000; 25: 1148-56. Chaiklieng S, Suggaravetsiri P, Boonprakob Y. Work ergonomic hazards for musculoskeletal pain among university office workers. Walailak Journal of Science and Technology 2010; 7(2): 169-76. Lwanga SK, Lemeshow S. Sample size determination in health studies: a practical manual. Geneva: World Health Organization; 1991. Mahmud N, Kenny DT, Zein RM, HassanSN. Ergonomic training reduces musculoskeletal disorders among office workers: results from the 6-month follow-up. Malaysian Journal of Medical Science 2011; 8(2): 16-26. Gourmelen J., Chastang J.-F., Ozguler J.-L., Lanoe J.-L., Ravaud J.-F. and Leclerc A. Frequency of low back pain among men and women age 30-64 years in France. Results of two national surveys Annales Readaptation Medicine Physigue 2007; 50: 640-4. Department of Health, Ministry of Public Health. Working health promotion manual, 6th ed. Nonthaburi: Ministry of Public Health, Thailand; 2007. Department of Mental Health, Ministry of Public Health, Stress manual. Bangkok: Ministry of Public Health, Thailand; 2010. Kuorinka I, Jonsson B, Kilbom A, Vinterberg H, Biering-Sorensen F, Andersson G, et al.Standardized Nordic questionnaires for the analysis of musculoskeletal symptoms. Applied Ergonomics 1987; 18: 233-7. Ministry of Labour, Thailand. Regulation on the standard of management of safety, occupational health and work environment; heat, light and sound, version 2006. [cited 2010 October 20]. Available from: http://www.labour. go.th/law/index.htm. Hush JM, Maher CG, Refschalke KM. Risk factors for neck pain in office workers: a prospective study. BMC Musculoskeletal Disorders 2006; 7(81): 1471-4. Bernard BP. Musculoskeletal disorders and workplace factors. A critical review of epidemiological evidence for work-related musculoskeletal disorders of the neck, upper extremity and low back. USA: Department of health and human services (NIOSH); 1997 Department of Disease Control, Ministry of Public Health, Thailand. MSDs among out-patients and in-patients; 2010. [cited 2011 November 6]. Available from URL http://occ.ddc.moph.go.th/

9