Self‐reported disease symptoms of stone quarry workers exposed to silica dust in Ghana

Abstract Background and aims Understanding the importance of using personal protective equipment (PPE) and the influence of work‐post (working distance to main dust source—crusher) in stone quarries is vital for designing tailored interventions in minimizing workers' exposure to silica dust. Nonetheless, studies on silica dust and disease symptoms in Ghana are nascent. This study assessed how work‐post and use of required PPE jointly influence exposure to silica dust and disease symptoms in Ghana. Methods Generalized linear models (complementary log‐log regression) were fitted to cross‐sectional survey data of 524 stone quarry workers in Ghana to assess the joint effect of work‐post and PPE usage on self‐reported disease symptoms while controlling for relevant compositional and contextual factors. Results Stone quarry workers who work between 1‐100 m and beyond 100 m from the crusher with the required PPE were 90% and 87% respectively less likely to report eye irritation compared with their counterparts who work between 1 and 100 m from the crusher without the required PPE. Individuals who work between 1‐100 m and beyond 100 m from the crusher with the required PPE were 94% and 95% respectively less likely to report breathing difficulty compared with the reference group. Workers who work between 1‐100 m and beyond 100 m from the crusher with the required PPE were 97% and 99% respectively less likely to report coughing compared with the reference group. Workers who work between 1‐100 m and beyond 100 m from the crusher with the required PPE were 93% and 97% respectively less likely to report common cold compared with their counterparts who work between 1 and 100 m from the crusher without the required PPE. Conclusion There are adverse health implications for people who work in silica dust polluted environments, suggesting the need for a national safety and health policy to target them.

adverse health effects of working in the stone quarry industry have been well documented. 4,[19][20][21][22][23] Some of the adverse health outcomes mostly experienced by stone quarry workers include eye irritation, breathing difficulty, coughing, and common cold. One way to protect oneself against the risks associated with silica dust exposure is by using the required protective equipment during quarrying operations.
All silica-related diseases are preventable when the appropriate exposure control measures are applied. Protecting workers from silica dust exposure can be achieved through various exposure control measures. 5 This can be achieved through eliminating tasks that expose workers to silica dust, substituting crystalline silica materials with noncrystalline silica materials, using engineering controls such as water spray or local exhaust ventilation, limiting workers access to areas of high RCS concentration and using PPE. 24 When an organization exhibits good safety culture, it reflects in the safety attitude of the workers. In such working environments, workers are less likely to take health and safety risks. However, workers who find themselves in organizations that exhibit poor organizational safety culture are more likely to have a careless attitude toward health and safety. While it is the responsibility of organizations to provide safe and healthy working environments, it is also the responsibility of workers to follow the laid down safety rules and procedures of the organization to avoid accidents, injuries and work-related diseases.
Many empirical studies have assessed the health implications of silica dust exposure and the use of required PEE separately. Sairanen and Rinne 17 observed that dust concentration in stone quarries decreases with increasing distance from the main dust source. The study also mentioned that variation in dust concentration is high within distances ranging from 10 to 200 m. Arcury et al, 25

| Study area
The study was carried out in the southern part of Ghana where most of the stone quarry sites in the country are located. The five regions selected for this study are noted for quarry activities among the eight regions in southern Ghana. With the exception of Greater Accra Region where we selected six quarry sites from one community (Shai Hills) for data collection, all the other four regions had three communities selected with two sites in each of the community-Central

| Data collection and sampling procedure
This study is part of a research project that assessed the human health risks of occupational exposure to silica dust by stone quarry workers in Ghana. In all, 30 stone quarries were visited in the five regions for data collection from May 2018 to February 2019. A total of 600 questionnaires were administered to 20 workers in each quarry site however, 524 filled questionnaires were retrieved from the participants.
Even though all participants agreed to take part in the study, 76 did not return their questionnaires. Participants were recruited through a random sampling method. In each quarry site, workers were assigned numbers and a random number generator was used to select 20 workers. Participants of the study were 18 years and above.
Before the commencement of the study, the survey instrument was subjected to content validity test to ensure that it included all the items that are essential and eliminates undesirable items. The judgmental approach was used to establish content validity through literature reviews and evaluations by experts. A pretest was conducted on a smaller sample (60 respondents) to identify errors associated with the survey instrument and also assess response latency. Internal consistency of the survey instrument was assessed using cronbach alpha before the actual survey was conducted. After the survey, follow up visits were made to one quarry site in each region to assess the reliability of the responses.
All the procedures carried out in the study were in accordance with the ethical standards of the Minerals Commission of Ghana. A prior permission was obtained from the management of the quarries visited, the aim and details of the study were also explained to them.
Oral and written consent was obtained from all the participants before the study. Participants were neither coerced nor financially induced to take part in the study, we explained that their participation was voluntary. They were also informed that the information provided will contribute to the improvement of safety culture in the stone quarry industry in Ghana.

| Response variables
The dependent variables considered in this study were eye irritation, breathing difficulty, coughing, and common cold. For each disease symptom, respondents were asked if they often experience any of the symptoms since they started working in the stone quarry industry.
The dichotomous response was coded as 0 (for no) and 1 (for yes).

| Key predictor variable
The key independent/explanatory variable was selected based on literature, parsimony, practical significance, and theoretical relevance.
The key explanatory variable was derived from combining two variables (working distance to the crusher and wearing of required PPE).
This produced the predictor variable called "work-post PPE usage" with four mutually exclusive groups: 1-100 m without PPE, 1-100 m with PPE, above 100 m without PPE, and above 100 m with PPE. Sairanen and Rinne 17 in their study on dust emission from crushing of hard rock aggregates concluded that variation in dust concentration is F I G U R E 1 A map showing the communities in which the quarry sites are located in the five selected regions high within distances ranging from 10 to 200 m from the main dust source. They also observed in their study that crushing is the most significant dust source in stone quarries. Based on literature, we considered individuals who work within high dust concentration distances (10-200 m) for the study. The study considered individuals who work between 1 and 100 m from the main dust source and workers who work beyond 100 m from the main dust source.

| Compositional and contextual factors
Compositional factors refer to biosocial and socio-cultural characteristics of the stone quarry workers. Biosocial factors include age, sex, race, and ethnicity while socio-cultural factors include marital status, income, education, occupation, and religion among others. 27,28 Contextual factors are location-specific opportunities in a region or a place. 29 In this study, the compositional factors included age (young adult: less than 35 years, middle-aged adult: 35-55 years), education (no formal education, secondary/higher), household size (small: 1-5, medium: 6-10), family status (head, member), marital status (single, married, divorced). The contextual factor was region of residence (Central, Western, Greater Accra, Eastern, and Ashanti).

| Data analyses
The data were subjected to univariate and multivariate statistical analyses to examine the relationships and proportions between factors that influence self-reported disease symptoms while controlling for theoretically relevant compositional and contextual factors. All statistical analyses were performed using Stata 14 (StataCorp, College Station, Texas) SE software.

| Univariate analyses
Univariate analyses of predictors of self-reported disease symptoms were carried out using Pearson chi-square and Cramer's V statistic.
Pearson chi-square was used to test the association between categorical variables. It is used to estimate if two or more groups of samples are independent or not. Cramer's V statistics assess the strength of association among categorical variables. 30 The univariate results output was presented as a contingency table. Statistical significance of 0.05 was set for all analyses with a confidence interval of 95%.

| Multivariate analyses
The relationship between self-reported disease symptoms and workpost PPE usage was determined using complementary log-log regression models and reported as exponentiated coefficients or odds ratios (OR). An OR of 1 means that the predictor does not affect the odds of reporting a specific disease symptom; OR >1 means that the predictor is associated with higher odds of reporting a specific disease symptom; and OR <1 implies that the predictor is associated with lower odds of reporting a specific disease symptom. Compositional (age, ethnicity, education, household size, family status, marital status, religion) and contextual variable (region of residence) that have been suggested in literature to affect self-reported disease symptoms were controlled for in the models. the assumption for the model. 31 Eye irritation (42% no, 58% yes), breathing difficulty (26% no, 74% yes), coughing (19% no, 81% yes), and common cold (20% no, 80% yes). The regression models used in this study are built under the assumption of independence of subjects, but the cross-sectional survey has a hierarchical structure with respondents nested within the survey clusters, which could potentially bias the standard errors (SE). [32][33][34] 4 | RESULTS

| Descriptive analyses
Self-reported disease symptoms across the study regions are shown in Figure 2. A large number of stone quarry workers in this study have experienced disease symptoms associated with silica dust exposure (eye irritation, breathing difficulty, coughing, and common cold) at one point during their working time in the stone quarry sector. Eastern region recorded the highest (22%) likelihood of experiencing eye irritation while Central region recorded the lowest (17%).
Breathing difficulty was highest among quarry workers from Greater Accra and Eastern regions with both regions recording 21%.
Reported cases of coughing was highest (23%) in the Eastern region.
Common cold was also investigated in this study. Results from Figure 2 shows that common cold was highest in three regions; regions; Greater Accra, Ashanti, and Eastern with all recording 22%. The descriptive statistics suggest that there is a growing number of stone quarry workers in Ghana suffering from silica-dust-related disease symptoms which has dire health implications. Figure 3 shows the percentage of stone quarry workers who were committed to wearing the required PPE at the workplace. The descriptive results revealed that majority of stone quarry workers in Ghana do not wear PPE at work. This is a worrying development due to the risk associated with silica dust exposure. All the study regions performed poorly when it came to workers commitment to wearing the required PPE for quarrying activities. The region that had the worst record was Ashanti, 92% of the workers did not wear PPE. The proportion of workers who did not wear PPE during quarrying in the other regions is as follows; Eastern (91%), Western (89%), Greater Accra (88%), and Central (87%). These outcomes show that there is a huge safety culture gap particularly on the use of PPE by workers in the stone quarry sector.

| Univariate analyses
Results for the univariate analyses are provided in Table S1. The Pearson chi-square statistic rejected the hypothesis that work-post-PPE usage is independent of self-reported disease symptoms. This means that there is a relationship between work-post PPE usage and reported disease symptoms. Cramer's V statistics (Table S1) show a strong relationship between work-post PPE usage and self-reported disease symptoms. The results (Table S1) revealed a weak relationship between age and eye irritation, region also had a weak association with common cold. All the other covariates had no association with any of the self-reported disease symptoms.
The strength of the association between disease symptoms and work-post PPE usage based on Cramer's V statistic in decreasing order of magnitude, are as follows: coughing > common cold > breathing difficulty > eye irritation.

| Multivariate analyses
Three models were developed at the multivariate level for eye irritation, breathing difficulty, coughing, and common cold; work-post PPE usage and biosocial factors (model 1), socio-cultural factors (model 2), and contextual factors (model 3) to assess how they cumulatively influenced exposure to silica dust-related air pollutants and disease symptoms. Results for models 1 and 2 are provided in the Table S2-S5 and results for model 3 are presented in Table 1.

| Relationship between work-post PPE usage and eye irritation
Under biosocial factors (Table S2), workers who work between 1 and 100 m from the crusher with the required PPE were 90% less likely to report eye irritation. Quarry workers who work beyond 100 m from  the crusher with the required PPE were 87% less likely to report eye irritation compared with their counterparts who work between 1 and 100 m from the crusher without the required PPE. Middle-aged adults were 37% less likely to report eye irritation compared with the young adult group.
When socio-cultural factors were controlled for in model 2 (Table S2), the relationship between work-post PPE usage and experience of eye irritation remained statistically significant. Workers who work between 1 and 100 m and beyond 100 m from the crusher with the required PPE were 90% and 87% less likely to report eye irritation respectively compared with those who work between 1 and 100 m from the crusher without the required PPE. The direction and magnitude of the odds of reporting eye irritation persisted when age was controlled for in this model. Middle-aged adults were 49% less likely to report eye irritation compared with those in the young-adults group. However, educational attainment, household size, family status, and marital status were not significant predictors of eye irritation.
Just like it was observed in the biosocial and socio-cultural models, the relationship between work-post PPE usage and experience of eye irritation remained robust in the third model (Table 1) in which contextual factors were added. Workers who work between 1-100 m and beyond 100 m from the crusher with the required PPE were again 90% and 87% respectively less likely to report eye irritation compared with their counterparts who work between 1 and 100 m without the required PPE. The relationship between age and eye irritation persisted under this model following the previous trends.
Middle-aged adults were 48% less likely to report eye irritation compared with young adult workers. Again, educational attainment, household size, family status, marital status, and region did not predict eye irritation in the socio-cultural model.

| Relationship between work-post PPE usage and breathing difficulty
In model 1 (Table S3), stone quarry workers who work between 1-100 m and beyond 100 m from the crusher with the required PPE were 94% and 95% respectively less likely to report breathing difficulty compared with individuals who work between 1 and 100 m from the crusher without the required PPE. Individuals in the middle-aged adult category were 26% less likely to report breathing difficulty compared with the young adult group.
Model 2 controlled for socio-cultural factors (

| Relationship between work-post PPE usage and coughing
After controlling for biosocial factors in model 1 (Table S4), the results show that stone quarry workers who work between 1-100 m and beyond 100 m from the crusher with the required PPE were 96% and 98% respectively less likely to report coughing compared with those who work between 1 and 100 m from the crusher without the required PPE. However, the relationship between age and coughing was not statistically significant in this model.
When socio-cultural factors were control for in model 2 (Table S4) Educational attainment, household size, family status, and marital status were not significant predictors of coughing in the socio-cultural model.
Geographical region was controlled for in model 3 (

| Relationship between work-post PPE usage and common cold
Bio-social factors were controlled for in model 1 (Table S5). The results revealed that stone quarry workers who work between 1-100 m and beyond 100 m from the crusher with the required PPE were 92% and 96% respectively less likely to report common cold considered the joint effect of working distance to a stone crusher (where a majority of the silica dust is generated) and the use of required PPE on adverse health outcomes of stone quarry workers in developing countries have generally been unreported. There is a health and safety disadvantage to people who work in silica dust polluted environments.
People engaged in stonework, masonry, and construction work are highly exposed to silica dust and are likely to suffer adverse health effects associated with silica dust exposure. 15 It has been widely reported that occupation influences health outcomes. 4,20,[35][36][37] The findings of this study give a strong indication of the influence of the joint effect of work-post and PPE usage (work-post PPE usage) on the likelihood of stone quarry workers experiencing disease symptoms (eye irritation, breathing difficulty, coughing, and common cold).
Work-post PPE usage significantly predicted all the four disease symptoms in all the three models, that is, biosocial, socio-cultural, and contextual models. Based on our findings, stone quarry workers who protected themselves appropriately with the required PPE and were either close or far away from the crusher (main dust source) are at a lower risk of experiencing any of the disease symptoms (eye irritation, breathing difficulty, coughing, and common cold) than those who did not protect themselves irrespective of the distance from which they worked from the crusher. This finding agrees with the study conducted by Rongo et al, 38 on occupational exposure and health problems among small-scale industry workers in Dar es Salaam, Tanzania which found that workers who did not wear PPE and were exposed to dust and fumes reported health complains related to their work.
Neves et al, 39 in their study reported that adherence to PPE usage is determined by individual safety values and beliefs. Several studies have found that stone crushing is the main source of dust in stone quarries. [16][17][18] This was observed in the quarries visited. Proximity to the crusher only poses a risk to workers who do not use the required PPE. Workers who work further away from the crusher and do not use the required PPE have a higher chance of experiencing adverse health outcomes than those who work closer to the crusher and wear the required PPE. This outcome is an indication that the long-held perception that the further away one works from the main dust source (crusher) particularly without wearing PPE, the more the person is protected against silica-dust related ailments is incorrect. It has been reported that respirable silica dust produced by industrial processes poses a potential risk to people working within meters in that environment for years without protection. 13 These findings give credence to the fact that protecting oneself with PPE against silica dust exposure has huge health and safety benefits.
Our findings also established a significant relationship between age and the likelihood of experiencing eye irritation in the biosocial, socio-cultural, and contextual models, however, there was no relationship between age and common cold in all three models. finding is a major cause for concern especially knowing that silicosis takes years to fully manifest. It is even more worrying because silicosis has no cure. However, with the right safety measures in place, silicosis can be prevented. A key strategy in preventing silicosis is by using job-appropriate PPE during stone quarrying operations.

| Limitations of the study
The key strength of this study is its empirical disposition as respondents were real workplace employees. However, one of the limitations of the study is the reliance on self-reported measures to assess the likelihood of experiencing silica-dust-related disease symptoms. Outcomes of the relationship among the measures may, therefore, be confounded by common method variance. Wagner and Crampton 46 indicated in their meta-analytic study that even though this problem continues to be cited regularly, the immensity of the distortions may be exaggerated. It has been well documented in literature that self-reported measures have proven to be effective for organizational safety studies. 47

| CONCLUSION
The profile of Ghana's silica dust exposed population is not well studied and documented. Stone quarry workers are at high risk of developing silicarelated symptoms and diseases due to the cumulative effect of silica dust exposure, a finding that is supported by a large body of epidemiologic evidence. This study assessed the joint effect of work-post (distance to main dust source-crusher) and PPE usage on the likelihood of workers experiencing adverse health outcomes (eye irritation, breathing difficulty, coughing, and common cold) in stone quarries in Ghana. The study found that stone quarry workers who work between 1-100 m and beyond 100 m from the crusher with the required PPE reported lower likelihoods of experiencing eye irritation, breathing difficulty, coughing, and common cold. Individuals who work between 1-100 m and beyond 100 m from the crusher without the required PPE reported a higher likelihood of experiencing adverse health outcomes. This relationship was robust and persisted even when it was subjected to compositional and contextual attributes. Middle-aged adults reported lower frequencies of adverse health outcomes.
Across the five study regions, we found that majority of workers were not using PPE during quarrying operations. This is a major health and safety risk that can potentially put workers in a vulnerable position