*3.2. Use of Respiratory Protective Device (RPD)*

The majority of the 112 (97.4%) coffee workers did not use any type of respiratory protective devices (RPD). Among the non-users of RPD, 109 (97.3%) of the coffee workers indicated that the reason for not using RPD was because it was not available or not provided at the work place. Others reported that the reasons for not using RPD were because it was not comfortable (one worker) (0.9%), and that the worker experienced that the RPD did not protect from the dust (one worker) (0.9%).

#### *3.3. Chronic Respiratory Symptoms*

The prevalence of chronic respiratory symptoms was in the range of 5.2–55% and 2.7–12.7% among the coffee workers and controls respectively (Table 2). Six of the coffee workers (5.2%), and none of the controls had chronic bronchitis. The prevalence ratio of all of the respiratory symptoms was significantly higher for the coffee workers compared to the controls after adjusting for age, education, years worked in other dusty factories, and previous respiratory disease (Table 2). As the number of current smokers were few, the analysis was also performed after excluding the smokers. This did not change the results substantially.


**Table 2.** Prevalence of chronic respiratory symptoms among coffee workers and controls.

CI, confidence interval while adjusting for age; education, years worked in other dusty factories and previous respiratory disease; *p*-value when comparing coffee workers vs. controls; "n": Number of study participants.

#### *3.4. Lung Function*

Table 3 shows the result of lung function stratified age among the coffee workers and controls. The coffee workers in the age group 28–39 years and ≥40 years, had significantly lower FVC and FEV1 compared with the controls in the similar age category. The FEV1/FVC-ratio was significantly lower among the coffee workers compared to the controls in the oldest age group (Table 3). The prevalence of airflow limitation (FEV1/FVC < 0.7) was higher among the coffee workers compared to the controls in all of the age categories (Table 3). The analyses were also performed again without including education level in the model, and the results were quite similar (data not shown).


**Table 3.** Lung function among primary coffee processing factories and control groups stratified by age (three age groups).

Analysis of covariance between coffee worker and controls while adjusting for height and education level. \* Fisher exact test between coffee workers and controls; *p*-value: significance level; "n": Number of study participants.

#### **4. Discussion**

This study found a significantly higher prevalence of respiratory symptoms and lower lung function among the coffee workers compared to controls. Our results are consistent with studies conducted in primary coffee processing factories in Papua New Guinea, Uganda, and Tanzania [8,9,11]. All of these studies show that coffee workers have high prevalence of respiratory health problems. However, our present study found a higher prevalence of some of the respiratory symptoms compared with the studies conducted among Arabica coffee workers in Tanzania, where the prevalence of

breathlessness was 14%, wheezing 13%, and chronic bronchitis 3.1% [12]. One of the reasons for this difference could be the higher personal total dust exposure in Ethiopian coffee factories (GM = 12.3 mg/m3) compared with the comparable job groups in Tanzanian primary coffee factories (GM = 2.1 mg/m3) [20]. The different methods of coffee pre-processing could be another reason; the Arabica coffee is pre-processed only by a wet-method in the Tanzanian factories, whereas in Ethiopia, Arabica coffee is pre-processed either by dry or wet method, based on the individual farmers' interests. In addition, only 3% of the workers used RPD in Ethiopia, compared to 33% in the coffee workers in Tanzania. The lack of RPD use makes it more likely that the workers actually were exposed to the dust levels measured in the factories. There may have been other additional factors that were not identified in the present study that may also have influenced the respiratory health of the workers.

Similarly, the prevalence of a cough with sputum and wheezing in our study was higher than in the study in Uganda; where the prevalence was 5.2% and 13.5%, respectively [9]. This might be due to the different types of coffee species between the two countries; the Robusta and Arabic coffee were processed in Ugandian factories, whereas only Arabica coffee was processed in Ethiopia.

For a cough with sputum, we found a higher prevalence than reported in the study in Papua New Guinea (8.7%) [8]. This difference might be due to higher dust exposure in the present study compared with what was measured in Papua New Guinea (0.7–10 mg/m3). In addition, the difference in working environments, coffee processing methods, and level of awareness among the coffee workers about the impact of dust exposure on their health might be the reason for the difference in the symptom prevalence. In addition, there may be differences between these countries regarding the presence of, for instance, lung infections or sequela after lung infections. Infections may cause respiratory symptoms and influence lung function. This possibility is not very likely, as the examined workers are performing hard physical work, but this factor needs to be considered, because of the high prevalence of tuberculosis as well as HIV in East-African countries [21]. However, it is not likely that this type of health problem is different among the workers in the two factory types included in our study.

The present study's results showed that in the two oldest age groups of coffee workers, the FVC and FEV1 were lower than among the controls, while FEV1/FVC was lower in the oldest age group of coffee workers than among the controls. In the Tanzanian study, there were no difference in the FVC and FEV1 between coffee workers and controls [12]. The considerably higher dust exposure in the Ethiopian study (GM = 12.3 mg/m3) [13] compared to the levels reported for the processing of Arabica coffee in Tanzania (GM = 2.1 mg/m3) [20] may have contributed to the difference in findings related to lung function. Also, the study conducted in Papua New Guinea, where the dust levels were lower than in the present study, did not find significant differences in FVC and FEV1/FVC between the coffee workers and controls.

Both FEV1 and FVC were reduced among the coffee workers, indicating both obstructive and restrictive lung effects. However, in the oldest age group, the FEV1/FVC ratio was less than 0.70 for about 27% of the coffee workers, which indicates the presence of an obstructive lung disease [18]. It is noteworthy that such a result was found, even though these workers were present at the workplaces in physically demanding work.

No statistically significant difference was observed in the incidence of past respiratory diseases between the coffee workers and controls; unfortunately, we have no information about when these past respiratory diseases occurred. For example, they could have been before starting work in coffee factories or after starting working in the factories. We only asked the participants if they had ever had any previous respiratory disease.

The control group in the present study was from another production factory, with very low dust levels. Another possibility would have been to choose a control group from the general population, but this would have introduced other types of bias in the study, related to socioeconomic differences between the factory workers and the population.

This is the first study to assess the prevalence respiratory symptoms and lung function among coffee workers in Ethiopia. However, as this study is a cross sectional study, the cause and effect association between dust exposure, and respiratory symptoms and lung function reduction cannot be drawn based on our findings. Our analyses were adjusted for other factors, including age, which may affect lung function. However, it is worth noting that there may be other variables present, which we have not identified. We would therefore recommend that a longitudinal study is undertaken in order to characterize the association between dust exposure and lung function reduction.

This study used a questionnaire-based interview to assess the respiratory symptoms that might result in recall and interviewer bias. However, similar questions were used to assess the respiratory symptoms in both the coffee workers and control groups. To minimize bias, a well-trained interviewer was involved in interviewing both groups. Also, the reported symptoms agreed with the objective measures from the spirometry. Symptoms such as coughing, wheezing, and breathlessness are often associated with obstructive lung disease [22].

This study included workers from all three coffee growing and processing regions of Ethiopia, and the factories are considered to be representative to all similar primary coffee processing factories in Ethiopia. Similar results might also be expected by any African coffee factory with a similar dust exposure level and similar production type of Arabica coffee beans.

As this study showed that the majority of the coffee workers did not use any type of respiratory protective devices, an immediate action to reduce respiratory health problems among coffee workers would be to provide proper facemasks.

#### **5. Conclusions**

Workers in primary coffee processing factories in Ethiopia had a higher prevalence of chronic respiratory symptoms and lower lung function than the controls. This might represent early signs of lung disease. A longitudinal study on the possible effects of coffee dust on respiratory health among coffee production workers is recommended.

**Author Contributions:** S.W.A. planned the study, collected and analyzed data, and drafted the manuscript in consultation with other authors. M.B., W.D., A.K., and B.E.M. participated in the design, collection and analysis of data, and provided scientific support throughout the project and commented on the manuscript. All of authors have read and approved the final manuscript.

**Funding:** This research was funded by the Norwegian Program for Capacity Building in Higher Education and Research for Development (NORHED).

**Acknowledgments:** We would like to thank the coffee factory management teams and workers for their permission to conduct the study and for participating in the study.

**Conflicts of Interest:** The authors declare that they have no competing interest.

#### **References**


© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

International Journal of *Environmental Research and Public Health*
