Originals
Changes in Psychosocial Work Conditions in Taiwanese Employees by Gender and Age from 2001 to 2010
2013 Volume 55 Issue 5 Pages 323-332
Details
2013 Volume 55 Issue 5 Pages 323-332
Objective: The aim of this study was to examine changes in working hours, shift work, psychological and physical job demands, job control and job insecurity in Taiwanese employees by gender and age during the period of 2001 to 2010. Methods: The study subjects were 36,750 men and 27,549 women, aged 25-64, from 4 rounds of cross-sectional surveys of representative employees. Psychosocial work conditions were assessed by a validated questionnaire. Results: Regression analyses with adjustment of education and employment grade showed that from 2001 to 2010, the proportions of workers with long working hours (>48 hours/week) (OR=1.4 in men and 1.5 in women) and workers with short working hours (<40 hours/week) (OR=1.3 in both genders) both increased over time, indicating an increasing polarization in the distribution of working hours. Furthermore, the proportions of nonstandard work shifts (OR=1.7 in men and 2.1 in women) and work with high physical demands (OR=1.5 for both gender) increased. There were signs of decreasing levels of job control from 2001 to 2007, which seemed to be more apparent in younger workers than in older workers. However, a slight recovery in decision latitude and opportunity for learning was noticed in later years. The trend in job insecurity was not linear, with the highest prevalence found in 2004. Conclusions: Our findings suggested that certain aspects of psychosocial work environment had deteriorated in Taiwan. There is a need to raise public awareness about the changing patterns of psychosocial health risks at work as well as their causes and their potential impacts on worker well-being.
(J Occup Health 2013; 55: 323-332)
Growing epidemiologic evidence indicates that a number of adverse work characteristics are associated with increased risks of cardiovascular diseases and other stress-related physical and mental health problems1-5). Major psychosocial work hazards identified include long working hours, shift work, high job demands, lack of job control, monotonous and repetitive work task, insufficient quality of social relationship, precarious employment, bullying, violence, family-work conflict, effort-reward imbalance and organizational injustice1, 3, 6, 7).
Several economically advanced countries have established surveillance systems or incorporated measures in population-based surveys to monitor the nature of psychosocial work conditions in the general working population8). In some Western countries, surveys of nationally representative employees have been carried out periodically9-13), and in Europe, the European Working Conditions Surveys (EWCS) are the major source of information, which have been conducted among representative employees and self- employed persons in EU member states every 5 years since 199014).
It is a commonly held notion that psychosocial work conditions have been deteriorating on a global scale due to globalization, labor deregulation and recent economic turmoil15-17). However, studies based on surveillance data, mostly from Western countries, have generated mixed findings—some reported negative changes9, 11), some reported little changes12, 14) and some reported improvement over time18). While the “export” of monotonous jobs to lower-income countries might help to improve the overall quality of work in wealthier countries9), for workers in lower-income countries, one might expect to observe an increasing number of unfavorable jobs along with the accelerating international division of labor19). Nevertheless, empirical evidence from lower-income countries concerning the distribution and changing trends of psychosocial work conditions have been very limited.
Another limitation in current literature is that few studies had looked into gender and age aspects when examining the trends of psychosocial work conditions. Taking into account the persistent gender division in labor and substantial increases in youth unemployment, one may expect to observe differential changes in psychosocial work conditions across gender and age groups. For instance, one may wonder if women are more vulnerable to the impacts of labor market changes, taking into consideration that women are more likely than men to experience unfavorable working conditions such as low decision authority, routine and repetitive work, job insecurity and low job rewards20, 21). With regard to age, it is known that young people in many countries have been harshly affected by the recent economic downturn22). In Taiwan, unemployment rates had been fluctuating since the early 1990s but overall have been on the rise—from 1.7 to 5.8% in men and from 1.6 to 4.5% in women during the period of 1990 to 2010. Younger people had a higher unemployment rate than other age groups and experienced a greater increase —for men and women aged 25-30 years, from 2.0 to 9.4% and 1.7 to 6.9% during the period of 1990 to 2010. Changes in psychosocial work conditions among young people deserve attention. On the other hand, the situations among older groups should be of concern when the workforce is aging rapidly.
In this study, we utilized data from 4 rounds of surveys of representative employees conducted in 2001, 2004, 2007 and 2010 to examine the trends in the distribution of working hours and a number of selected psychosocial work factors. In addition to age- and gender-stratified descriptive analyses, we further examined changes in psychosocial work conditions with adjustment of changes in educational attainment and employment grade across the four survey cycles, taking into consideration that demographic shifts that might have occurred over the studied period.
The Council of Labour Affairs of Taiwan has conducted nationwide cross-sectional surveys with an interval of 3 to 5 years since 1988. The surveys are designed to assess multiple dimensions of work conditions and occupational safety and health issues in the working population.
Subjects were selected through a two-stage random sampling process. In the first stage, all districts and villages throughout Taiwan were grouped into strata according to their levels of urbanization. A random sample of districts and villages was chosen from each stratum. In the second stage, a random sample of households was selected within each district or village, and residents of the sampled households who were currently working at the time of the survey were identified. Self-administered questionnaires were delivered to the selected households by trained interviewers. Subjects were informed of the purpose of the survey and invited to participate in a face-to-face manner. After one week, completed questionnaires were collected, and on-site checking was performed by the same interviewer. The official report of the latest survey is available online (Survey of Perception of Safety and Health in the Work Environment in Taiwan, 2010; http://www.iosh.gov.tw/Book/ Report_Publish.aspx?P=1607).
Data for this study was drawn from 4 rounds of national surveys conducted in September of 2001, 2004, 2007 and 2010. The response rates of the 4 surveys were 82, 81, 86 and 87%, respectively. For this study, we restricted analyses to employees aged between 25 and 65 years old. This resulted in a total of 36,750 men and 27,549 women for analyses.
Measures for psychosocial work conditionsPsychosocial work demands and job control were assessed by the Chinese version of the Job Content Questionnaire (C-JCQ) based on Karasek's job strain model23, 24). This model postulates that a combination of high demands and low control causes high job strain that leads to negative health outcomes.
Five core items for the demands scale (work fast, hectic, excessive work, not enough time and concentrate on job for long time) were included in the questionnaires for all four surveys. Cronbach's alpha coefficients for the 5-item job demands scale were 0.56 for men and 0.63 for women in the survey in 2010.
The job control scale consists of two sub-components: 6 items for the skill discretion subscale (learning new things, high level of skills, non-repetitive work, creative work, various tasks, develop one's abilities) and 3 items for the decision authority subscale (allowed to make own decisions, freedom to make decision, opinions influential). All 9 items were included in the questionnaires for the surveys in 2001, 2004 and 2007, with Cronbach's alpha coefficients well above 0.77. In 2010, however, the items for the job control scale were reduced to 3 items because the questionnaire was reoriented and items for other competing dimensions were expanded.
All items mentioned above were listed as a statement with response recorded on a four-point Likert scale ranging from 1 (strongly disagree) to 4 (strongly agree). Items stated oppositely were reversely coded in order to make the items additive in the same direction. The mean scores were then calculated and standardized to the range from 0 to 100 by the following formula: ((summed score − minimum score/score range)* 100).
Also included in this study were 1 item for job insecurity (my job is secure) and 1 item for physical demands (my work is physically demanding). Their responses were recorded on a four-point scale (strongly agree to strongly disagree) and re-coded dichotomously (agree/disagree) for analyses.
Subjects were also asked to provide information regarding their job title, total working hours and the status of shift work within the week prior to the survey date. Employment grade was classified according to the following 6 categories based on self-reported job titles: Grade 1 (administrators and managers), Grade 2 (professionals), Grade 3 (non-manual skilled workers), Grade 4 (non-manual low-skilled workers), Grade 5 (manual skilled workers) and Grade 6 (manual low-skilled workers).
Statistical analysisDescriptive analyses were performed to examine distribution of work conditions by gender and across 3 age groups (aged 25-34, 35-54 and 55-64). Differences in work conditions of each year (dummy coded) compared with the reference year, 2010, were examined by regression models, with adjustment of education level and employment grade. Logistic regression models were used for dichotomous outcomes, and linear regression models were used for continuous outcomes. To test if there were linear trends in work conditions across the 4 surveys, tests for trends were performed by assigning scores of 1 to 4 to the four sequential years and treating the scored variable as a continuous predicting variable. Because psychosocial work factors are known to differ by gender, all analyses were stratified by gender. SAS 9.1 (SAS Institute, Cary, NC, USA) was used for all of the analyses.
As shown in Table 1, the proportion of workers aged between 55 and 64 years increased from 6.1 to 11.1% in men and from 3.5 to 7.2% in women during the study period from 2001 to 2010. Over the same period, the proportion of employees with a college or higher education also increased, and in the female workforce, there was a substantial decrease in the proportion of low-skilled manual workers, from 27.0% in 2001 to 20.8% in 2010. When comparing the levels of education and employment grades across the 3 age groups, it was found that as compared with older groups, younger groups had higher levels of education and were more likely to be employed as professional or skilled workers (data not shown).
| Men | Women | |||||||
|---|---|---|---|---|---|---|---|---|
| 2001 (n=8,692) | 2004 (n=8,893) | 2007 (n=9,622) | 2010 (n=9,543) | 2001 (n=5,979) | 2004 (n=6,377) | 2007 (n=7,394) | 2010 (n=7,799) | |
| Age group (years) | ||||||||
| 25.34 | 36.6% | 33.6% | 34.1% | 32.9% | 41.1% | 39.0% | 39.1% | 38.6% |
| 35.54 | 57.3% | 59.7% | 57.4% | 56.0% | 55.4% | 56.9% | 55.3% | 54.2% |
| 55.64 | 6.1% | 6.6% | 8.6% | 11.1% | 3.5% | 4.1% | 5.6% | 7.2% |
| Education (school years) | ||||||||
| Junior school or below (≤9) | 31.1% | 28.4% | 25.9% | 20.9% | 28.7% | 24.1% | 22.3% | 17.6% |
| High school (10-12) | 34.2% | 33.3% | 35.0% | 34.5% | 35.3% | 34.2% | 33.6% | 33.6% |
| College (13-16) | 31.0% | 33.1% | 33.5% | 37.4% | 34.0% | 38.1% | 40.3% | 43.7% |
| Graduate (>16) | 3.8% | 5.2% | 5.6% | 7.3% | 2.1% | 3.6% | 3.8% | 5.1% |
| Employment grade | ||||||||
| G1: Managers | 4.4% | 4.9% | 4.3% | 4.3% | 1.2% | 1.6% | 1.3% | 1.3% |
| G2: Professionals | 8.3% | 9.2% | 9.5% | 10.0% | 11.2% | 12.6% | 11.6% | 11.9% |
| G3: Non-manual skilled | 22.9% | 22.9% | 22.4% | 22.9% | 21.1% | 22.4% | 22.9% | 23.8% |
| G4: Non-manual low-skilled | 13.4% | 12.9% | 12.9% | 14.9% | 36.2% | 35.5% | 37.0% | 39.6% |
| G5: Manual skilled | 21.5% | 20.6% | 21.8% | 19.5% | 3.3% | 2.0% | 2.6% | 2.7% |
| G6: Manual low-skilled | 29.5% | 29.5% | 29.2% | 28.5% | 27.0% | 25.9% | 24.7% | 20.8% |
As shown in Table 2, most of the examined psychosocial working conditions showed negative developments, i.e., trends toward worsening working conditions, with the exception of a nonlinear change in job insecurity—which showed the highest level in 2004. Taken as a whole, the proportions of workers with short working hours (<40 hours per week) and long working hours (>48 hours per week) both increased. When we further examined the full range of changes in the distribution of working hours, as shown in Fig. 1, we found that the proportion of workers with working hours in the range of 41-48 hours per week had decreased steadily, while the proportions of workers with working hours longer than 48 hours per week had increased. However, the patterns of working hours shorter than 40 hours per week appeared to be nonlinear, with the lowest prevalence found in 2004 and a decrease thereafter.
Distribution of weekly working hours in male and female employees of the 4 surveys
| Men | Women | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 2001 (n=8,692) | 2004 (n=8,893) | 2007 (n=9,622) | 2010 (n=9,543) | pa | 2001 (n=5,979) | 2004 (n=6,377) | 2007 (n=7,394) | 2010 (n=7,799) | pa | |
| Work hours <40 h/wk | ||||||||||
| All | 8.3% | 5.4% | 7.3% | 9.1% | ** | 6.9% | 5.4% | 7.3% | 7.8% | ** |
| 25-34 | 6.5% | 3.8% | 4.1% | 5.8% | 4.6% | 3.2% | 4.8% | 5.3% | ||
| 35-54 | 9.3% | 5.9% | 8.1% | 10.0% | * | 8.1% | 6.6% | 8.2% | 8.6% | |
| 55-64 | 10.1% | 9.7% | 14.7% | 14.0% | * | 16.0% | 11.2% | 14.9% | 15.1% | |
| Work hours>48 h/wk | ||||||||||
| All | 10.3% | 12.7% | 14.6% | 13.7% | ** | 8.2% | 9.1% | 10.6% | 10.8% | ** |
| 25-34 | 11.0% | 12.5% | 13.8% | 13.4% | * | 9.0% | 9.3% | 10.0% | 11.0% | * |
| 35-54 | 9.8% | 12.9% | 15.0% | 14.0% | ** | 7.5% | 9.1% | 10.7% | 10.7% | ** |
| 55-64 | 10.3% | 12.2% | 14.8% | 13.3% | 9.7% | 8.1% | 13.7% | 11.0% | ||
| Nonstandard shift (evening/night/rotating/irregular shift vs. fixed day shift) | ||||||||||
| All | 16.6% | 21.9% | 20.3% | 24.0% | ** | 12.0% | 16.3% | 17.3% | 20.4% | ** |
| 25-34 | 17.0% | 20.8% | 20.9% | 26.4% | ** | 14.5% | 17.8% | 20.1% | 24.2% | ** |
| 35-54 | 16.4% | 22.5% | 19.6% | 22.4% | ** | 10.3% | 15.0% | 15.6% | 17.4% | ** |
| 55-64 | 15.2% | 21.9% | 22.2% | 23.9% | ** | 8.7% | 17.4% | 13.9% | 20.6% | ** |
| Physically demanding (yes vs. no) | ||||||||||
| All | 52.0% | 55.8% | 55.8% | 55.8% | ** | 38.9% | 42.5% | 41.0% | 44.9% | ** |
| 25-34 | 52.2% | 56.0% | 55.6% | 59.7% | ** | 34.2% | 38.8% | 38.0% | 43.5% | ** |
| 35-54 | 51.7% | 56.3% | 56.6% | 57.4% | ** | 41.2% | 44.8% | 42.2% | 45.0% | |
| 55-64 | 52.9% | 50.7% | 50.9% | 53.9% | 56.8% | 45.7% | 51.0% | 51.3% | ||
| Job insecurity (yes vs. no) | ||||||||||
| All | 49.0% | 55.8% | 53.2% | 46.4% | ** | 51.1% | 56.4% | 54.5% | 48.5% | ** |
| 25-34 | 50.2% | 55.9% | 53.7% | 49.0% | 49.4% | 54.6% | 54.6% | 50.3% | ||
| 35-54 | 49.1% | 56.4% | 53.2% | 45.8% | ** | 51.7% | 57.7% | 54.5% | 47.6% | ** |
| 55-64 | 41.0% | 49.8% | 51.2% | 41.2% | 59.2% | 55.8% | 53.6% | 45.8% | ** | |
| Psychological job demands (5 items) | ||||||||||
| All | 51.5 | 52.6 | 52.6 | 52.4 | ** | 51.3 | 52.2 | 51.9 | 52.3 | ** |
| 25-34 | 52.4 | 53.5 | 53.3 | 53.7 | ** | 52.1 | 53.1 | 53.1 | 53.7 | ** |
| 35-54 | 51.1 | 52.4 | 52.7 | 52.3 | ** | 50.7 | 51.8 | 51.4 | 51.7 | * |
| 55-64 | 50.1 | 49.4 | 49.6 | 49.4 | 50.2 | 48.7 | 48.8 | 49.6 | ||
| Job control (9 items) | ||||||||||
| All | 55.6 | 53.3 | 53.2 | — | ** | 52.4 | 50.9 | 50.2 | — | ** |
| 25-34 | 56.1 | 53.5 | 53.2 | — | ** | 54.0 | 52.0 | 50.9 | — | ** |
| 35-54 | 55.6 | 53.3 | 53.4 | — | ** | 51.7 | 50.5 | 50.1 | — | ** |
| 55-64 | 52.6 | 52.2 | 51.4 | — | 45.7 | 45.8 | 46.5 | — | ||
* p<0.01; ** p<0.001. a Test for trend.
There were significant increases in nonstandard work shifts (including evening shift, night shift, rotating shift and irregular shift), physically demanding jobs and psychological job demands, and significant decreases in job control. Similar trends were observed in most of the gender and age subgroups.
The responses of individual items for psychological job demands and job control were examined graphically, and the results are presented in Fig. 2 and Fig. 3, respectively. No apparent trend was found for job demands, but a closer examination revealed that the proportions of employees reporting high “excessive work” and “not enough time” were higher in 2004 than in other years.
Changes in responses of items for job demands from 2001 to 2010.
Changes in responses for items concerning job control from 2001 to 2010.
The 9-item scale for job control was used in the first 3 surveys, but the scale was reduced to 3 items in 2010. When examined individually, most of the item responses showed declining trends during the period from 2001 to 2007. However, a slight recovery in two of the 3 remaining items—“learn new things” and “allowed to make own decisions”—was noticed in 2010.
When education and employment grade were adjusted in the multivariate regression models, the proportions of workers with working hours fewer than normal (40 hour/week) were found to increase significantly from 2001 to 2010; meanwhile, the proportion of workers with long working hours (>48 hours/week) had also increased. Furthermore, it was noticed that nonstandard work shifts became more prevalent from 2001 to 2010. In both men and women, significant increases in physical job demands and psychological job demands and significant decreases in job control were found in the young and middle-aged groups over the four surveys but not in the older age groups.
The findings of this study indicated that during the period of 2001 to 2010, some psychosocial work conditions had deteriorated in Taiwanese employees, even though the overall level of educational attainment had increased substantially. After adjusting for changes in educational level and employment grades across the four surveys, we found increasing trends in the proportion of employees with long working hours (>48 hours per week) and the proportion of employees with short working hours (<40 hours per week). This suggests that the distribution of working hours might have become more polarized. In addition, there was evidence of increasing jobs with a nonstandard work shift and jobs with high physical demands.
According to Taiwan's official statistics, there has been a downward trend in average working hours since the early 1980s. However, the trend had flattened and even reversed in recent years. While prolonged working hours are known to increase health risks25-27), work with insufficient working hours should be of concern because it is often associated with underemployment or precarious employment in the labor market context of Taiwan and other East Asian countries, where long working hours are the norm and social security programs for part-time workers are much weaker as compared with those in Western countries28, 29). Our observation of more unevenly distributed working hours might be associated with organizational downsizing and an increase in nonstandard employment in later years. For instance, a survey conducted by Taiwan's Directorate-General of Budget, Accounting and Statistics (DGBAS) in 2005 showed that 16% of organizations admitted having downsized to some extent30), and a later survey, also conducted by the DGBAS, indicated that workers working part-time or on a temporary basis had increased substantially in recent years. Decreases in work with normal working hours and standard working times are expected to result in negative consequences, not only in terms of occupational safety and health but also in terms of work-family balance and social participation.
| Men | Women | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 2001 | 2004 | 2007 | 2010 | 2001 | 2004 | 2007 | 2010 | |||
| (n=8,692) | (n=8,893) | (n=9,622) | (n=9,543) | pb | (n=5,979) | (n=6,377) | (n=7,394) | (n=7,799) | pb | |
| Work hours <40 h/w | ||||||||||
| All | OR=1 | 0.6 (0.6, 0.7)** | 0.9 (0.8, 1.0) | 1.3 (1.2, 1.5)** | ** | OR=1 | 0.8 (0.7, 0.9)* | 1.1 (1.0, 1.3) | 1.3 (1.1, 1.5)** | ** |
| 25-34 | OR=1 | 0.6 (0.5, 0.7)** | 0.7 (0.5, 0.8)** | 1.1 (0.9, 1.4) | OR=1 | 0.7 (0.5, 0.9) | 1.1 (0.8, 1.4) | 1.2 (0.9, 1.5) | ||
| 35-54 | OR=1 | 0.6 (0.5, 0.7)** | 0.9 (0.8, 1.0) | 1.3 (1.1, 1.5)** | ** | OR=1 | 0.8 (0.7, 1.0) | 1.1 (0.9, 1.3) | 1.2 (1.1, 1.5)* | ** |
| 55.64 | OR=1 | 1.0 (0.7, 1.5) | 1.6 (1.1, 2.3)* | 1.7 (1.2, 2.4)* | ** | OR=1 | 0.7 (0.4, 1.2) | 1.0 (0.6, 1.6) | 1.1 (0.7, 1.7) | |
| Work hours>48 h/w | ||||||||||
| All | OR=1 | 1.3 (1.2, 1.4)** | 1.5 (1.4, 1.7)** | 1.4 (1.3, 1.6)** | ** | OR=1 | 1.2 (1.0, 1.4)* | 1.4 (1.3, 1.6)** | 1.5 (1.3, 1.7)** | ** |
| 25-34 | OR=1 | 1.2 (1.0, 1.4) | 1.4 (1.2, 1.6)** | 1.4 (1.2, 1.6)** | ** | OR=1 | 1.1 (0.9, 1.4) | 1.3 (1.1, 1.6)* | 1.5 (1.3, 1.8)** | ** |
| 35-54 | OR=1 | 1.4 (1.2, 1.5)** | 1.6 (1.5, 1.9)** | 1.5 (1.4, 1.7)** | ** | OR=1 | 1.3 (1.1, 1.5)* | 1.6 (1.3, 1.8)** | 1.6 (1.4, 1.9)** | ** |
| 55.64 | OR=1 | 1.3 (0.9, 1.9) | 1.7 (1.2, 2.3)* | 1.5 (1.1, 2.1) | OR=1 | 0.9 (0.5, 1.7) | 1.6 (0.9, 2.7) | 1.3 (0.7, 2.2) | ||
| Nonstandard shift (evening/night/rotating/irregular shift vs. fixed day shift) | ||||||||||
| All | OR=1 | 1.5 (1.3, 1.6)** | 1.3 (1.2, 1.4)** | 1.7 (1.5, 1.8)** | ** | OR=1 | 1.5 (1.3, 1.6)** | 1.6 (1.5, 1.8)** | 2.1 (1.9, 2.3)** | ** |
| 25-34 | OR=1 | 1.3 (1.2, 1.5)** | 1.3 (1.2, 1.5)** | 1.8 (1.6, 2.1)** | ** | OR=1 | 1.3 (1.2, 1.6)** | 1.6 (1.4, 1.9)** | 2.2 (1.9, 2.5)** | ** |
| 35-54 | OR=1 | 1.5 (1.4, 1.7)** | 1.3 (1.1, 1.4)** | 1.5 (1.4, 1.7)** | ** | OR=1 | 1.6 (1.4, 1.9)** | 1.7 (1.5, 2.0)** | 2.1 (1.8, 2.4)** | ** |
| 55.64 | OR=1 | 1.6 (1.2, 2.2)* | 1.6 (1.2, 2.2)* | 1.9 (1.4, 2.5)** | ** | OR=1 | 2.3 (1.3, 4.2)* | 1.8 (1.0, 3.1) | 2.9 (1.7, 5.0)** | ** |
| Physically demanding (yes vs. no) | ||||||||||
| All | OR=1 | 1.2 (1.2, 1.3)** | 1.2 (1.2, 1.3)** | 1.5 (1.4, 1.6)** | ** | OR=1 | 1.2 (1.1, 1.3)** | 1.2 (1.1, 1.3)** | 1.5 (1.4, 1.6)** | ** |
| 25-34 | OR=1 | 1.2 (1.1, 1.4)** | 1.3 (1.1, 1.4)** | 1.6 (1.5, 1.8)** | ** | OR=1 | 1.2 (1.1, 1.4)** | 1.2 (1.1, 1.4)** | 1.6 (1.4, 1.8)** | ** |
| 35-54 | OR=1 | 1.3 (1.2, 1.4)** | 1.3 (1.2, 1.4)** | 1.5 (1.4, 1.6)** | ** | OR=1 | 1.3 (1.1, 1.4)** | 1.2 (1.1, 1.3)* | 1.5 (1.3, 1.6)** | ** |
| 55.64 | OR=1 | 1.0 (0.7, 1.2) | 0.9 (0.7, 1.2) | 1.2 (1.0, 1.5) | OR=1 | 0.7 (0.5, 1.0) | 0.9 (0.7, 1.4) | 1.2 (0.9, 1.7) | ||
| Job insecurity (yes vs. no) | ||||||||||
| All | OR=1 | 1.4 (1.3, 1.5)** | 1.2 (1.2, 1.3)** | 1.0 (0.9, 1.0) | OR=1 | 1.3 (1.2, 1.4)** | 1.2 (1.1, 1.3)** | 1.0 (0.9, 1.1) | ||
| 25-34 | OR=1 | 1.3 (1.2, 1.5)** | 1.2 (1.1, 1.4)** | 1.0 (0.9, 1.2) | OR=1 | 1.3 (1.2, 1.5)** | 1.3 (1.2, 1.5)** | 1.1 (1.0, 1.3) | ||
| 35-54 | OR=1 | 1.4 (1.3, 1.5)** | 1.2 (1.1, 1.3)** | 1.0 (0.9, 1.0) | OR=1 | 1.4 (1.3, 1.6)** | 1.2 (1.1, 1.4)** | 1.0 (0.9, 1.1) | ||
| 55.64 | OR=1 | 1.6 (1.3, 2.1)** | 1.6 (1.3, 2.1)** | 1.2 (0.9, 1.5) | OR=1 | 1.0 (0.7, 1.4) | 0.9 (0.6, 1.3) | 0.7 (0.5, 1.1) | ||
| Psychological job demands (5 items) | ||||||||||
| All | β=0 | 0.9 (0.6, 1.3)** | 0.9 (0.6, 1.3)** | 0.5 (0.2, 0.9)* | * | β=0 | 0.7 (0.3, 1.2)* | 0.4 (−0.1, 0.8) | 0.7 (0.2, 1.1)* | |
| 25-34 | β=0 | 1.0 (0.4, 1.6)* | 0.6 (0.0, 1.2) | 1.0 (0.4, 1.6)* | * | β=0 | 0.8 (0.1, 1.5) | 0.5 (−0.2, 1.3) | 1.0 (0.3, 1.7)* | |
| 35-54 | β=0 | 1.2 (0.7, 1.6)** | 1.4 (1.0, 1.9)** | 0.8 (0.4, 1.3)** | ** | β=0 | 0.9 (0.3, 1.5)* | 0.5 (−0.1, 1.0) | 0.7 (0.2, 1.3)* | |
| 55.64 | β=0 | −0.8 (−2.0, 0.5) | −0.6 (−1.7, 0.6) | −0.8 (−1.9, 0.3) | β=0 | −1.5 (−3.6, 0.7) | −1.2 (−3.1, 0.8) | −0.3 (−2.2, 1.5) | ||
| Job control (9 items) | ||||||||||
| All | β=0 | −2.7 (−3.1, −2.3)** | −2.8 (−3.2, −2.4)** | — | ** | β=0 | −2.1 (−2.6, −1.7)** | −2.9 (−3.3, −2.4)** | — | ** |
| 25-34 | β=0 | −3.0 (−3.6, −2.3)** | −3.5 (−4.2, −2.8)** | — | ** | β=0 | −2.5 (−3.2, −1.7)** | −3.6 (−4.3, −2.8)** | — | ** |
| 35-54 | β=0 | −2.7 (−3.2, −2.2)** | −2.5 (−3.0, −2.0)** | — | ** | β=0 | −2.0 (−2.6, −1.4)** | −2.6 (−3.2, −2.0)** | — | ** |
| 55.64 | β=0 | −1.2 (−2.7, 0.3) | −1.4 (−2.8, −0.0) | — | β=0 | −0.8 (−1.7) | −0.3 (−2.5, 2.0) | — | ||
a Adjustment for education (4 groups) and employment grade (3 groups: G1&G2, G3&G4, G5&G6); logistic regression models are applied for dichotomized outcomes, and linear regression models are applied for continuous outcomes. b Test for trend. * p<0.01; ** p<0.001.
In this study, aggregated scores for job control were available only for the period from 2001 to 2007, which showed decreasing trends. Of the three remaining items adopted in the 2010 survey, two items—learning new things and allowed to make own decisions—showed a slight recovery, but the item for “work not repetitive” continue to decline. Consistencies in changing patterns of different dimensions of job control were not expected and deserved further investigation.
The highest level of job insecurity was found in 2004. In the same year, we also observed increases in two dimensions of job demands, i.e., “excessive work” and “not enough time”, and the lowest prevalence of workers with working hours less than 40 hours. We speculated that the peculiar pattern surrounding 2004 might be attributed to a massive pension reform implemented in July of the same year. It required employers to deposit a certain proportion of employees' pay into their individual pension accounts, regardless of the size of enterprise and seniority of workers. Many employers reacted to the sudden increase in financial burden by reducing their workforce and replacing regular employees with contractors or dispatched workers. Slight improvement in job security and some items of job control observed in 2010 might be related to the country's economic recovery after the global financial crisis that occurred in 2008-2009.
The negative developments in some psychosocial work factors observed in our study were consistent with findings of some studies in Western countries. For instance, a study by Pejtersen and Kristensen among Danish employees reported negative trends during the period from 1997 to 2005, including increases in work pace, more role conflicts and workplace violence, and decreases in authority latitude, skill discretion, coworker social support and sense of community9). Similarly, a study among Canadian employees reported negative changes from 1994 to 2005, including increases in long working hours and nonstandard times and declines in decision authority, skill discretion and coworker support11). However, data from the British national surveys reported little changes in the 6 dimensions of psychosocial work conditions, namely, demand, control, managerial and peer support, role conflict, social relationships at work and organizational change, between 2004 an 201012), and the last European Working Condition Survey found little or no change in job autonomy across European countries between 2000 and 2010, despite noticeable increases in the proportion of workers on temporary contracts and in the level of job insecurity14). In contrast, an earlier study found improvement in the work environment of Danish employees, including increases in job control, over the period from 1990 to 200018).
Previous studies have showed that in many Asian countries including Japan, South Korea and Taiwan, recent increases in unemployment rates were closely related to male suicide mortality rates31, 32). In Taiwan, suicide mortality increased substantially in recent years and was 8.1 to 22.7 per 100,000 men and 5.2 to 10.9 per 100,000 women during the period from 1990 to 2010. Furthermore, a recent publication by Fu and colleagues documented that the prevalence of probable common mental disorders doubled in Taiwan—from 11.5% in 1990 to 23.8% in 201033). The findings of this study suggested that certain psychosocial work conditions had been worsening, providing an explanation for substantial increases in stress-related health problems and compensation claims observed in recent years34).
At the country-specific level, it is obvious that the presence and extent for employment-related welfare programs and labor protection policies set the frame of working life experiences. Nevertheless, cross-country differences in these social contextual characteristics as well as their influences on the changing patterns of psychosocial work conditions and workers' stress experiences and health risks have rarely been addressed in the existing literature. Furthermore, whether or not workers of different countries or workers with different socioeconomic positions are affected differentially by globalization and recent economic turmoil deserves further investigation.
In Taiwan, as well as in Japan and South Korea, the surge of stress claims asking for workers' compensation has been a thorny policy issue in recent years. Most of the claims were cases of sudden deaths and severe cardiovascular events that had been associated with heavy workloads and long working hours34-37). While the problems of long working hours have been targeted, nevertheless, other dimensions of psychosocial work hazards are still largely overlooked by the general public and policy participants. The establishment of national surveillance of psychosocial hazards at work is essential for a better understanding of the nature and the changing trends in psychosocial work environment, and documentation of national trends can also help to raise social awareness and to aid public communication.
Our study had several strengths. First, the study population was randomly sampled based on household registry with participation rates well above 80%; thus it can be considered as a representative sample of employees in Taiwan. Second, the instruments for assessment of psychosocial work conditions were well validated, and the survey methods were consistent. Third, we examined changing patterns in psychosocial work conditions by gender and age sub-groups with adjustment of changes in education and employment grades over the four surveys.
A major limitation of this study is that we assessed only a few of psychosocial work hazards. Items and scales for important factors such as low rewards, violence and bullying at work, work-family conflict and organizational climate should be included in our further surveys. Secondly, it was noticed that Cronbach's alphas for the job demands scale were low, indicating that this scale should be improved further. Thirdly, our data were self-reported and collected on an individual basis, which means that our results might have been confounded by personality factors or have problems related to common-method variance. More objective measures and measures for organizational characteristics will be needed to understand the pathways linking macro-level factors to task- level work conditions. Furthermore, although the GDP per capita of Taiwan (20,006 US dollars, 2010) was much lower than most of the OECD countries (average 35,021 US dollars in 2011), it is not considered as a low-income country. Empirical evidences on psychosocial work conditions from low-income countries are still lacking.
In conclusion, our findings suggested that certain aspects of psychosocial work environment had deteriorated in Taiwan. There is a need to raise public awareness about the changing patterns of psychosocial health risks at work as well as their causes and potential impacts on worker well-being.
Acknowledgment: This study was supported by a research grant from the National Science Council (NSC 99-2410-H002-171-MY3).
