Association between occupational risk factors and lung function among small-scale gold miners in Mongolia: a cross-sectional study

Abstract

Objectives

This study aimed to assess the association between occupational risk factors and lung function among gold miners in a northern Mongolian province.

Methods

A cross-sectional study was conducted among 143 gold miners in July 2024. Data were collected through structured interviews, physical examinations, chest X-rays, and spirometry. Occupational exposures included workplace type (administrative, factory, or underground), dust exposure level (low, moderate, high, based on work duration and underground working hours), and N95 mask use. Lung function outcomes were assessed using forced expiratory volume in 1 second (FEV₁), evaluated both both as an absolute value (liters) and as impaired lung function, defined by FEV₁ <80% of predicted. Multivariable logistic regression was used to examine associations with impaired lung function, and linear regression was applied to assess determinants of absolute FEV₁, adjusting for age, sex, and smoking status.

Results

Among 143 workers, mean FEV₁ was 3.39 L; and 23.8% had FEV₁ <80% of predicted. No occupational exposure was significantly associated with impaired lung function (FEV₁ <80%). However, underground work was significantly associated with reduced absolute FEV₁ (β = −0.22 L; p=0.026). N95 mask use was not significantly associated with impaired lung function after adjustment.

Conclusions

Early lung function decline was not captured by the standard threshold. Therefore, monitoring absolute FEV₁ may offer a more sensitive measure of early pulmonary changes among miners. Regular health monitoring and enhanced protective measures are urgently needed to prevent occupational lung diseases in resource-limited mining settings, including Mongolia and other developing countries.

Introduction

The modern mining industry in Mongolia officially began on December 25, 1922, with the operation of the Nalaikh coal deposit, the country’s first official mine. A major turning point occurred in the 1990s, when Mongolia transitioned from socialism to democracy and a free-market economy, marking a new era of rapid mining development¹⁾. Over the past three decades, gold mining has become one of the country’s most important industries, contributing significantly to foreign exchange reserves and export revenues. Mongolia ranks first in the world in terms of gold reserves and production. Since 1992, the gold production and mining in Mongolia has dramatically increased and the gold purchase by the central bank grew 10 times in the last decade²⁾.

As of 2022, Mongolia had 554 valid gold mining licenses, with 462 for placer mines and 92 for hard rock deposits. A significant portion of Mongolia’s gold production comes from artisanal and small-scale miners, with an estimated 40,000 to 60,000 people engaged in this activity across 14 out of 21 provinces³⁾. However, this sector presents significant occupational health challenges. Miners in these settings are exposed to high levels of dust, gas, and other hazardous substances, often under rotating shifts and physically demanding workloads^4,5,6). They often operate in dangerous conditions with limited use of personal protective equipment (PPE), poor access to healthcare, and insufficient insurance coverage^7,8). Even when healthcare is technically available, many miners perceive travel to distant clinics as a waste of time. Moreover, small-scale companies rarely conduct regular health checkups for their workers.

Chronic exposure to dust can impair respiratory function and lead to long-term health conditions, including chronic obstructive pulmonary disease, lung cancer, and pneumoconiosis^4,9). Pneumoconiosis accounts for 53.8% of all occupational disability cases¹⁰⁾. In Mongolia, between 1980 and 2006, a study found that pneumoconiosis reduced miners’ quality of life by 45%¹¹⁾. These findings underscore the importance of regular monitoring of lung function to detect abnormalities early and prevent further complications.

Lung function is known to decline with age and varies by anthropometric characteristics, such as age, height, and weight¹²⁾. However, for mining workers, lung function may decline more rapidly due to dust exposures, beyond the normal decrease associated with aging and body composition. Spirometry is widely used to assess pulmonary function. Yet, even when results are near the lower limit of normal, individuals are often classified as “normal,” potentially masking early signs of respiratory impairment¹³⁾. Therefore, it is crucial to regularly monitor the trajectory of lung function over time, not merely to check whether levels have crossed into the abnormal range, especially among mining workers. Several international studies have examined the relationship between dusty workplaces and lung function among small-scale gold miners, particularly in low-income countries^14,15,16). However, no such studies are currently available in Mongolia. Given the country’s unique mining practices, environmental conditions, and healthcare context, exposure levels and lung function outcomes may differ significantly from those observed elsewhere. Therefore, the aim of this study is to assess the correlation between occupational risks, in particular work type, use of PPE, and dust exposure, and pulmonary function among small-scale gold miners in Mongolia. Our findings will provide much-needed localized data to inform occupational health guidelines and contribute to global understanding of lung health in artisanal mining — filling a critical knowledge gap in Central Asia.

Methods

Study setting, design, and subjects

This cross-sectional study was conducted in July 2024 as part of The Mongolian Miners’ Health Study: An International Collaborative Initiative at a gold mining company located in Mongolia’s northern province. Mongolia’s known gold deposits fall into three economic categories: veins, mineralized zones, and placers. The mine involved in this study is a high-grade, narrow-vein gold operation, with gold content in quartz veins ranging from 0.1 g/ton to 746 g/ton¹⁾.

A total of 477 workers were employed at the mining site at the time of study. Of these, 147 workers (30.9%) were on duty during the shift when the research team conducted interviews and health examinations. Among them, 143 workers with complete data on lung function tests and work history were included in the final analysis, yielding a response rate of 97.3%. Due to time constraints and logistical challenges, including the remote location of the mining site, which made repeated visits during subsequent shifts impractical, the study was restricted to a single work shift. Consequently, participants were recruited using a convenience sampling approach.

Data collection and measurement

Data were collected through structured interviews using a pre-administered questionnaire, physical examination conducted by pulmonologists, medical imaging, and lung function testing via spirometry. Due to time and logistical constraints, we conducted a cross-sectional survey among workers present during a single scheduled shift at the gold mining site.

The questionnaire consisted of 43 items, organized into six main domains: (1) general information, including demographic information; (2) occupational risk; (3) injury-related information; (4) health-related quality of life; (5) lifestyle factors; and (6) additional information. The physical examination included a thorough respiratory assessment, with anamnesis conducted to collect participants’ current complaints. Chest X-rays were performed for all participants to evaluate structural changes in the lungs, while spirometry tests were conducted to assess lung function.

In the current analysis, we extracted data on demographic characteristics, occupational risk factors, and lung function. Occupational risk factors served as exposure variables and included work duration, daily working hours, workplace type, and the use of PPE. Workplace type was categorized by comparing participants working in factory and underground (undermining) environments to those in administrative roles. Work duration was categorized into two groups: up to 2 years or more than 2 years. Exposure intensity was defined by underground work hours and categorized as up to 6.0 hours or more than 6.0 hours per day. Based on the combination of work duration and intensity, participants were further grouped into three levels of dust exposure: low, medium, and high (Figure 1). PPE usage was defined strictly as the use of N95 dust-proof masks; other types of masks, including gas masks, surgical masks, cotton fabric masks, neck gaiters, and neck cloths, were excluded from this definition.

Fig. 1.

Framework of study methods

Demographic variables and smoking status were extracted as confounding variables. Participants who had smoked within the past 10 years were classified as “smokers,” while those who had never smoked or had quit at least 6 months prior were classified as “non-smokers.”

The lung function was assessed using spirometry, a widely accepted method for diagnosing and monitoring lung function. We used the Medikro-Pro, a computer-based spirometer compliant with the American Thoracic Society (ATS) Standards (2019). The spirometry procedure adhered strictly to ATS guidelines, including patient preparation, assessment, and interpretation of results^17,18). During spirometry, participants were seated, with feet flat on the floor in a relaxed environment. The spirometry software required the following inputs: participant identification number, name, age, sex, ethnicity (categorized as “Chinese” in the system), address, height, weight, smoking status, and relevant medical history. Participants were informed of the test purpose and given standardized instructions. A disposable mouthpiece flowmeter and a nose clip were used. Each participant performed up to three spirometry attempts unless a valid result was obtained in the first attempt.

We assessed lung function using two FEV₁ measurements: absolute FEV₁ in liters (L), representing the volume of air forcefully exhaled in 1 second; and FEV₁ categorized as normal or abnormal based on the percentage of the predicted value, with FEV₁ considered normal if it exceeded 80% of the predicted value¹⁹⁾.

Statistical analysis

First, study participants were descriptively analyzed by demographic characteristics and occupational risk factors. Mean FEV₁ (in liters) and the proportion of participants with normal FEV₁ values were reported across these variables.

To examine the relationship between demographic and occupational factors and lung function, two statistical approaches were applied: multivariable logistic regression and linear regression.

Multivariable logistic regression was used to assess the association between occupational and demographic factors and the likelihood of having abnormal lung function, defined as FEV₁ <80% of the predicted value. The dependent variable in this model was binary: 0 = normal FEV₁ (≥80%) and 1 = abnormal FEV₁ (<80%). Results were presented as odds ratios (ORs) with 95% confidence intervals (CIs).

Linear regression analysis was conducted to explore the relationship between occupational and demographic variables and absolute FEV₁ values (in liters), providing a continuous measure of lung function. Results from the linear regression were expressed as regression coefficients with 95% CIs, quantifying the associations between predictors and FEV₁.

For both logistic and linear regression analyses, three models were constructed: i) model 1 with only occupational exposure variables (eg, dust exposure level, PPE use, and workplace type); and ii) model 2 with occupational exposure variables plus demographic factors (eg, age and sex) and smoking status. A p-value <0.05 was considered statistically significant for all analyses.

To assess the robustness of the models, variance inflation factors (VIFs) were calculated to detect multicollinearity among the predictor variables. No significant multicollinearity was observed.

All statistical analyses were performed using SPSS version 26.0 (IBM Corp, Armonk, NY, USA).

Ethical approval

Ethical approval for this study was obtained from the Ethics Committee of Medical Research of the Mongolian National University of Medical Sciences (2022/3-06) and the Ethics Committee of the University of Occupational and Environmental Health, Japan (ER24-005). Participation in this study was voluntary and non-coercive. Workers were informed that refusal to participate would not result in any penalty, disadvantage, or change in employment conditions. All participants provided written informed consent after reviewing the study details.

Results

Table 1 presents the general characteristics of the study population and their lung function, as measured by both absolute FEV₁ (liters) and categorized FEV₁ (normal or abnormal, based on the 80% threshold).

Table 1. General characteristics of the study participants.

	N (column %)	FEV1, L	FEV1 <80%
		Mean (SD)	N (row %)
Total	143 (100.0)	3.39 (0.74)	34 (23.8)
Sex
Male	124 (86.7)	3.52 (0.68)	27 (21.8)
Female	19 (13.3)	2.55 (0.53)	7 (36.8)
Age, years
20–29	57 (39.8)	3.67 (0.72)	11 (19.3)
30–39	56 (39.2)	3.29 (0.71)	16 (28.6)
40–59	30 (20.9)	3.05 (0.67)	7 (23.3)
Smoking
Never/past	35 (24.5)	2.98 (0.81)	11 (31.4)
Current	108 (75.5)	3.52 (0.66)	23 (21.3)
Working duration, years
≤2	97 (67.8)	3.27 (0.72)	26 (26.8)
>2	46 (32.2)	3.64 (0.75)	8 (17.4)
Working hours, hours per day
≤6	73 (51.0)	3.28 (0.71)	19 (26.0)
>6	70 (48.9)	3.50 (0.76)	15 (21.4)
Worksite
Administration	40 (27.9)	3.27 (0.86)	9 (22.5)
Factory	33 (23.1)	3.45 (0.65)	6 (18.2)
Underground	70 (48.9)	3.42 (0.71)	19 (27.1)
Use of N95 respirators
Yes	109 (76.2)	3.43 (0.68)	21 (19.4)
No	34 (23.8)	3.27 (0.90)	13 (36.1)
Dust exposure
Low	53 (37.1)	3.25 (0.65)	14 (26.4)
Moderate	59 (41.3)	3.25 (0.77)	14 (23.7)
High	31 (21.7)	3.25 (0.78)	6 (19.4)

FEV₁, forced expiratory volume in 1 second; SD, standard deviation.

Most participants were male (86.7%), and males had a higher mean FEV₁ (mean 3.52; standard deviation [SD], 0.68 L) compared to females (mean 2.55; SD, 0.53 L). The proportion of participants with impaired lung function (FEV₁ <80% of predicted) was also higher among females (36.8%) than males (21.8%).

The youngest age group (20–29 years), comprising 39.8% of the sample, had the highest mean FEV₁ at 3.67 (SD, 0.72) L, with 19.3% of individuals in this group exhibiting FEV₁ <80%. In contrast, the oldest group (40–59 years, 20.9% of the sample) had the lowest mean FEV₁ (3.05; SD, 0.67 L), although the proportion with FEV₁ <80% (23.3%) was slightly lower than in the 30–39 age group.

Regarding occupational risk factors, 76.2% of participants reported wearing N95 masks. These individuals had a higher mean FEV₁ (3.43; SD, 0.68 L) compared to those who did not use N95 masks (mean 3.27; SD, 0.90 L). Similarly, the prevalence of impaired lung function was lower among mask users (19.4%) than non-users (36.1%).

In terms of dust exposure, 37.1% of workers were categorized as having low exposure, 41.3% as moderate, and 21.7% as high exposure. The three groups had similar mean FEV₁ values, but the percentage of participants with FEV₁ <80% varied: 26.4% in the low exposure group, 23.7% in the moderate group, and 19.4% in the high exposure group.

Nearly half of the participants (48.9%) were underground workers. Factory workers had a slightly higher mean FEV₁ (3.45 L; SD, 0.65 L) and a lower proportion of abnormal FEV₁ (18.2%) compared to those in other work environments.

Table 2 presents the association between occupational risk factors and impaired lung function, defined as FEV₁ <80% of the predicted value (categorized as abnormal).

Table 2. Logistic regression analysis of occupational risk factors associated with impaired lung function (FEV₁ <80% predicted).

	Univariate			Multivariate*
	OR	(95% CI)	P value	OR	(95% CI)	P value
Worksite
Administration	Reference			Reference
Factory	0.65	(0.24–1.74)	0.392	1.11	(0.30–4.04)	0.875
Underground	1.44	(0.66–3.13)	0.356	2.27	(0.71–7.29)	0.169
Use of N95 respirators
Yes	Reference			Reference
No	2.31	(1.01–5.31)	0.048	2.57	(0.98–6.69)	0.054
Dust exposure
Low	Reference			Reference
Moderate	1.00	(0.46–2.18)	0.991	0.51	(0.17–1.52)	0.227
High	0.72	(0.27–1.93)	0.515	0.51	(0.14–1.83)	0.299

CI, confidence interval; FEV₁, forced expiratory volume in 1 second; OR, odds ratio.

^*Adjusted for sex, age, smoking, worksite, dust exposure, and use of N95 respirators.

No statistically significant differences in the prevalence of impaired lung function were observed among workers across different workplace settings (administration, factory, and underground).

Participants who did not use N95 masks had 2.31 times higher odds of impaired lung function compared to those who used N95 masks (95% CI, 1.01–5.31). However, after adjusting for age, sex, and smoking status, the odds ratio increased slightly to 2.57 (95% CI, 0.98–6.69), though this association was no longer statistically significant.

In both the unadjusted and adjusted models, participants with moderate and high dust exposure did not show a statistically significant increase in the odds of impaired lung function compared to those in the low exposure group.

Table 3 shows the relationship between occupational risks and average FEV₁ in liters. In adjusted model, underground workers experienced a significant reduction of −0.22 L (95% CI, −0.62 to −0.04) compared to administrative workers. For use of N95 masks, participants who did not wear masks had a reduction in FEV₁ compared to those who used masks in both models, however, it did not show statistical significance. Compared to those categorized as low exposure to dust, FEV₁ increased by 0.01 (95% CI, −0.26 to 0.30) for moderate exposure and 0.09 (95% CI, −0.16 to 0.49) for high exposure; however, these differences were not significant.

Table 3. Linear regression analysis of occupational risk factors associated with absolute FEV₁ (liters).

	Univariate			Multivariate*
	Coefficient	(95% CI)	P value	Coefficient	(95% CI)	P value
Worksite
Administration	Reference			Reference
Factory	0.05	(−0.21 to 0.38)	0.563	−0.11	(−0.51 to 0.13)	0.238
Underground	0.05	(−0.18 to 0.31)	0.578	−0.22	(−0.62 to −0.04)	0.026
Use of N95 respirators
Yes	Reference			Reference
No	−0.09	(−0.44 to 0.12)	0.269	−0.03	(−0.32 to 0.21)	0.675
Dust exposure
Low	Reference			Reference
Moderate	0.01	(−0.24 to 0.26)	0.935	0.01	(−0.26 to 0.30)	0.879
High	0.17	(0.00–0.59)	0.047	0.09	(−0.16 to 0.49)	0.313

CI, confidence interval; FEV₁, forced expiratory volume in 1 second.

^*Adjusted for sex, age, smoking, worksite, dust exposure, and use of N95 respirators.

Discussion

This study examined the association between occupational risks such as work type, N95 respirator use, working duration and intensity and pulmonary function among 143 workers employed in a gold mining company in a Northern province of Mongolia. Pulmonary function was evaluated using two measures: the percentage of predicted FEV₁, where values below 80% are considered abnormal, and the absolute FEV₁ in liters. The findings did not reveal any association between the percentage of predicted FEV₁ and occupational risks. However, we observed a significant decline in absolute FEV₁ among underground miners compared to workers at other sites. This result highlights the potential for early, subtle changes in lung function among underground miners that may not yet meet the threshold for “abnormality” based on percentage-predicted FEV₁. Consequently, monitoring absolute FEV₁ may be a more sensitive approach to detecting early lung function deterioration in occupational settings.

In parallel, radiographic findings from same study population have been reported separately²⁰⁾, demonstrating a silicosis prevalence of 16.3% and significant associations with occupational exposures, particularly high dust exposure and engagement in blasting-related tasks. These findings provide important context for interpreting the present results. While chest X-rays capture structural lung abnormalities associated with silica exposure, spirometry reflects functional impairment. The discrepancy between the strong exposure–disease associations observed in radiographic outcomes and the weaker associations with FEV₁-based outcomes in this study suggests that structural changes may occur earlier or be more readily detectable than measurable functional decline.

The significance of this finding is supported by previous studies. Research has shown that miners with initially comparable mean FEV₁ levels to reference groups often experience higher rates of FEV₁ decline during their working careers, later reporting increased rates of respiratory symptoms and illnesses^19,21). Bhakta et al. underscored the limitations of evaluating lung function solely on percentage-predicted FEV₁, noting that this approach may overlook gradual declines in absolute FEV₁²²⁾. Additionally, follow-up studies among coal miners revealed that those with declining FEV₁ exhibited significantly greater airway resistance and air trapping, further linking these changes to long-term respiratory issues²³⁾. Community-based studies have also documented a relationship between lower FEV₁ levels and increased mortality risk^23,24). Collectively, these findings emphasize the importance of absolute FEV₁ as a critical measure for early detection of lung function changes, even before clinical symptoms arise.

Our study reinforces the necessity of regular health monitoring for miners, particularly those working underground, who are at heightened risk of gradual lung function decline. Unfortunately, routine follow-up is often lacking in Mongolia, where mid-sized mining companies typically do not employ occupational health professionals, and workers seek medical care only when symptoms significantly impact their daily lives. Health screenings are often limited to external projects rather than forming part of routine health protocols. This gap in regular monitoring presents a significant challenge for the early detection and prevention of occupational lung diseases. Our findings call for mining companies to prioritize routine health assessments and adopt proactive measures to safeguard the health and well-being of their workforce.

The study also examined the role of N95 respirator use in mitigating lung function decline. Although non-use of N95 respirators was associated with higher odds of impaired lung function (adjusted OR 2.57), this association did not reach statistical significance after adjustment. This finding contradicts with many studies showing that N95 respirators are significantly associated with preventing occupational lung disease including silicosis²⁵⁾. Therefore, the current findings should be interpreted cautiously, as the study may have been underpowered to detect this effect. The exposure distribution was highly imbalanced, with 76.2% of participants reporting N95 use, and the total number of abnormal lung function events was limited. Such exposure skewness increases standard errors, resulting in wider confidence intervals and higher p-values. Therefore, the absence of statistical significance does not necessarily indicate the absence of a clinically meaningful association²⁶⁾. The magnitude of the point estimate suggests a potentially important protective role of N95 respirators, supporting continued promotion of appropriate respiratory protection in mining settings. In addition, the lack of significance could stem from improper mask usage, poor fitness, misreporting, or low-quality masks, since the effectiveness of N95 respirators depends on proper fit, consistent usage, and regular maintenance²⁷⁾. Therefore, the quality of masks must be assured and improper use of mask needs to be checked. Globally, including in Mongolia, workers often face barriers to consistent respirator use, such as physical discomfort, limited awareness of dust exposure risks, and the absence of stringent PPE policies^28,29). To address these issues, mining companies should ensure the provision of high-quality masks and enforce strict PPE usage policies. Training programs that educate workers about dust exposure risks and proper respirator use are essential.

Regarding the dust exposure defined based on combination of working duration and intensity working in their own worksite, although the average FEV₁ appeared slightly higher in the medium and high dust-exposure groups compared to the low exposure group, this difference was not statistically significant. This unexpected trend may reflect random variation or possible selection factors, such as the healthy worker effect, where individuals with better baseline lung function are more likely to remain in or be assigned to more intensive work. However, given the lack of statistical significance, no definitive conclusion can be drawn, and further research with larger sample sizes and more precise exposure assessment is warranted.

This study has several strengths and limitations. One notable strength is the ability to collect data from a mid-sized mining company, despite the general reluctance of such companies to disclose workers’ health information due to concerns about their reputation. By engaging with the company, the study was able to highlight the importance of regular health monitoring, conveying a crucial message to company managers and health professionals about the benefits of consistent follow-ups in lung function rather than relying on abnormality of lung function via sporadic health checks. However, as a cross-sectional study, our findings are limited to a snapshot of the current situation and cannot establish causality between exposure and lung function outcomes. Additionally, the relatively small sample size of 143 workers may have reduced the study’s statistical power, limiting the precision of results and their generalizability to other mining populations. Future research with larger sample sizes is necessary to confirm these findings and improve their applicability to broader worker populations. In the future, there is a need for larger studies to confirm the findings and improve generalizability.

Conclusion

Our findings contribute to the growing body of evidence that questions the sole reliance on percentage-predicted FEV₁ as a measure of lung function. While this metric remains the standard for diagnosing pulmonary dysfunction, it may fail to capture the full spectrum of occupational risks faced by miners. Incorporating absolute FEV₁ measurements offers a more sensitive approach for detecting early changes in lung function. Furthermore, transitioning underground miners to less hazardous work environments based on early detection of lung function decline could help prevent occupational lung diseases. Mining companies, particularly in resource-limited settings like Mongolia, must take proactive measures to implement routine health monitoring and protective strategies to safeguard workers’ respiratory health.

Funding

This work was supported by the Grant for the Occupational Health for SDGs: Research Development Project in Asia, provided by the Japanese Society for Occupational Health.

Conflicts of interest

The authors declare no conflicts of interest.