Subjects/Materials

The subjects for this study were employees belonging to the 89 job categories of

Standard Industrial Classification System. Data were collected using two separate databases as sources which connecting by 89 job categories: (1) Taiwan's occupational disease and injury notification system (ODIS) and (2) Survey of Perceptions of Safety and Health in the Work Environment.

3.2.1 Standard Industrial Classification System

The Standard Industrial Classification (SIC) is a system for classifying industries by a four-digit code, which was established in the United States in 1937. It is used by government agencies to classify industry areas. The SIC system is also used by agencies in other countries. In this study, we used the Standard Industrial Classification System of the Republic of China (Rev.9, 2011).

The SIC codes can be grouped into progressively broader industry classifications:

industry group, section (major group) and division. The first three digits of the SIC code indicate the industry group, and the first two digits indicate the major group. Each division encompasses a range of SIC codes. From 0100 till 0999, it stand for the division Agriculture, Forestry and Fishing; from 1000 till 1999, it stand for the division Mining; from 2000 till 3999, it stand for the division Manufacturing; from 4000 till 4999, it stand for the division Transportation, Communications, Electric, Gas and Sanitary service; from 5000 till 5199, it stand for the division Wholesale Trade; from

5200 till 5999, it stand for the division Retail Trade; from 6000 till 6799, it stand for the division Finance, Insurance and Real Estate; from 7000 till 8999, it stand for the division Services and from 9100 till 9729, it stand for the division Public Administration. The codes that start with 99 are No classifiable [36].

The 19 major groups and 89 divisions were listed in Table 1.

3.2.2 Collected and classified the WMSDs cases nearly four years (2008-2011)

In 2003, the system of occupational disease and injury services was launched, and the first center for Center of Occupational Disease and Injury Service (CODIS) was established. In 2007, a coordination center, Center of occupational Disease and Injury Management and Service (CODIMS), was also established [37]. In 2009, there were nine CODISs to provide the services, including three centers in northern, three in central, two in southern, and one in eastern Taiwan. Furthermore, each CODIS collaborates with surrounding hospitals to develop a network for occupational disease and injury services.

By the end of 2012, 56 network hospitals were established (Figure 2). The CODISs and network hospitals provide occupational health services, including diagnosing and preventing occupational diseases or injuries, occupational rehabilitation or return-to-work, and insurance compensation assistant.

The main tasks of CODIMS are:

(1) To provide the professional consultation on the management of occupational disease and injury.

(2) To establish the standard operation procedures for CODISs’ services, to ensure the quality of CODISs’ services.

(3) To investigate the clusters of occupational diseases, to develop the strategy for the prevention of occupational diseases or injuries.

(4) To maintain the reporting system of occupational diseases and injuries.

The reporting system of occupational diseases and injuries was formally launched in September 2007, which was based on the principle of previous reporting system at Department of Health [38].

The workers with injury or disease went the CODISs or network hospitals for help, and assessing, diagnosing and comparing with epidemiology studies that the occupational exposure factors were related more than 50% from the disease. Then, the doctor will report the case to Taiwan's occupational disease and injury notification

system (ODIS).

In this study, we collected the reported-cases from ODIS from 2008 to 2011 and exported data as excel form for data analysis. The reported cases used the “date of visit”

to collect cases, and its quality was examined by the reviewer who classified the cases to “probable”, “possible”, “not related” and “cannot be determined”. In our study, we

only choose the “probable” and “possible” cases for analysis.

We classified all the reported cases to occupational pulmonary disease, occupational liver disease, occupational skin disease, occupational cardiovascular disease, occupational neurological disease, occupational hematological disease, occupational renal diseases, occupational musculoskeletal diseases, occupational hearing loss, occupational eye disease, occupational physical disease, biological hazard and others. Base on the total number of employed population, the incidence of cases per 10⁶ populations per year could be calculated.

3.2.3 Selected and recognized the disease types of WMSDs

The WMSDs reported cases from 2008 to 2011 were 2,239 cases in total, accounting for 46.8% of all the occupational disease notification. We used International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) to classify WMSDs. It was divided into upper extremities (including hand / wrist, elbow, shoulder), cervical and lumbar region and lower extremities (knee, foot / ankle) three categories according to the diseased parts and two or more diseased parts of the case will be classified in others.

In the upper extremities, we selected carpal tunnel syndrome (CTS) (ICD 354.0 Carpal tunnel syndrome), and in the cervical and lumbar region 722.93 Other and

unspecified disc disorder of lumbar region and 722.73 Lumbar region of intervertebral disc disorder with myelopathy were classified in herniated intervertebral disc (HIVD).

Furthermore, the joint region and lower extremities did not include in this study.

3.2.4 Survey of Perceptions of Safety and Health in the Work Environment

In order to provide workers with an equitable, humane, secure, and dignified environment it has always been the objective and responsibility of the Government. The Institute of Occupational Safety and Health (IOSH) has conducted surveys every three years since 1994 to characterize demand and perception of employees, employers and self-employed people for safety and health in the workplace [39].

Our study used the Survey of Perceptions of Safety and Health in the Work Environment in 2010 to get the self-reported information about working conditions such as workplace environmental ergonomic factors and musculoskeletal discomfort in the workers. We took the proportion of ergonomic factors, and the proportion of musculoskeletal discomforts in each job category.

3.3 Data analysis

Flow chart of the statistical methods was shown in Figure 4. 

First, sex and age adjusted standardized Incidence Rate (SIR) for CTS and HIVD in every job category (89 divisions) were calculated. Then, the Survey of Perceptions of

Safety and Health in the Work Environment 2010 were used to count the situation of workplace in self-reported ergonomics, and musculoskeletal discomforts in each job category. Finally, simple and multiple linear regressions were to identify the relationship between WMSDs SIR and the risk factors.

3.3.1 Standardized incidence rates (SIR)

Because musculoskeletal disorders are, more common in women and will increase with age [40]. And the population component in every industry was not equivalent, it was necessary to standardize the incidence rate. In this study, SIR calculated by direct standardization using sex and age group, and use national total workforce as standard.

Incidence was calculated by 2008-2011 reported WMSDs from Taiwan's occupational disease and injury notification system (ODIS) by estimated numbers of employees in each job category. The denominator was the averaged manpower in each job category in 2008-2010 from Yearbook of manpower survey statistics. The calculation process of SIR was shown in Figure 3.

3.3.2 Linear regression

Simple and multiple linear regressions were to identify the relationship between SIR of WMSDs and the risk factors. Since the incidences of every industry were not normal distribution, we did a log-transformed to the SIR in every job category before

simple linear regression. Considering the different working population, we used the manpower survey statistical working population as weight in the statistical analysis. In the statistical model, dependent variables were log-transformed SIR of WMSDs and independent variables were the workplace environmental ergonomic factors, and musculoskeletal discomfort from Survey of Perceptions of Safety and Health in the Work Environment.

Then chose those p-values less than 0.05 from the results of simple linear regression as variables of multiple linear regression. The backward stepwise multiple linear regression was adopted here. The remained risk factors had greater contribution to WMSDs. Finally, the remained variables from multiple linear regression were chosen.

By using the median as the cut point and setting the dummy variables for the multiple linear regression, it can be checked whether there was any combination of risk factors for WMSDs.

3.3.3 Geometric mean of SIR

In order to show the relationship between the SIR of WMSDs and the risk factors clearly when finding the high correlation in WMSDs and risk factors, transforming the log-SIR back to the original SIR was needed. Furthermore, divided the exposure factors into four groups (Q1-Q3 as the cut point), and showed the relationship between the SIR

od WMSDs and the percentage level of exposure. By comparing the four groups to its SIR, we got the trend of exposure factors in every industries and the risk level of suffering from WMSDs.