Cost-Effectiveness of Coal Workers' Pneumoconiosis Prevention Based on Its Predicted Incidence within the Datong Coal Mine Group in China

We aimed to estimate the economic losses currently caused by coal workers’ pneumoconiosis (CWP) and, on the basis of these measurements, confirm the economic benefit of preventive measures. Our cohort study included 1,847 patients with CWP and 43,742 coal workers without CWP who were registered in the employment records of the Datong Coal Mine Group. We calculated the cumulative incidence rate of pneumoconiosis using the life-table method. We used the dose-response relationship between cumulative incidence density and cumulative dust exposure to predict the future trend in the incidence of CWP. We calculate the economic loss caused by CWP and economic effectiveness of CWP prevention by a step-wise model. The cumulative incidence rates of CWP in the tunneling, mining, combining, and helping cohorts were 58.7%, 28.1%, 21.7%, and 4.0%, respectively. The cumulative incidence rates increased gradually with increasing cumulative dust exposure (CDE). We predicted 4,300 new CWP cases, assuming the dust concentrations remained at the levels of 2011. If advanced dustproof equipment was adopted, 537 fewer people would be diagnosed with CWP. In all, losses of 1.207 billion Renminbi (RMB, official currency of China) would be prevented and 4,698.8 healthy life years would be gained. Investments in advanced dustproof equipment would be total 843 million RMB, according to our study; the ratio of investment to restored economic losses was 1:1.43. Controlling workplace dust concentrations is critical to reduce the onset of pneumoconiosis and to achieve economic benefits.


Introduction
Coal workers' pneumoconiosis (CWP) is the most common occupational disease among coal workers [1][2][3][4][5][6]. These workers develop CWP because they are exposed to coal dust, mixed dust, Coal Mine Group is representative of the coal mining industry in China for its experience in coal production and dustproof efforts.
For our study, we evaluated dust-exposed workers with and without CWP from the Datong Coal Mine Group. The goals of our study were to analyze the prevalence of CWP and predict the number of CWP cases among workers who first experienced dust exposure between 1970 and 2010; to calculate economic losses and healthy life losses due to CWP; and to calculate economic and social benefits achieved as a result of dustproof investments. This is important information for occupational health management departments and coal mining enterprises. We aimed to provide a theoretical basis for the formulation and revision of laws, regulations, and standards for the prevention of pneumoconiosis in China.

Study population
Our study cohort were conducted in 2012 in Datong Coal Mine Group at a total of 15 mines. Coal workers were included in our study if their dust exposure started between January 1, 1970 and December 31, 2010 as part of the Datong Coal Mine Group and if they were exposed to dust for at least one year. We required physical examination cards, detailed records of occupational history, and posterior-anterior chest radiographs for all workers enrolled in our study. The study was approved by the Medical Ethics Committee of China Medical University (permit number CMU6206-3008). The need for written consent was not deemed necessary and was waived by the Medical Ethics Committee. The study was carried out in accordance with the approved guidelines. In order to protect the privacy of research subjects and the confidentiality of their personal information, Identifying information, including names, initials, identity card numbers, or job serial numbers of the subjects, did not exsit in our written descriptions.
We constructed a database, which included demographic details, work history records including the dates of dust exposure, individual medical records, pneumoconiosis diagnosis records, dust concentrations of the workplace, economic losses attributed to pneumoconiosis patients, and costs associated with dustproof facilities. The demographic details and work history records were obtained from personnel files in the human resources department. The individual medical and pneumoconiosis diagnosis records were obtained from the occupational disease prevention and treatment hospital. The dust concentrations of the workplaces were obtained from the department of dust detection and monitoring. The economic loss data were obtained from department of industrial injury insurance, hospital, and finance. The costs of dustproof equipment and supplies were obtained from the equipment department and finance department.

Diagnosis of pneumoconiosis
The diagnosis of CWP was based on the "Diagnostic criteria of pneumoconiosis" and corresponding standard films of pneumoconiosis in China [25]. The Chinese "Diagnostic criteria of pneumoconiosis" was formulated based on the "Guidelines for the use of the ILO international classification of radiographs of pneumoconioses" [26]. They are based on the same principle. Pneumoconioses were classified as stage Ⅰ, Ⅱ, and Ⅲ in Chinese Diagnostic criteria of pneumoconiosis based on the ILO criteria. Chest radiographs of subjects were read and the diagnosis was made independently by five qualified experts who were members of the pneumoconiosis diagnosis committee. If the diagnoses differed among the five experts, the opinion of the majority ruled. The experts identified disabilities associated with pneumoconiosis based on the "Standard of work-related disability identification" (Table A in S1 File) [27].

Occupational categories
We defined four work areas in the underground mines of the Datong Coal Mine Group by reviewing the work history of the subjects [28,29]: tunneling, mining, combining, and helping. Workers who consistently worked in the same area were defined by their respective work area titles: tunneling, mining, combining, or helping. The duration of dust exposure for each coal worker was the sum of years of each dust exposure job. The duration of each dust exposure job was measured from the start date to the end date of the job. Workers were classified as "tunneling" if their duration in tunneling areas accounted for more than half of the entire duration of dust exposure. Workers were classified as "mining" if they worked in tunneling areas for less than 2 years and their duration in mining areas accounted for more than half of the entire duration of dust exposure. Workers were classified as "combining" if their duration in tunneling areas was more than 2 years but less than half of the entire duration of dust exposure. Workers were classified as "helping" if they could not be included in the tunneling, mining, or combining categories.

Dust exposure data
Dust was collected on a membrane filter and measured by gravimetric method, which is the national standard of measurement for dust concentration [30,31]. This information was then used to calculate the cumulative dust exposure (CDE) for each coal worker [32]. CDE is usually estimated according to the following formula [8] (Eq 1): In the formula, CDE is calculated in milligrams/cubic meter-years (mgÁyears); n is the total number of job titles held by the individual during his work history; Cj is the 8-hour timeweighted mean concentration of dust in milligrams/cubic meter for the jth job title within a facility and an employment period; and Tj is the duration of employment in years in the jth job.

Calculation of the cumulative incidence rate
We established four subcohorts according to occupational categories: tunneling, mining, combining, and helping. We calculated the cumulative incidence rate of CWP in the corresponding years of observation for each cohort using the life-table method [23,28]. The rates were analyzed by Peto's log-rank test. We calculated the cumulative incidence rates, using the tunneling cohort as an example in Table B in S1 File. Cumulative incidence rates of the other cohorts were calculated in the same manner.

Predicted future incidence of CWP
We calculated the dose-response relationship between cumulative incidence density (CID) and CDE for the four occupational categories. We also introduced the variable of person-years in order to account for time in the equation. The incidence density of each dose group was the ratio of incidence number and the adjusted observed person-years. CID was calculated by the life-table method. We translated CDE into its natural logarithm and translated CID into its logit value, according to the following formula (Eq 2): logit ¼ ln P 1 À P By this method, we obtained the linear regression equation for the logit and the natural logarithm of CDE. We calculated the estimated values of incidence density from the equation that was used to predict the future incidence of CWP at the dust concentration level of 2011 and after adopting dust-reduction measures.
Compensation standards of direct economic loss caused by CWP Direct economic losses included medical costs, lump-sum grants for disability, disability allowances, lump-sum grants for death, funeral grants, nursing costs, food allowances, traffic fees, and pension costs for dependent relatives (Table C in S1 File) [14,33,34]. The total amount of compensation was calculated according to the 2010 revision of "Ordinance of industrial injury insurance" [19].
Medical costs included inspection fees, hospitalization fees, and medicine and treatmentrelated costs. The total medical cost for each CWP patient was calculated. Each CWP patient was compensated with a lump-sum grant for disability (Table D in S1 File). Disability allowances were only distributed to patients whose disability grades ranged from 2 to 6 before retirement (Table E in S1 File). Funeral grants equals the average monthly wage of the staff and workers in a region multiplied by 6 months, which were distributed to relatives of deceased CWP patients. If the disabled workers died during the period of suspension with pay caused by CWP, their relatives were compensated with a lump-sum grant for death. Nursing costs were compensated to CWP patients who required nursing care from industrial injury insurance funds (Table F in S1 File). Food allowances were paid to those who required medical treatment, which is 35 RMB per day in Shanxi Province. Dependent relative pension costs were paid to the families of deceased CWP patients whose disability grades ranged from 2 to 4 or CWP patients who died during the period of suspension with pay regardless of disability grade, if the relatives relied on CWP patients when they were alive. Traffic fees were compensated to CWP patients who traveled to other areas for medical treatment. In our study, very few CWP patients traveled to other areas for medical treatment, so we did not calculate this factor into the total economic loss.

Compensation standards of indirect economic loss caused by CWP
Indirect economic losses included social productivity losses caused by CWP patients, costs of training new employees, traffic fees for people who accompanied CWP patients, social productivity losses caused by people who accompanied CWP patients, and losses owed to stopping or cutting production (Table C in S1 File) [15].
Social productivity losses attributed to pneumoconiosis patients were calculated using ∑DALY, gross domestic product (GDP) per capita, productivity weight [22]. DALY was used to evaluate life loss caused by pneumoconiosis, which includes years lived with disability (YLD) and years of life lost (YLL) [21,22]. If coal mine workers suffered from CWP before retirement, new workers were recruited to supplement their jobs; the cost of training new supplemental employees was included in the economic losses caused by pneumoconiosis. Traffic fees for people who accompanied hospitalized CWP patients were also included in the economic losses caused by pneumoconiosis. According to travel standards of China's residents, the average annual per capita traffic fee was 2,000 RMB. Traffic fees for accompanying people were the average daily traffic costs multiplied by the number of days spent in the hospital. Social productivity losses caused by accompanying people were included in the economic losses caused by pneumoconiosis; it was calculated as the product of GDP, productivity weight, and length of disability of pneumoconiosis patients. Since coal mining facilities did not stop or limit production because of dust hazards, there was no economic loss related to stopping or cutting production.

Economic effectiveness of investment in dustproof equipment and practices
We calculated the economic benefits of dustproof to be the difference between the monetary investment in dustproof equipment and practices and the restored economic losses owed to the investment. We used current CWP patients as the example for calculating economic losses and we used prices and costs in 2011 as the standards for compensation and costs. Further, social benefits were calculated by assessing the restoration of healthy life years of the workers exposed to dust.

Baseline characteristics
There were 49210 people in Datong Coal Mine Group who started to expose to dust from 1970 to 2010. 3621 people were excluded because of incomplete data, such as any drop outs, workers loss to follow up, or workers who die from other disease. 45589 subjects met the requirements, the coincidence rate was 92.6%. We evaluated a total of 45,589 coal workers in our study, including 1,847 patients with CWP and 43,742 workers without CWP. The average age of CWP onset for the 1,847 CWP patients was 45.9 ± 8.3 years and the average duration of dust exposure for these patients was 19.8 ± 7.2 years. In 2011, the average age of the 43,742 coal workers without CWP was 42.3 ± 9.6 years and the average duration of dust exposure for these workers was 18.9 ± 9.5 years. We observed statistically significant differences between the dustexposed workers with CWP and without CWP in years of first dust exposure, occupational category, and CDE, according to the chi square test ( Table 1).
The cumulative incidence rates of CWP of tunneling, mining, combining, and helping cohorts were 58.7%, 28.1%, 21.7%, and 4.0%, respectively, during the 42-year observation period ( Fig 1A). We observed significant differences between occupational categories, according to the log-rank test. The cumulative incidence rate of CWP in the group of workers first exposed to dust in the 1970s was 12.7% and the rate in the group first exposed to dust in the 1980s was 7.6% during the 32-year observation period ( Fig 1B). The cumulative incidence rates were 25.4%, 7.3%, and 3.2% among coal workers whose CDEs were ! 2000 mgÁyears, ! 100 mgÁyears, and <100 mgÁyears during the 42-year observation period ( Fig 1C). The dust concentration decreased with time in different work areas (Table 2). We observed a doseresponse relationship between cumulative incidence rate and CDE (S1 Fig).

Prediction model of CWP
The cumulative incidence densities were calculated for each group of cumulative dust exposure by using method of life table for tunneling workers (Table G in S1 File). We observed a doseresponse relationship between CDE and cumulative incidence density estimation (Fig 2A1). After logarithmic transformation of variables, the following linear equation of the two variables was obtained (Eq 3) (Fig 2A2): In the same way, we observed a dose-response relationship between CID and CDE for mining workers (Fig 2B1). The linear equation for the mining workers was (Eq 4) ( Fig 2B2): The dose-response relationship for combining workers is shown in Fig 2C1. The linear equation of the combining workers was (Eq 5) ( Fig 2C2): The dose-response relationship for helping workers is shown in Fig 2D1. The linear equation of the mining workers was (Eq 6) ( Fig 2D2): The incidence density estimation for each segment was calculated by these equations. The tunneling cohort is used as an example in Table H in S1 File.

Prediction of future CWP cases after decreasing dust levels
Dustproof equipment could decrease workplace dust concentrations by eighty percent of the present levels [35]. To assess the potential decline in CWP after dustproof, we calculated the CDE for each coal worker until his retirement assuming a dust concentration at 20% of the 2011 level. The assumed dust concentrations were 6.3 mg/m 3 , 4.6 mg/m 3 , 4.6 mg/m 3 , and 0.04 mg/m 3 in the tunneling, mining, combining, and helping areas, respectively. According to the incidence density estimations of the tunneling, mining, combining, and helping cohorts, we predicted the number of future CWP cases within the life expectancy of each subject to be 1,245.5, 1,084.7, 1,020.5, and 412.5 among tunneling (Table N in S1 File), mining (Table O in S1 File), combining (Table P in S1 File), and helping cohorts (Table Q in S1 File), respectively. We predicted that 3,763 coal workers would develop CWP after the implementation of advanced dustproof equipment and practices, which equates to 537 fewer people with CWP (Table 3). In this study, 1,047 CWP patients had disability grades that ranged from 2 to 6 before retirement. The total disability allowance for CWP patients was calculated to be 1,199,356,930.5 RMB, based on the costs in 2011 (Table T in S1 File). The average monthly wage in Shanxi Province was 3,269.2 RMB in 2011 [36]. Funeral grants for the 1,847 CWP patients were calculated to be 36,228,941.9 RMB (3,269.2 RMB × 6 months × 1847 persons), based on the costs in 2011. Funeral grants for persons who did not die from work are 2,000 RMB [37]. Therefore, in our study, funeral grants totaled 32 In 2011, the average dependent relative pension cost for deceased CWP patients from the Datong Coal Mine Group was 17,782.32 RMB. By the end of 2011, the average age of death of the 210 CWP patients who died was 55.4 years. The average life expectancy of the patients was 80.1 years. The average life loss was 24.7 years. Therefore, we calculated the total dependent relative pension costs to be 294,480,198.3 RMB (17,782.32 RMB × 671 persons × 24.7 years) for 655 CWP patients whose disability grade ranged from 2 to 4 and for 16 CWP patients whose disability grade was 7 but died during the period of suspension with pay.

Direct economic losses attributed to CWP patients
For our analysis, we calculated the direct economic losses caused by 1,847 CWP patients to total 3,488,134,634.1 RMB, assuming prices and costs of 2011. The average direct economic loss was 1,888,540.7 RMB per patient with CWP (Table 4). Medical costs accounted for 38.7% and disability allowances accounted for 34.4% of total direct economic loss.

Indirect economic losses attributed to CWP patients
At the end of our observation period, among 1,847 patients with CWP, 1,637 patients were alive and 210 cases were deceased. Eight patients had a disability grade of 2, 147 had a grade of  According to travel standards for China's residents, the average annual traffic fee per capita was 2,000 RMB. Based on our investigation, the average duration of pneumoconiosis was 31.6 years and patients were hospitalized for an average of 75 days per year. For this analysis, we assumed that each hospitalized patient required one accompanying family caregiver. Therefore, we calculated traffic fees to be 23,962,927.4 RMB (2,000 RMB × 1,847 persons × 31.6 years × 75 days/365 days/year).
In 2011, China's GDP per capita was 35,181 RMB [36] and the productivity weight of the total population was 0.5. Therefore, we calculated the social productivity loss caused by accompanying family members, assuming one per hospitalized patient, to be 210,759,937.2 RMB (35,181 RMB × 1,847 persons × 0.5 × 31.6 years × 75 days/365 days/year).
The total indirect economic losses caused by CWP patients totaled 663,187,570.9 RMB. The average loss was 359,062.0 RMB per patient (663,187,570.85 RMB / 1,847 persons). Social productivity loss caused by CWP patients accounted for 64.3% and social productivity loss caused by accompanying family members accounted for 31.8% of the total indirect cost ( Table 5).

Total economic losses caused by CWP patients
On the basis of our calculations, we concluded that the total economic loss caused by 1,847 CWP patients who experienced their first dust exposure between January 1,1970 and December 31, 2010 was 4,151,322,205.0 RMB (3,488,134,634.1 RMB + 663,187,570.9 RMB). The average loss attributed to each patient was 2,247,602.7 RMB (4,151,322,205.0 RMB / 1,847 persons). Direct economic loss accounted for 84.0% and indirect economic loss accounted for

Cost-effectiveness analysis of future dustproof investments
If the 43,742 coal mine workers in our study continued working from January 1, 2012 until their retirement, the average dust exposure time would be 17.73 years. If new equipment, such as an underground water purification system and a high-pressure spray system with pneumatic power, was adopted, workplace dust concentrations would decrease by 80% of the present levels [35]. According to the results of our study, the average number of working areas (tunneling, mining, and combining) and trans-shipment points were 13

Discussion and Conclusions
Occupational injuries and diseases occur as a result of the production of goods and services. In other words, they are economic phenomena [38]. The failure to recognize this central aspect of occupational health limits the effectiveness of interventions ostensibly designed to prevent disease and injury [13]. Currently, pneumoconiosis is the primary occupational disease in China [11] and the government has assigned great importance to the prevention and treatment of pneumoconiosis. In 1987, "Pneumoconiosis disease prevention and control regulations of the People's Republic of China" was promulgated and implemented [39]. Additionally, in 2001, "Law of the People's Republic of China on Prevention and Control of Occupational Diseases" was promulgated; it was revised in 2011 [40]. The implementation of these laws and regulations provided legal guarantees for the prevention and treatment of pneumoconiosis.
Despite laws aimed at prevention, the cumulative incidence rates of CWP remained high among the Datong Coal Mine Group. We evaluated the number of patients with CWP who experienced their first dust exposure between 1970 and 2010 in the Datong Coal Mine Group. We predicted that, among all these workers, 4,300 would develop CWP in the future if workplace dust concentrations remained at the levels of 2011. We predicted that 3,763 coal workers would develop CWP if advanced dustproof strategies were adopted, which equates to 537 In our survey, direct economic loss exceeded indirect economic loss and accounted for 84.0% of total economic loss. This result was different from previous research in China [15], but it was similar to reports from other developed countries [20]. Our finding might be the result of increases in lump-sum grants for disability, lump-sum grants for death, and medical costs, which was a promotion for the prevention and treatment of pneumoconiosis.
Dust concentrations in the workplaces of the Datong Coal Mine Group have decreased significantly since the implementation of a series of dustproof measures, but the levels are still not controlled under the standards defined in "Occupational exposure limit (OEL) for hazardous agents" [8,41]. Dust is still the primary occupational hazard in the Datong Coal Mine Group, as well as other coal enterprises. If dust concentrations decrease, the incidence of CWP will decrease.
The implementation of advanced dustproof equipment, such as a high-pressure spray system with pneumatic power and an underground water purification system, would require an investment of 843 million RMB, according to our analysis, but the dust concentrations would subsequently decline in the coal mines. For example, we expect the dust level to be reduced to 6.3 mg/m 3 in the tunneling area. At this level, if a worker was exposed to dust for 40 years, his CDE would be 252.8 mgÁyears. According to the results of our study, the risk probability of developing CWP of tunneling workers with a CDE of 200 mgÁyears was only 0.0064. The majority of workers would, therefore, be free from pneumoconiosis, and 1.207 billion RMB would be restored. We calculated the ratio of dustproof investments and restored economic losses to be 1:1.43 (843/1,207). Controlling workplace dust concentrations is critical to reduce the onset of pneumoconiosis and to achieve economic benefits.
It is important to recognize pneumoconiosis prevention as an economic activity [13]. In our investigation, we observed substantial economic losses and healthy life losses for CWP patients who were first exposed to dust between 1970 and 2010 in the Datong Coal Mine Group. Improving dust protection could bring considerable economic benefits to the coal mining industry owing to a reduced number of CWP cases. Dust also causes a variety of diseases of the respiratory system and it can increase the incidence rate of other chronic diseases [42,43]. If the economic losses caused by other diseases related to dust were considered, including physical and mental pain to patients and their families, broken families, and social instability, the losses would total more than our original estimates. Such estimates would be a more powerful illustration of the importance of dust control and exposure prevention. If funds are invested early in dustproof equipment and strategies, great benefits will be realized owing to delayed effects of pneumoconiosis [44].
In conclusion, implementing dustproof devices and practices would not only reduce the incidence of CWP but also, in the long run, increase financial profits. Occupational health management departments and coal mining enterprises can benefit from an understanding of this information.
Supporting Information S1 Fig. Dose-response relationship between cumulative incidence rate of CWP and cumulative dust exposure. (TIF) S1 File. Supplementary tables A-AA. Table A. The relationship between the disability grade of occupational disease and the stage of pneumoconiosis. Table B. Cumulative incidence rate of CWP in the tunneling cohort.