Tungu, Alexander Mtemi MD; Bråtveit, Magne PhD; Mamuya, Simon H. PhD; Moen, Bente E. PhD
Previous studies have suggested a relationship between dust exposure and adverse respiratory health effects among workers in cement factories.1–5 A multicenter cross-sectional study among European cement workers found a higher prevalence of chronic respiratory symptoms among cement workers compared with controls.4 A higher prevalence of chronic respiratory symptoms and chronic obstructive pulmonary disease (COPD) were found among Tanzanian cement production workers compared with maintenance and administrative controls.5 Few follow-up studies on respiratory health and dust exposure exist in cement factories.6,7 A prospective follow-up study among highly exposed Ethiopian cement workers reported an increase in prevalence of chronic cough (with or without phlegm) and dyspnoea, as well as reduction of lung function (forced expiratory volume in 1 second [FEV1], and the ratio of FEV1 to forced vital capacity [FEV1/FVC]) among the exposed after only 1 year, but not among controls.8
Some studies conducted in developed countries found no association between dust exposure and adverse chronic health effects.9–12 This could be due to low dust-exposure levels as a result of more efficient dust-control measures in these factories. Nevertheless, none of the previous studies in cement factories have assessed the impact of reducing personal dust exposure levels on the prevalence of chronic respiratory symptoms, lung function, and/or COPD. A follow-up study among Austrian workers exposed to quartz demonstrated a slowing of the annual decrease in FEV1 and FVC after implementation of a Threshold Limit Value (TLV) of 10 mg/m3 for inhalable dust.13 Nevertheless, the Austrian study has no personal dust exposure levels and the type of exposure is different from that in this study.
Dust exposure among workers in cement factories occurs in all stages of cement production.14,15 High dust-exposure levels exceeding recommended international exposure levels16 have been reported in several studies.17,18
In a Tanzanian cement factory, the highest dust-exposure levels were found in the crane, packing, and crusher in 2002.18 Conditions related to high dust-exposure levels among workers were reportedly defective doors and windows of crane cabins, and lack of or poor ventilation systems in the packing areas and open flow lines.18 A reduction of dust-exposure levels among factory workers was recommended in this factory. As part of the recommended measures, maintenance and repair activities were carried out in the cement factory (Table 1). In addition, a new production line was established alongside the old cement factory and started production in early 2010.
In this study, we have used group-based data from two cross-sectional studies to evaluate the effect of the collective measures19,20 in reduction of exposure levels and the targeted respiratory health outcomes. We compared baseline data collected in 2002 and the data obtained after intervention (8 years after the baseline study) in the same cement factory. We aimed at comparing total dust-exposure levels, prevalence of chronic respiratory symptoms, lung function, and COPD among Tanzanian cement production workers in 2002 and 2010–2011.
Study Design and Setting
This study compared data collected in 2002 with similar data obtained in 2010–2011 among workers in the production line (exposed) at a Portland cement factory in Tanzania. Because of the absence of individual data obtained in 2002, we compared summarized data between 2002 and 2010–2011 (an ecological analysis).
In 2002, all exposed workers (n = 120) were included in respiratory health examinations.5 The controls were maintenance and administrative workers (n = 107) in the same cement factory in 2002. The 2002 data on total dust exposure, chronic respiratory symptoms, lung function, and COPD among cement workers have been published previously.2,5,18
In 2010, a total of 210 out of 411 exposed workers were randomly selected and invited to participate in the study.14 Thirty-nine exposed workers did not participate giving a response rate of 82.4%. A total of 105 out of 349 production workers in the mineral water factory were randomly selected to form the control group in 2010. Seven controls did not participate giving a response rate of 93.3% in 2010. Data were collected (same variables as in 2002) from exposed workers and controls in 2010. In 2011, additional data for personal total dust exposure alone were collected for the same months (June to August) as in 2002.
All participants in both 2002 and 2010–2011 were males. The cement factory18 and mineral water factory14 were located in Dar es Salaam, Tanzania.
In both sampling periods (2002 and 2010–2011), personal total dust samples were collected on preweighed 37-mm cellulose acetate filters with a pore size of 0.8 μm in a closed faced three-piece Millipore-cassette connected to an SKC pump (Sidekick Casella; SKC Limited, Blandford Forum, United Kingdom) with a flow rate of 2.0 L/min. The analysis of total dust samples was performed using similar methods in both periods. Nevertheless, the analysis was performed at the X-lab AS laboratory in Bergen, Norway, in 200218 and at the Eurofins Product Testing in Denmark in 2010–2011, as the X-lab had been sold to the Eurofins laboratory.14
In 2002, a total of 79 dust samples were collected from 52 exposed workers,18 whereas among controls, 24 samples from 16 maintenance workers and 17 samples from administrative controls were collected in 2002.
In 2010–2011, a total of 192 dust samples were collected from 115 exposed workers, whereas 47 dust samples were collected from 43 mineral water factory controls. Nevertheless, 13 dust samples from the exposed workers and 3 dust samples from controls were not analyzed in 2010–2011, leaving a total of 179 and 44 samples for final analysis among the exposed workers and controls, respectively. Reasons for exclusion in the analysis were low pump flow rate for nine exposed workers and three controls. Four samples obtained from cleaning workers using vacuum cleaners were also excluded because these appliances were used in multiple sections of the cement factory.
Chronic respiratory symptoms among study participants were assessed using a modified British Medical Research Council questionnaire21 in both examination periods. The questionnaire was translated from English to Swahili and back to English, using two independent translators. Questions used in the assessment of chronic respiratory symptoms are shown in Table 2. The questionnaire also comprised sociodemographic data, occupational history, past chest illnesses, use of respiratory protective equipment (RPE), and smoking habits. Current smokers or those who had stopped smoking less than 1 year ago were categorized as smokers, whereas never-smokers or those who had stopped smoking more than 1 year ago (ex-smokers) were categorized as nonsmokers. The pack years of smoking were calculated as the number of cigarettes per year divided by 20.
In 2002, participants self-administered the questions and returned the completed questionnaire to the investigator on the next day. In 2010, the first author administered the questionnaire by interviewing each participant.
Lung function tests were performed in accordance with the American Thoracic Society recommendations for spirometry22 in both examination periods. In 2002, a Vitalograph Spirometer was used,2 and in 2010 a digital Spirare Spirometer (SPS310) was used. The Vitalograph spirometer was calibrated daily using a 2L syringe, whereas no calibrations were required for the Spirare Spirometer. Lung function examinations were conducted between 10:00 AM and 12:30 PM in 2002 and between 14:00 PM and 16:00 PM in 2010. The lung function examinations were performed in a sitting position without a nasal clip in both periods. Participants performed three to eight spirometric tests.
For comparison purposes, the choice of the lung function indices in 2010 was based on the indices examined in 2002. The lung function indices included FVC, FEV1, percentage predicted FVC (FVC%), percentage predicted FEV1 (FEV1%), and FEV1/FVC ratio. The maximum values for FVC and FEV1 were used in the statistical analysis.
Predicted values for FVC and FEV1 in both periods were calculated on the basis of the regression equations for Tanzanian male adults as follows:
FVC(l) = 0.0604H − 0.016A − 6.14, and FEV1(l/s) = 0.046H − 0.022A − 3.864, where H is height (cm) and A is age (years).23
In 2010, nine spirograms (six unacceptable and three missing data) among exposed workers and five spirograms among controls were excluded, leaving 162 and 93 spirograms among the exposed workers and controls, respectively, for statistical analysis.
Diagnosis of COPD
In both examination periods, all participants with airflow limitation (FEV1/FVC ratio < 0.70) were considered to have COPD based on the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria.24
The study was approved by the Western Norway Regional Committee on Medical Research Ethics and the Muhimbili University of Health and Allied Sciences Research and Ethics Committee, in both 2002 and 2010–2011. Each study participant gave a written informed consent. No information about the individuals was available to the employers at any time during the study.
Total dust-exposure levels were skewed and were log-transformed before analysis. The geometric mean (GM) and geometric standard deviation for total dust exposure in 2010–2011 were calculated. The GM for total dust exposure from each section of the production line in the cement factory in 2002 was obtained from a previous publication.18 Statistical testing of the overall GM for total dust exposure between the two time periods was not possible because of missing geometric standard deviation in 2002. In 2010–2011, the dust exposure levels in the raw mill and kiln were combined because workers in these sections worked interchangeably; hence, they were not compared with the dust levels in 2002. An independent t test was used for comparison of continuous variables. A two-sample t test was used for comparison of summarized data between the two time periods. A chi-squared test was used to compare categorical variables.
Linear regression analysis was used to compare FVC, FEV1, and the FEV1/FVC ratio between the exposed workers and controls in the respective years, adjusting for age, height, duration of employment, and pack years. Additional linear regression analysis was used to compare FVC%, FEV1% while adjusting for pack years of smoking and duration of employment. A chi-squared test with a Breslow-Day test of homogeneity of the odds ratios (ORs) was used to compare the difference in risks (unadjusted OR) of chronic respiratory symptoms between 2002 and 2010.
Logistic regression analysis was used in the comparison of chronic respiratory symptoms between the exposed workers and controls within the same time period of examination while adjusting for age, duration of employment, education level, and pack years of smoking. Among the exposed workers, a subgroup analysis of chronic respiratory symptoms and lung function among workers who had less than 8 years at work (examined in 2010 only) and those with 8 years or more at work (examined in both 2002 and 2010) was performed. Statistical analyses were performed using Statistical Package of Social Sciences, version 19 (International Business Machines [IBM] SPSS Statistics, NY) for Windows and statistical significance was reached at 95% confidence interval (CI).
A total of 120 exposed workers and 107 controls participated in the study in 2002 (Table 3). In 2010, a total of 171 exposed workers and 98 controls participated in the study (Table 3). Age, duration of employment, and education level were lower among the exposed workers than among maintenance and administrative controls in 2002. These characteristics were similar among the exposed workers and controls in 2010, except for height where the exposed workers were significantly shorter than controls (P < 0.001) (Table 3).
Height and duration of employment among the exposed workers were significantly lower in 2010 than in 2002 (Table 3). The proportion of the exposed workers who reported using RPE was lower in 2002 than in 2010 (38% vs 91%). Age, education, pack years of smoking, and weight among the exposed workers did not differ significantly between the two periods.
Among controls, age and duration of employment were significantly lower in 2010 than in 2002, whereas height and pack years of smoking did not differ significantly between the two periods.
In 2010, five exposed workers and four controls reported a history of tuberculosis but had been treated and declared cured. Six exposed workers and eight controls had a history of asthma.
The number of personal total dust-exposure levels obtained in the cement factory in 2002 and 2010–2011 were 79 and 179, respectively (Table 4).
In 2002, the overall GM for total dust exposure among exposed workers was 10.6 mg/m3. The proportion of dust levels above TLV of 10 mg/m3 for particles not otherwise specified (PNOS) in 2002 was 58.2%. The highest exposure levels in 2002 were found in the crane followed by packing and crusher (Table 4).
In 2010, the overall GM for total dust exposure was 5.8 10 mg/m3, and the proportion of the dust levels above the TLV16 value of 10 mg/m3 were 31% (Table 4). The highest exposure level was found in the packing, followed by crusher and cement mill.
Among the exposed workers, the overall dust exposure levels were significantly lower in 2010–2011 than in 2002. Similarly, the dust-exposure levels were consistently lower among the exposed job groups in 2010–2011 than in 2002. Nevertheless, there were significantly higher dust-exposure levels in the cement mill in 2010–2011 than in 2002 (Table 4).
Among controls, total dust exposure among maintenance workers (GM: 1.16 mg/m3, range 0.11 to 10.23) was higher than that among administrative workers (GM: 0.29 mg/m3, range: 0.01 to 2.36) and among mineral water factory controls (GM: 0.36 mg/m3, range: 0.05 to 1.8). Nevertheless, the dust exposure levels among controls did not differ significantly.
Chronic Respiratory Symptoms
In 2002, the exposed workers had a significantly higher prevalence of chronic respiratory symptoms than controls (Table 5). In 2010, work-related shortness of breath was the only symptom that was significantly higher among the exposed workers than among controls (OR: 3.3; 95% CI 1.3 to 8.3) while adjusting for age, duration of employment, educational level, and pack years of smoking.
Among the exposed workers, the prevalence of chronic cough, chronic sputum production, and chronic bronchitis were considerably lower in 2010 than in 2002 (Table 5). Among controls, there was a lower prevalence of chronic cough and chronic sputum production in 2010 compared with 2002.
The OR for developing chronic bronchitis among the exposed workers compared with controls was significantly lower in 2010 than in 2002 (P = 0.022) (Table 5). No significant difference in the ORs between 2002 and 2010 for the remaining symptoms was found.
In a subgroup analysis among the exposed workers, there was no significant difference in chronic respiratory symptoms between those who had worked less than 8 years and those who had worked for 8 years or more (Table 5).
In 2002, a total of 115 and 102 spirometric results among exposed workers and controls were analyzed, respectively. In 2010, the spirometric results analyzed were 162 and 93 among exposed workers and controls, respectively.
The exposed workers had significantly lower FVC, FEV1, FVC%, FEV1%, and FEV1/FVC ratio than controls in 2002 (Table 6). In 2010, there was no significant difference in FVC, FEV1, and FEV1/FVC ratio, whereas FVC% and FEV1% were significantly higher among the exposed workers than among controls.
Among the exposed workers, FEV1, FVC%, and FEV1% were lower in 2002 than in 2010, while FVC and the FEV1/FVC ratio did not differ significantly between the two periods (Table 6). The fraction of airflow limitation (FEV1/FVC < 0.7) was significantly higher among the exposed workers in 2002 than in 2010 (23% and 2.7%, respectively; P = 0.003). Among controls, there was no difference in lung function indices between the two periods except for FVC%, which was significantly lower in 2002 than in 2010 (P < 0.001).
A subgroup analysis for lung function indices revealed no significant differences between the exposed workers who had worked less than 8 years and those with 8 years or more in the cement factory (Table 6).
In 2002, there was a significantly higher prevalence of COPD among the exposed workers than among controls (Table 6). The prevalence of COPD did not differ significantly between the exposed workers and controls in 2010.
Among the exposed workers, the prevalence of COPD was higher in 2002 than in 2010 (Table 6). All the exposed workers who had COPD in 2010 had worked less than 8 years in the cement factory. Two of seven exposed workers (28.6%) who had COPD in 2010 were smokers.
The prevalence of COPD among controls did not differ significantly between the two periods (Table 6).
Total dust-exposure level in the cement factory was considerably lower in 2010—2011 than in 2002. The exposed workers had a considerably lower prevalence of chronic cough, chronic sputum production, chronic bronchitis, and COPD in 2010 than in 2002. The OR for chronic bronchitis among the exposed workers was significantly lower in 2010 whereas FEV1, FVC%, and FEV1% were significantly higher in 2010 than in 2002.
We observed an overall reduction in personal total dust-exposure levels in 2010–2011 compared with the dust levels reported previously in the same cement factory.18 The exposure level in 2010–2011 (GM: 5.8 mg/m3) was considerably lower than that reported among Ethiopian cement cleaners8 (GM: 549 mg/m3) but was still higher than those among Norwegian12 (GM: 2.3 mg/m3), German25 (GM: 3.0 mg/m3), and American10 (GM: 2.9 mg/m3) cement workers. In this study, the reduction of dust-exposure levels in the crusher may have resulted from the installation of an air conditioning system in the crusher control room in the old factory and the use of bag filters for dust suppression in the new production line. Nevertheless, attendants in this section still had 59% of the dust-exposure levels exceeding the limit value of 10 mg/m3 in 2010–2011, possibly because of the manual cleaning process.17 In the cranes, installation of air conditioning systems and repaired windows and doors in the old cement factory may possibly have reduced dust entry into the crane cabins, thereby reducing dust exposure among crane operators. In addition, instead of the old cranes, the new production line uses “stacker” machines that feed raw materials automatically into a conveyor belt system. The stacked materials are ready to be transported for further processing, and this may reduce dust-exposure levels among workers. In the packing area, installed local ventilation systems in the old factory and the use of bag filters in the new production line possibly resulted in reduced exposure among packing workers. Nevertheless, loaders (75%) in the packing section still had dust exposure levels higher than the limit value in 2010–2011. This may be due to poor ventilation when cement bags are loaded into trucks. The dust-exposure levels were higher in the cement mill in 2010–2011 compared with 2002. This might be explained by the fact that fewer (only 8) dust samples were collected among attendants in 2002 (compared with 38 dust samples in 2010–2011). The samples collected might not have provided a representative estimate of the exposure level for these workers.
The prevalence of chronic cough, chronic sputum production, dyspnoea, and chronic bronchitis among the exposed workers was lower in 2010 than in 2002. For chronic cough and chronic sputum production among the exposed workers in 2010, the prevalence was lower than that reported in a cross-sectional study involving less-exposed American workers from 16 cement factories.10 The latter finding can possibly be explained by differences in smoking habits between the two studies; the prevalence of smoking was higher in the American study than in this study (43.6% vs 19.8%).
In comparing the odds ratios, there was a significant reduction of chronic bronchitis among the exposed workers in 2010 compared with 2002, whereas for the other symptoms, the odds ratios were not reduced. The lack of a significant difference in the odds ratios between the two periods can be explained by two factors. First, although exposed workers had a lower prevalence of symptoms in 2010 than in 2002, controls also had a proportionately lower prevalence of symptoms in 2010, which may in turn result in nonsignificant changes in odds ratios between these two examination periods. Likewise, the control group in 2002 included maintenance workers who had higher dust exposure than administrative controls.18 Therefore, inclusion of the maintenance workers is likely to have increased the prevalence of respiratory symptoms among controls, thereby lowering the odds ratio between the exposed workers and controls. Thus, comparison of the possibly “diluted” odds ratios in 2002 might have led to lack of a true difference in the odds ratios between the two periods. Second, comparison of the odds ratios between 2002 and 2010 involved unadjusted estimates, which possibly underestimated our results. Nevertheless, this is unlikely because of similar age, smoking habits, and pack years of smoking among the exposed workers in the two examination periods.
There were significantly higher FEV1, FVC%, and FEV1% among the exposed workers in 2010 than in 2002. Although the lung function among cement workers was lower in 2002 than among controls, they generally had lung function similar to that of the controls in 2010. It is well established that highly exposed cement workers have decreased lung function indices.2,6–8,26 At lower dust-exposure levels, there was no significant difference in lung function indices among American10 and Norwegian9 cement workers compared with their respective controls, similar to what we found in 2010. In addition, we observed a considerably lower prevalence of COPD among the exposed workers in 2010 than in 2002. In 2010, the prevalence of COPD was also lower than that among Norwegian cement workers,9 possibly because of the older age and a larger number of smokers among the participants in the Norwegian study than in this study.
It is less likely that the lung function and COPD findings among the exposed workers in this study could be explained by differences in baseline characteristics between 2002 and 2010. The lower height among the exposed workers in 2010 than in 2002 might instead have caused underestimation of the difference in lung function indices. On the contrary, shorter duration of employment among the exposed workers in 2010 compared with 2002 might have caused an overestimation of our results. At the presently found dust-exposure levels, it is unlikely that the 2-year difference in duration of employment between the cement workers in 2002 and 2010 should fully explain the difference presently found in FEV1 of 430 ml/s. Other factors20 such as increased use of personal RPE in 2010 compared with 2002, and regular training on health and safety among cement factory workers might also have played a role.
A “healthy worker effect” might have been present in this study. All three participants who had COPD in 2010 had worked less than 8 years in the cement factory. This suggests that workers previously diagnosed with COPD5 were indeed not examined in 2010, possibly because the workers with such respiratory problems had quit their jobs27 or had changed positions in the cement factory. The healthy worker effect may cause an underestimation of the prevalence of COPD associated with long-term dust exposure in the cement factory.
To our knowledge, this is the first study to report on the association between reduction of occupational dust exposure, chronic respiratory symptoms, lung function, and COPD after implementation of dust-control measures in a cement factory in a developing country. This study enabled the evaluation of group-based interventions and hypothesis generation regarding the impact of reduction of dust exposure on respiratory health outcomes among cement workers.19,20,28 Our study was strengthened by assessment of current total dust exposure levels, by repeated measurements for total dust exposure, by comparison of exposure level before and after intervention in the same cement factory, by baseline characteristics being similar among exposed workers (except for height and tenure), and by presence of control groups in both examination periods.
The sample size in 2010 had a power of 90% to detect a difference in chronic cough at a significance level of 0.05, given a prevalence of 28.5% and 12.1% among exposed workers and controls, respectively. At low dust-exposure levels, previous studies with different sample sizes found no significant differences in the prevalence of chronic respiratory symptoms, COPD, and lung function between exposed workers and controls.9–11 One study had a relatively smaller sample size,9 whereas others had larger sample sizes10,11 than this study. This suggests that adverse respiratory health effects of dust exposure are relatively small when the exposure is low, consistent with our findings in 2010.
Nevertheless, this study was limited by a number of methodological aspects. Causal relationship between reduction of dust-exposure levels and the lower prevalence of chronic respiratory symptoms and COPD cannot be drawn in the current study.20 No adjustments for potential confounders were made during the comparison of variables between 2002 and 2010. Nevertheless, the potential confounders such as age, education level, height, and pack years of smoking between the exposed workers and controls were adjusted in each examination period, thus minimizing bias. In accordance with the National Tuberculosis and Leprosy Control Programme in Tanzania, a cough lasting for 2 weeks or more may be indicative of pulmonary tuberculosis after a regimen of broad spectrum antibiotics.29 Therefore, it is likely that workers with a persistent cough resulting from dust exposure might have been treated as tuberculosis cases and thus were included in the analysis in 2010. Nevertheless, a similar number of participants among the exposed workers and controls had a history of tuberculosis; thus, the impact on our results may be minimal. Controls obtained from the maintenance group in the cement factory in 2002 had relatively high dust-exposure levels, thereby biasing our results toward no difference. Nevertheless, these controls resided in the same geographical area as the administrative controls and the controls from the mineral water factory, thus minimizing bias resulting from exposure outside workplaces.
Another limitation of the study is that only total dust exposure was evaluated in relation to respiratory health effects. Other dust fractions such as the thoracic fraction that is presumed to be relevant to obstructive diseases related to dust exposure might have been evaluated.4 Nevertheless, the thoracic fraction lacks a limit value causing difficulties in interpretation and comparison of study findings. The use of total exposure in this study allowed comparisons with total dust-exposure levels measured previously in the same cement factory and with the TLV of 10 mg/m3 for PNOS.16
Furthermore, the dust samples were analyzed at different laboratories in 2002 and 2010–2011, but we assume that any interlaboratory differences27 did not explain the observed difference in dust exposure between the two periods, as the same analytic method was used. We also used a similar sampling technique in the two periods. Another difference was that the questionnaire for chronic symptoms was self-administered in 2002 whereas a questionnaire-guided interview was conducted in 2010, causing possible interviewer and response bias.27,30,31 Nevertheless, similar questions were used to assess chronic respiratory symptoms in both periods; all interviews in 2010 were conducted by one investigator (AM.T.), and the interviewer was unaware of the disease status of participants during the interviews, thus minimizing bias.
Lung function tests were performed using different devices and at different times of the day.32 The lung function indices might have been influenced by diurnal variations and by acute effects of dust exposure 33 because the measurements were conducted in the afternoon in 2010 and at noon in 2002. Nevertheless, studies on diurnal variations of lung function have reported conflicting results.34,35 In a study among patients referred for lung function tests, higher values of lung function indices were reported in the afternoon and the lowest were found at noon.34 On the contrary, a study among two Dutch populations reported high lung function indices at noon than in the afternoon.35 Therefore, we do not know how the diurnal variations influenced our results. Nevertheless, similar criteria for acceptability and reproducibility of spirometry22 were used in both examination periods.
Although we observed a substantial reduction in dust-exposure levels in 2010–2011 compared with 2002 in the cement factory, some working groups, such as attendants in the crusher and loaders in the packing areas, were exposed to dust levels above the limit value of 10 mg/m3 for PNOS.16 The high dust-exposure levels pose an increased risk for chronic respiratory health problems among these workers. Additional efforts to reduce dust-exposure levels and the use of efficient personal respiratory equipment are recommended in the cement factory.
There was a considerable reduction in total dust-exposure levels in the cement factory after intervention in 2010–2011 compared with 2002. We observed a lower prevalence of chronic respiratory symptoms and COPD, and higher lung function indices among cement workers in 2010 than in 2002.
We thank the cement and mineral water factory management teams for permission to conduct this study, the factory workers for participating in the study, Mr Elly Kapinda for conducting spirometry, Drs Valborg Baste, Gloria Sakwari, Akwilina Kayumba, and Aiwerasia Vera Ngowi Aiwerasia Vera Ngowi for their valuable contributions in the manuscript, and the Eurofins Laboratory in Denmark for analyzing the dust samples. Last, we thank the Norwegian State Loan Fund (Lånekassen) and the Norwegian Council of Universities Committee for Development Research and Education (NUFU)for funding this study.
1. Neghab M, Choobineh A. Work related respiratory symptoms and ventilatory disorders among employees of a cement industry in Shiraz, Iran. J Occp Health. 2007;49:273–278.
2. Mwaiselage J, Bråtveit M, Moen B, Mashalla Y. Cement dust exposure and ventilatory function impairment: an exposure–response study. J Occup Environ Med. 2004;46:658–667.
3. Noor H, Yap C, Zolkepli O, Faridah M. Effect of exposure to dust on lung function of cement factory workers. Med J Malays. 2000;55:5–7.
4. Nordby K-C, Fell AKM, Notø H, et al. Exposure to thoracic dust, airway symptoms, and lung function in cement production workers. Eur Respir J. 2011;38:1278–1286.
5. Mwaiselage J, Bråtveit M, Moen BE, Mashalla Y. Respiratory symptoms and chronic obstructive pulmonary disease among cement factory workers. Scand J Work Environ Health. 2005;31:316–323.
6. Saric M, Kalacic I, Holetic A. Follow-up of ventilatory lung function in a group of cement workers. Br J Ind Med. 1976;33:18–24.
7. Siracusa A, Forcina A, Volpi R, Mollichella E, Cicioni C, Fiordi T. An 11-year longitudinal study of the occupational dust exposure and lung function of polyvinyl chloride, cement and asbestos cement factory workers. Scand J Work Environ Health. 1988;14:181–188.
8. Zeleke Z, Moen B, Bråtveit M. Lung function reduction and chronic respiratory symptoms among workers in the cement industry: a follow up study. BMC Pulm Med. 2011;11:50.
9. Fell AKM, Thomassen TR, Kristensen P, Egeland T, Kongerud J. Respiratory symptoms and ventilatory function in workers exposed to Portland cement dust. J Occup Environ Med. 2003;45:1008–1014.
10. Abrons H, Petersen M, Sanderson W, Engelberg A, Harber P. Symptoms, ventilatory function, and environmental exposures in Portland cement workers. Br J Ind Med. 1988;45:368–375.
11. Vestbo J, Rasmussen FV. Long-term exposure to cement dust and later hospitalization due to respiratory disease. Int Arch Occup Environ Health. 1990;62:217–220.
12. Fell AKM, Notø H, Skogstad M, et al. A cross-shift study of lung function, exhaled nitric oxide and inflammatory markers in blood in Norwegian cement production workers. Occup Environ Med. 2011;68:799–805.
13. Hochgatterer K, Moshammer H, Haluza D. Dust is in the air: effects of occupational exposure to mineral dust on lung function in a 9-year study. Lung. 2013;191:257–263.
14. Tungu AM, Bråtveit M, Mamuya SD, Moen BE. Fractional exhaled nitric oxide among cement factory workers: a cross sectional study. Occup Environ Med. 2013;70:289–295.
15. Fairhurst S, Phillips A, Gilles C, Brown RH. Portland Cement Dust: Criteria Document for an Occupational Exposure Limit. London: Health and Safety Executive; 1997.
16. American Conference of Governmental and Industrial Hygienists. Threshold Limit Values for Chemical Substances and Physical Agents and Biological Exposure Indices. Cincinnati, OH: American Conference of Governmental and Industrial Hygienists; 2011.
17. Zeleke ZK, Moen BE, Bråtveit M. Excessive exposure to dust among cleaners in the Ethiopian cement industry. J Occup Environ Hyg. 2011;8:544–550.
18. Mwaiselage J, Bråtveit M, Moen B, Yost M. Variability in dust exposure in a cement factory in Tanzania. Ann Occup Hyg. 2005;49:511–519.
19. Blanc PD, Menezes AMB, Plana E, et al. Occupational exposures and COPD: an ecological analysis of international data. Eur Respir J. 2009;33:298–304.
20. Morgenstern H. Uses of ecologic analysis in epidemiologic research. Am J Public Health. 1982;72:1336–1344.
21. British Medical Research Council. Standardised questionnaire on respiratory symptoms. BMJ. 1960;2:1665.
22. American Thoracic Society. Standardization of Spirometry, 1994 Update. American Thoracic Society. Am J Respir Crit Care Med. 1995;152:1107–1136.
23. Mustafa KY. Spirometric lung function tests in normal men and African ethnic origin. Am Rev Respir Dis. 1977;116:209–213.
24. Pauwels RA, Buist AS, Calverley PMA, Jenkins CR, Hurd SS. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: NHLBI/WHO Global Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop summary. Am J Respir Crit Care Med. 2001;163:1256–1276.
25. Peters S, Thomassen Y, Fechter-Rink E, Kromhout H. Personal exposure to inhalable cement dust among construction workers. J Environ Monit. 2009;11:174–180.
26. Zeleke Z, Moen B, Bråtveit M. Cement dust exposure and acute lung function: a cross shift study. BMC Pulm Med. 2010;10:19.
27. Hennekens CH, Buring JE, eds. Epidemiology in Medicine. Philadelphia, PA: Lippincott Williams & Wilkins; 1987:133–162.
28. Bonita R, Beaglehole R, Kjellström T. Basic Epidemiology. Geneva, Switzerland: World Health Organization; 2006.
29. Ministry of Health and Social Welfare. Manual of the National Tuberculosis and Leprosy Programme in Tanzania. 5th ed. Dar es Salaam, Tanzania: Ministry of Health and Social Welfare; 2006:11–22.
30. Bergmann M, Jacobs E, Hoffmann K, Boeing H. Agreement of Self-Reported Medical History: comparison of an in-person interview with a self-administered questionnaire. Eur J Epidemiol. 2004;19:411–416.
31. Cook DJ, Guyatt GH, Juniper E, et al. Interviewer versus self-administered questionnaires in developing a disease-specific, health-related quality of life instrument for asthma. J Clin Epidemiol. 1993;46:529–534.
32. Künzli N, Ackermann-Liebrich U, Keller R, Perruchoud A, Schindler C. Variability of FVC and FEV1 due to technician, team, device and subject in an eight centre study: three quality control studies in SAPALDIA. Swiss Study on Air Pollution and Lung Disease in Adults. Eur Respir J. 1995;8:371–376.
33. Ali B, Ballal S, Albar A, Ahmed H. Post-shift changes in pulmonary function in a cement factory in eastern Saudi Arabia. Occup Med (Lond). 1998;48:519–522.
34. Medarov BI, Pavlov VA, Rossoff L. Diurnal variations in human pulmonary function. Int J Clin Exp Med. 2008;1:267–273.
35. Borsboom GJJM, van Pelt W, van Houwelingen HC, van Vianen BG, Schouten JP, Quanjer PH. Diurnal variation in lung function in subgroups from two Dutch Populations: consequences for longitudinal analysis. Am J Respir Crit Care Med. 1999;159:1163–1171.
Copyright © 2014 by the American College of Occupational and Environmental Medicine