Article Text

Download PDFPDF

The risk of hospitalisation for infectious pneumonia in mineral dust exposed industries
  1. Dong-Hee Koh1,
  2. Ki-Tae Moon2,
  3. Jeong-Youn Kim3,
  4. Seong-Weon Choe1
  1. 1Occupational Safety and Health Research Institute (OSHRI), Korea Safety and Health Agency (KOSHA), Bupyeong-gu, Incheon, Republic of Korea
  2. 2Research Center for Review and Assessment, Health Insurance Review & Assessment Service, Seocho-dong, Seocho-gu, Seoul, Republic of Korea
  3. 3Worker's Health Protection Division, Occupational Safety and Health Bureau, Ministry of Labor, Joongang-Dong, Gwacheon-city, Gyeonggi-do, Republic of Korea
  1. Correspondence to Dr Dong-Hee Koh, Occupational Safety and Health Research Institute (OSHRI), Korea Safety and Health Agency (KOSHA), #34-4, Gusan-dong, Bupyeong-gu, Incheon, 403-711, Republic of Korea; koh.donghee{at}


Objectives The purpose of the study was to elucidate the relationship between industries characterised by mineral dust exposure and infectious pneumonia.

Methods The authors applied a retrospective record linkage design and used the nationwide specific health examination database and identified industries where mineral dust exposure occurs and a control group composed of workers who had been exposed to noise-only during a 2000–2004 period. The database was matched with National Health Insurance claim records to identify pneumonia admissions from 2000 to 2005. The indirectly standardised admission ratios (SARs) for pneumonia admissions were estimated by comparing mineral dust exposed industry workers with noise-only exposed workers.

Results The authors found significantly elevated SARs in both men (1.54, 95% CI 1.13 to 2.05) and women (3.23, 95% CI 1.40 to 6.37) working in the cement, lime, plaster and plaster products industries, and only in men working in the cast-metals industry (foundry) (1.64, 95% CI 1.25 to 2.11).

Conclusions These results support the association between mineral dust exposure, as well as metal fumes, and infectious pneumonia.

  • Dust
  • pneumonia
  • silica
  • mineral dust
  • respiratory
  • inorganic dusts

Statistics from

What this paper adds

  • Metal fume exposure is regarded as a risk factor for infectious pneumonia.

  • The cement, lime, plaster, plaster products industry and cast-metals industry present increased risk of admission due to pneumonia, and these results support the potential association of exposure to mineral dusts, as well as metal fumes with infectious pneumonia.

Policy implication

  • These results might offer general implications with regard to environmental particulate matter.


Infectious pneumonia can occur as the result of employment in specific occupations. Abattoir workers, farmers and ranchers, healthcare workers, hunters and trappers, industrial plant workers, laboratory workers, manual labourers, military personnel, textile mill workers and veterinarians are employed in occupations related to specific bacteria, viruses and fungi.1 In industrial settings, outbreaks of disease due to Legionella pneumophila have been reported among workers exposed to contaminated aerosols.

Three important analyses of occupational mortality in England and Wales investigated the risk of pneumonia in welders from 1959 to 1972.2 Increased mortality due to pneumonia suggested that welding fumes reversibly increased the susceptibility of the lungs to pulmonary infections. In this study, workers who were definitely exposed to fumes, such as welders, molders, coremakers, and furnacemen; those possibly exposed to fumes, such as foundry labourers, metalworkers, tin platers and galvanisers; as well as those exposed to metal dust showed increased standardised mortality ratios (SMRs) or proportionate mortality ratios (PMRs) with regard to pneumonia.3 Following these reports, the Norwegian Labor Inspection Authority (NLIA) issued a warning to Norwegian physicians about the potentially lethal risk associated with the inhalation of fumes from thermal metalwork.4

Similar mortality patterns in molders, coremakers and foundry workers suggest that mineral dust, particularly crystalline silica, in addition to metal fumes might increase the risk of infectious pneumonia. Foundry workers are exposed to crystalline silica and metal fumes, both of which can damage lung tissue. Mineral dusts such as crystalline silica and coal, and fumes of transition metals cause oxidative damage to lung tissue, and might increase susceptibility to infection.5 6 Therefore, we sought to elucidate the relationship of mineral dust exposure with infectious pneumonia.

Methods and materials

Data collection

In Korea, manufacturing workers exposed to specific hazardous agents such as benzene and noise must undergo particular health examinations on an annual basis.7 For example, foundry workers exposed to fume and high noise concurrently must undergo medical tests such as chest x-ray and pure tone audiometry. These periodic medical examinations are conducted by more than 120 occupational service institutes throughout Korea. The institutes send the specific health examination data to the Occupational Safety and Health Research Institute (OSHRI), which has been computerising these data since 2000.

This database provides information about workers such as residence registration number (a unique identification number assigned to all Koreans), date of medical examination, materials to which workers were exposed and information about the employer, including its standard industrial code. The database includes all workers employed in numerous companies throughout Korea who are legally required to participate in the specific health examination programme.

We identified two groups in the database: workers in industries where mineral dust exposure occurs and noise-only exposed workers (excluding workers in dust-exposed industries) using the Korean Standard Industrial Classification (KSIC). The KSIC was revised based on the International Standard Industrial Classification (ISIC), and we distinguished industry groups according to major sector categories.8 We used silica exposure as a proxy of industries where mineral dust exposure occurs, and these included manufacturers of glass and glass products; ceramic wares; cement, lime, plaster and plaster products; other non-metallic mineral products (eg, cutting, shaping and finishing of stone); basic iron and steel; basic precious and non-ferrous metals (eg, manufacture of smelting, refining and alloys of non-ferrous metals); and metal casts (foundry).

An unexposed comparison group was selected to reduce the potential for confounding variables, including those associated with healthy workers, owing to differences between workers and the general population. This control group consisted of Korean workers in all industries except mineral dust exposed industries, who had taken the special annual health examinations for noise-only from 2000 to 2004 (as did those in mineral dust exposed industries).

Definition of pneumonia and data matching

We used the database of National Health Insurance claims for the 2000–2005 periods and identified those involving more than 1 day of hospitalisation due to pneumonia; we matched the specific health examination database with these data using unique identifiers. National Health Insurance claim records include admission dates, and diagnostic codes for relevant disorders. The database for National Health Insurance claims is more than 99% complete because all Koreans are covered by National Health Insurance. Diagnoses were classified according to the Korea Classification of Diseases and Causes of Death, fourth edition, based on the tenth revision of the International Classification of Diseases (ICD-10).

Pneumonia was defined as viral (J12); bacterial (J13–J15); caused by other infectious agents such as fungi, Chlamydia, and parasites (J16–J17); or organism-unspecified pneumonia (J18). We used the major code assigned to the disease in the database to identify pneumonia cases that led to hospital admission.

Deaths were identified by the Korean National Statistical Office (KNSO), a registry estimated to cover more than 95% of deaths. KNSO records provide the date of death and the cause of death. Study subjects were matched to the KNSO database using a unique identifier for the 2000–2005 periods.

Methods of analysis

We followed up the database retrospectively as for cohort data for the 6 years from 1 January 2000 to 31 December 2005. The entry point into the database was the date of the first specific health examination in 2000–2004. Individuals were included if they were aged 20–69 years at this point of entry. The person-year (denominator) was calculated as the period between the date of the first specific health examination and the date of the first pneumonia admission or death. If the worker survived without admission due to pneumonia until the termination point of the study, person-years were calculated as from the date of the first medical examination to 31 December 2005.

In cases involving multiple admissions due to pneumonia, we used only the first admission for calculating the person-year, irrespective of the number of pneumonia-related admissions.

The indirectly standardised admission ratios (SARs) for pneumonia were estimated by comparing the observed number of admissions for workers in mineral-dust-exposed industries with the expected number of those for noise-only exposed workers. The expected numbers were calculated from the age-specific admission rates for 10-year age groups of noise-only exposed workers.

SAS macros using a Lexis diagram were used to tabulate the number of pneumonia admissions/person-years, and SMRs and exact Poisson CIs were calculated with the SAS programme.9 10


Person-years were calculated in a similar way to the cohort data. Person-years in glass and glass products manufacturers (35 582 men, 4372 women), ceramic ware manufacturers (10/14 657 men, 10 428 women), cement, lime, plaster and plaster products manufacturers (41 629 men, 3597 women), manufacturers of non-metallic mineral products (38 288 men, 7331 women), manufacturers of basic iron and steel (169 952 men, 4177 women), manufacturers of basic precious and non-ferrous metals (60 228 men, 2889 women), cast-metal manufacturers (46 538 men, 4445 women) and the control group: noise-only exposed workers (1 107 033 men, 400 609 women) are also shown in table 1.

Table 1

Comparisons of indirectly standardised admission ratios by industry for noise-only exposed workers

Both male and female workers in the cement, lime, plaster and plaster products industries showed significantly increased SARs (1.54, 95% CI 1.13 to 2.05 for men and 3.23, 95% CI 1.40 to 6.37 for women). We also found an increased risk of pneumonia admissions among men working in the cast-metals industry (foundry) (1.64, 95% CI 1.25 to 2.11).

Male workers in the glass, glass products, ceramic wares and other non-metallic mineral products industries and female workers in the basic iron and steel and cast metals industries showed an increased risk for pneumonia admission, but these differences were not significant.


The aim of this study was to investigate the relationship between industries where mineral dust exposure occurs and occurrence of infectious pneumonia. Almost all studies related to occupational exposure and pneumonia have been performed with workers exposed to metal fumes. A mortality study on shipyard workers in northeastern England reported that mortality due to pneumonia was increased in welders (SMR 814, 90% CI 100 to 314).11 In a recent nationwide mortality study in England and Wales, Palmer et al reported excessive mortality from infectious pneumonia.12 A case–control study of men aged 20–64 years admitted to 11 hospitals for pneumonia13 found that pneumonia was associated with reported occupational exposure to metal fumes during the previous year (adjusted OR 1.6, 95% CI 1.1 to 2.4). The risk was greatest for lobar pneumonia and for those with recent exposure to ferrous fumes (OR 2.3, 95% CI 1.2 to 4.3). A 10-year observational study of absences due to sickness among 36 welders and members of a control group found that the average annual duration of all respiratory diseases was 2.6 (p<0.01) times greater for welders than for controls, and the average annual duration of lower respiratory diseases was 4.6 (p<0.01) times greater for welders than for controls. In this study, four welders but no controls had absences attributed to pneumonia.14 However, a multicentre cohort study failed to show increased mortality from pneumonia in welders (SMR 63, 95% CI 32 to 109).15

Palmer et al suggested at least two pathogenic mechanisms with regard to welding fumes and infectious pneumonia.13 The first relates oxidative damage and direct toxicity to macrophages, and the second proposes that free iron serves as a nutrient for micro-organisms and promotes their growth. Welders are exposed to fumes containing various transition metals, and these transition metals increase oxygen free radicals.6 16 They also might cause lung disease such as asthma, chronic obstructive pulmonary disease (COPD), and interstitial lung diseases.17 18 One study described enhanced deterioration, independent of smoking, in the lung functioning of welders.19 In addition to metal fumes, mineral dusts such as crystalline silica and coal have also been known to promote reactive oxygen species.5 16 From a perspective that underscores the mechanism of oxidative stress, mineral dust can be assumed to act similarly to metal fumes as a cause of pneumonia. Inflammation and fibrosis of the lung might form an environment conducive to the growth of pathogens eventually. In a case report, three men who had worked as welders or grinders of steel castings in a foundry were diagnosed as having pneumonia caused by Acinetobacter calcoaceticus.20 Two of these individuals died, and forensic examinations revealed mixed-dust pneumoconiosis with iron particles in the lungs. These findings suggest that chronic exposure to such particles might increase susceptibility to infection, even those caused by this atypical organism.

There have been several studies on the relation between various occupational exposures and respiratory mortality including pneumonia. The Registrar General listed the 15 occupations with the highest standardised mortality ratios for respiratory diseases, including bronchitis and pneumonia but excluding tuberculosis and lung cancer.21 Sandblasters showed the highest SMR (733); followed by those who worked underground in tin and copper mines (725); potters, ware-makers, casters and finishers (252); men working with kilns and ovens (247); cotton strippers and grinders (213); iron-foundry furnace men and labourers (205); and coal boat loaders and dischargers (197). These results reflect a strong association between dust exposure and mortality due to non-malignant respiratory diseases. A mortality study reported that pneumonia was more prevalent in certain occupations, including potters, cotton strippers and grinders, glazers, polishers, buffers, moppers, iron-foundry furnace men and metal molders: men in these occupations were found to be at higher risk of mortality due to pneumonia, compared with their wives.22 The author of this study suggested that inhaling silica dust rendered the subjects especially susceptible to bronchitis and bronchopneumonia, even before definite silicosis developed. These results suggest that exposure to mineral dust as well as metal fumes is associated with occurrence of pneumonia.

There have been a few community-based studies on occupational exposure and pneumonia. A case–control study on risk factors for community-acquired pneumonia in adults admitted to hospitals found that occupational exposure to lint, coal, asbestos, stone or vegetable dust was significantly more prevalent in cases than in controls (28% vs 14%, p<0.001).23 The adjusted OR for dusty occupations was 2.49 (95% CI 1.15 to 1.93). However, a mail-in case–control study of 29 general practices in Nottingham, UK, involving those diagnosed by general practitioners as having pneumonia, showed that dusty occupations were associated with an elevated unadjusted OR of 1.71 (95% CI 0.96 to 3.04), but this figure did not reach statistical significance.24

In our study, the cast-metals industry (foundry) where metal fumes are exposed presented an increased risk of pneumonia admission. However, in the cast-metals industry, workers can also be exposed to crystalline silica dust. Workers in the cement, lime, plaster and plaster products industries also showed a significantly increased risk of pneumonia admission. In these industries, workers can be exposed to cement dust, limestone dust, crystalline silica and other various mineral dusts. These results support the potential relationship between mineral dust exposure and infectious pneumonia.

We used nationwide data for the analysis of events. Although the national database used in this study represented an extremely valuable resource for identifying occupational groups at increased risk of diseases, it also had some crucial limitations.25 First, we could not identify precisely to what workers were exposed. The quality of information on materials to which workers were exposed was not so good. Therefore, we used industry as a proxy of exposure. Workers in each industry were selected using standard industrial codes irrespective of the content of their exposure; thus, only a portion of these respondents were exposed to dusts, and the remainder worked under non- or low-exposure conditions. However, some workers could have been exposed to fumes, gases and chemicals simultaneously. Except for the cast metals and cement, lime, plaster and plaster product industries, no other industry showed significantly increased risks. Usually glass and glass product, basic iron and steel, and basic precious and non-ferrous metals industries were composed of large factories, and it is thought that the proportion of workers exposed to dusts in these industries is lower than that of cast metals or cement, lime, plaster and plaster product industries. These situations might have led to risk dilution and moved the association towards null.26 In addition, each worker was assigned to only one occupation during the study period, and we did not consider job change or modifications to exposure circumstances.

Second, we could not control strong confounding factors such as smoking and other coexisting lung diseases, such as COPD.27 Reduced forced expiratory volume in 1 s and low education are also regarded as associated factors. These confounding factors could play a crucial role, but because all the study subjects were manufacturing workers, we thought the effect was not so large as to alter the association tendency.

Third, the accuracy of diagnosis (outcome) only depended on disease codes of database. The admission rates do not accurately represent true pneumonia incidence rate because admission for pneumonia is usually decided by severity and comorbidity, such as COPD.27 But in case of admission, diagnosis would be made by chest x-ray, so the accuracy of pneumonia might be high enough to be considered significant.

When the observed number of pneumonia admissions was relatively large, the indirectly standardised admission ratios in women were generally greater than those in men. Two explanations are possible. First, because women have lower smoking rates, these results might reflect the confounding effect of the risk associated with smoking. Second, women might be more susceptible to respiratory exposure.

The species of pathogen associated with metal fumes was not clearly identified, but Streptococcus pneumoniae has been reportedly linked to exposure to these fumes.13 In our study, more than half of the cases were classified as organism-unspecified pneumonia, and so we did not calculate the pathogen-specific RR.

Workers in these industries were exposed to mineral dusts with relatively high content levels compared with general environment. Our study might offer general implications with regard to environmental particulate matter. But this study is just a record linkage study, so it is important to take notice of limitations and avoid overinterpretation.


The purpose of our study was to elucidate the relationship between industries characterised by mineral dust exposure and infectious pneumonia. We used the nationwide specific health examination database to compare mineral dust exposed industry workers to control group: noise-only exposed workers.

We found significantly increased indirectly standardised admission ratios (SARs) in men (1.54, 95% CI 1.13 to 2.05) and women (3.23, 95% CI 1.40 to 6.37) working in the cement, lime, plaster and plaster products industries, and in men (1.64, 95% CI 1.25 to 2.11) working in the cast-metals industry (foundries). These results support the potential association between occupational mineral dust exposure, as well as metal fume exposure, and hospitalisation for pneumonia.



  • Funding This study was supported by Occupational Safety and Health Research Institute, Korea Occupational Safety and Health Agency (KOSHA).

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.