Occupational exposure to inhalable agents, such as chemicals, vapors, gas, dust, and fumes, is a common occurrence. In Norway, an estimated 23% of workers are exposed to such agents (1). Inhalable occupational exposure is associated with different respiratory diseases, such as asthma, chronic obstructive pulmonary disease (COPD), lung cancer, hypersensitivity pneumonitis, and pneumoconiosis. The population attributable fraction varies from 10% to 30%, depending on the type of exposure and respiratory disease (2). Occupational exposure to inhalable substances is a common, modifiable risk factor and an important contributor to the development of respiratory diseases.

The concept of work ability can be defined as “How good are workers at present and in the near future, and how able are they to do their job with respect to work demands and health and mental resources?”(3). Measuring work ability is complex; however, the Work Ability Index (WAI) is a validated and reliable instrument that has been used in epidemiological studies. Previous studies have shown that a low score predicts future work disability and may help identify employees who require support to improve their work ability (3). The WAI is not a static measure and is influenced by age, health status, weight, and level of physical work (4). The complete WAI battery is a comprehensive instrument that is inconvenient for large surveys; therefore, shorter questionnaires have been developed (5). One such questionnaire is the Work Ability Score (WAS), where the participants are asked to grade their current work ability on a scale from 0 (“I cannot work at all”) to 10 (“My work ability is at its best right now”). The WAS has been shown to quite effectively represent the entire WAI (4). In a study of health care workers, respiratory diseases were associated with reduced work ability measured using the WAS, with an odds ratio (OR) of 2.11 (95% confidence interval [CI], 1.77–2.51), compared with healthy employees (6). To our knowledge, no studies have assessed whether exposure to inhalable agents, such as vapors, gas, dust, and fumes (VGDF), can predict future reduced work ability.

Therefore, the main aim of this study was to use self-reported data on inhalable occupational exposure and data from the asthma-specific northern European job-exposure matrix (N-JEM) to assess work ability after 5 years measured with the WAS. Second, we assessed whether physician-diagnosed asthma modified the effect estimates.

From 2013 to 2018, the WAS in 478 of the 4423 participants (11%) changed from normal to reduced. The mean WAS was 9.45 (SD, 0.77) in 2013 and 8.99 (1.53) in 2018. Table 1 shows baseline characteristics of the study population. The study population comprised a majority of women (57%), and approximately half (52%) the participants were in the oldest age category of 41–50 years. The mean participant age in 2013 was 38.7 years with a standard error of 8.8. Regarding smoking status, 11% of participants smoked daily, 9% smoked occasionally, 23% were past smokers, and 57% were never-smokers. Nine percent of the participants had physician-diagnosed asthma.

Table 2 shows the unadjusted and adjusted associations between reduced WAS in 2018 and exposure to VGDF in 2013 as well as the collapsed exposure groups. Ever exposure to VGDF was associated with reduced WAS in 2018 (OR, 1.24; 95% CI, 1.01–1.55). Weekly and daily exposure to VGDF in the last 5 years in 2013 was also associated with reduced WAS in 2018 (OR, 1.43; 95% CI, 1.02–2.02 and OR, 1.88; 95% CI, 1.42 –2.50, respectively). Exposure to irritants was associated with reduced WAS (OR, 1.46; 95% CI, 1.15–1.86). No statistically significant interactions were observed between asthma and exposure (data not shown).

Table 3 shows the different specific exposures derived from the N-JEM, with reduced WAS in 2018 as the outcome variable. In the present study, exposure to wood and paper dust (OR, 1.81; 95% CI, 1.04–3.14) and exposure to mixed agricultural agents (OR, 2.24; 95% CI, 1.00–4.97) were associated with reduced WAS. An increased estimate was observed for exposure to welding fumes (OR, 2.05; 95% CI, 0.94–4.48), but did not reach statistical significance. Other exposures assessed in 2013 were not associated with reduced WAS in 2018. No statistically significant interactions were observed between asthma and exposure (data not shown).

In Table 4, the results related to the exposure to VGDF and main exposure groups are stratified by physician-diagnosed asthma. In the stratum showing results for those without asthma, weekly (OR, 1.47; 95% CI, 1.03–2.10) and daily exposure (OR, 1.92; 95% CI, 1.43–2.58) to VGDF, as well as exposure to irritants (OR, 1.41; 95% CI, 1.07–1.86) was associated with reduced WAS. For participants with asthma, ever exposure (OR, 2.35; 95% CI, 1.24–4.42) and daily exposure to VGDF (OR, 3.45; 95% CI, 1.56–7.61) was associated with reduced WAS. Exposure to high-molecular-weight chemicals and irritants resulted in higher, but not statistically significant, estimates for participants with asthma compared to those without. Interaction analyses showed no statistical significant interactions between asthma and inhalable exposures on WAS.

As a sensitivity analysis we assessed subjects with reduced WAS in 2013 (n=585) and found that age was an important predictor of WAS in 2018; older age was associated with continued reduced WAS, whereas younger age was associated with improved WAS in 2018. A statistically non-significant trend of individuals exposed to VGDF continuing to have reduced WAS in 2018 was also observed (data not shown). In the case of N-JEM and more specific exposures, we lacked the power to fit the regression models.

In the present study, self-reported exposure to VGFD was associated with reduced WAS, and there was a trend towards further reduction as exposure increased. Exposure to irritants was also associated with reduced WAS when the N-JEM data was analyzed. When assessing the specific exposure variables derived from the N-JEM, we found that exposure to wood and paper dust and mixed agricultural agents were associated with reduced WAS. There was an indicated association with exposure to welding fumes, but it did not reach statistical significance. The estimates were higher in subjects with asthma than in those without asthma but followed a similar pattern and interaction analyses showed no statistically significant results. This suggests that asthma does not modify the association between inhalable exposures and reduced WAS.

Previous studies have found that psychosocial factors, such as poor management and job satisfaction, and high physical demands, such as increased muscular work and poor ergonomic conditions, are work-related factors associated with reduced work ability (14). Results regarding the association between the type of work and level of work ability are conflicting (15). Several respiratory diseases, such as asthma, as well as respiratory symptoms, are associated with inhalable exposures (16, 17). Patients with these diseases also report reduced work ability, and this may indicate that inhalable occupational exposure could be a risk factor for reduced work ability. To our knowledge, no other studies have assessed on inhalable occupational exposure as a risk factor for reduced work ability by assessing specific exposures or by using the N-JEM or any others JEMs.

Other studies have shown that VGDF exposure is associated with an increased risk of respiratory disability (18). In a previous study from the Telemark study, carpenters and joiners, gardeners, welders, and agricultural workers were among the occupations with a higher risk of job change because exposures during work affected their breathing (20). In addition, exposure to irritants has been described as a risk factor for developing asthma (21). Further, occupational peak exposure to gases in the pulp mill industry has been reported to be a potent factor responsible for job change and respiratory work disability (22). These findings are consistent with our findings, indicating that VGDF exposure may affect work ability.

In our study, exposure to wood and paper dust was associated with reduced WAS assessed 5 years later. A large majority of the 104 participants in this exposure category were carpenters, builders, or other construction workers (90%), and only two participants had a job title associated with exposure to paper dust. In another Norwegian study using the same question to measure work ability, 12% of the total population reported reduced work ability (1). Among construction workers, 16–18% reported reduced work ability. These results indicate that construction workers may be at a higher risk of developing reduced work ability than other workers. In a large Dutch cross-sectional study of construction workers, researchers concluded that highly physically demanding jobs and psychosocial and physical work-related factors were the most important factors associated with work ability (23). They also reported a negative association between pulmonary problems and work ability, but when the indicators were included in the regression models, the explained variance increased by only 0.2%. The limited influence of respiratory problems was thought to be caused by the healthy worker effect, as the study population did not include participants receiving long-term disability benefits or those who had changed jobs because of health problems, leading to a likely underestimation of the effects of respiratory problems on work ability.

In the present study, increased exposure to VGDF was associated with reduced WAS in both individuals without and with asthma. This indicates that this exposure may affect all workers. The estimated ORs were higher in the participants with asthma than in those without asthma; however, not all effect estimates were statistically significant, as one might have expected. This may be because the group of participants with asthma comprised those with all types of asthma, including those with childhood asthma without symptoms in adulthood, and not just those with occupational asthma. This and the relatively small number of exposed subjects with asthma may have contributed to the lack of statistically significant associations between WAS and more well-known exposures, such as HMW agents, in subjects with asthma.

The measurement of work ability is complex. The complete WAI battery is a comprehensive instrument that is inconvenient for large surveys; therefore, shorter questionnaires have been developed (5). The WAS have been shown to effectively represent the entire WAI (4). In a study comparing the WAI to other instruments, the WAS exhibited poorer discriminating properties for detecting the absence of long-term sickness and correctly identified 74% of non-manual laborers and 72% of manual laborers, while the complete WAI battery identified 79% and 82%, respectively (5). Based on two large samples and on using other questionnaires, such as the WAI, Ebener and Hasselhorn concluded that the WAS could be used in occupational health research surveys (24).

The WAS can also be influenced by several other factors, both occupational and non-occupational; however, our questionnaire did not contain all possible risk factors that were previously found to be associated with reduced WAS (14). We did have data on body mass index (BMI) and regular physical activity during leisure time, but including these variables in the regression models had minimal effects on the estimates and associations (data not shown). Psychosocial factors, such as a lack of autonomy at work and high mental work demands, have been shown to influence work ability; however, we did not have information on these factors (14).

Work ability is not a static measure; those with reduced WAS in 2013 could have increased their work ability by 2018 and then be considered to have normal work ability. In the current study, 10% of individuals with reduced WAS in 2013 had a normal WAS in 2018. We found that age was an important predictor of WAS in 2018, and a statistically non-significant trend showing that individuals who were exposed to VGDF continued to have reduced WAS in 2018. This might indicate that cessation of exposure to inhalable agents improves work ability and that younger age is a positive predictor of reduced WAS. As an alternative approach, we analyzed WAS as a continuous score and calculated the change in WAS between 2018 and 2013 (See Tables, Supplemental Digital Content 1–3, http://links.lww.com/JOM/B779). We compared the group with a negative change to those with no change or a positive change. The results of these analyses showed somewhat lower effect estimates but a similar pattern, except that statistically significant results were not observed in the case of exposure to wood and paper dust (OR, 1.14; 95% CI, 0.78–1.67) but were observed in the case of exposure to cleaning chemicals (OR, 1.63; 95% CI, 1.08–2.45). These analyses showed that participants with reduced WAS at baseline experienced less subsequent reduction in work ability than those with normal WAS. Exposure had a greater effect in individuals with normal WAS and asthma than in those with reduced WAS and asthma, possibly because the latter group was more likely to benefit from interventions, such as being provided a disability pension or reduction of exposure.