A practical framework to develop and prioritize safety interventions to improve underground coal miners’ safety performance

Abstract

BACKGROUND:

Improvement of workers’ safety performance is an integral and essential part of safety management. Relevant safety interventions to improve workers’ safety performance are generally difficult to establish when there is a wide range of occupational hazards and at-risk individuals’ features.

OBJECTIVE:

This study aimed at formulating a practical approach to develop and prioritize potential safety interventions based on occupational and individual risk factors perceived by workers to promote workers’ safety performance.

METHODS:

A simple framework developed to identify and prioritize the suitable safety interventions. This framework made use of data collected using standardized and validated questionnaire and domain experts’ opinions. Pearson correlation coefficients, exploratory factor analysis, and multiple linear regression were used to identify significant risk factors associated with workers’ safety performance. Data were collected by interviewing 202 coal mine workers with occupational injuries, and their immediate supervisors from three mines.

RESULTS:

Safety performance was associated with the occupational factor-domain (poor working condition, poor safety environment, poor job satisfaction, and high job stress) only (regression coefficient = 2.14, p < 0.01). The following interventions were identified and prioritized to promote workers’ safety performance: provide fair compensation to workers, job-specific and safety training, promotion policy, achievable targets, relevant perks/benefits, safety training awareness, workplace lighting, ventilation network, sensitize the management, associate safety performance to promotion, and develop team spirit.

CONCLUSION:

Our approach helps to identify and prioritize the most relevant interventions to promote safety at work when there are multiple risk factors.

Keywords

Safety intervention and control safety management occupational injuries risk assessment data analysis

1 Introduction

Worldwide, around 340 million occupational accidents occur annually which have severe consequences for workforce’s health and morale as well as organisations [1]. The mining industry is one of the major industries where occupational safety is of prime concern. Despite decades of safety management research, occupational accidents and injuries are still common, though the trend has been downward in the recent past. The coal mining sector, which produces the prime energy source in most developing countries, is a big part of the mining industry and is regarded as a high-risk sector [2, 3]. In India, 925 workers have died, and 4991 workers have been seriously injured between 2008 and 2019 in the coal mining sector [4]. Along with India, several other coal-producing countries, such as China, Australia, and the United States of America, suffers from occupational injuries and fatalities [2 , 5]. Hence, the coal mining industry needs relevant workplace intervention strategies to improve state of safety in mines.

Previously, the improvement of occupational health and safety was focused on technology improvement, which was relatively successful. Nowadays, it is widely acknowledged that technological improvement alone is insufficient to improve state of safety in an organisation [6]. There is a need to investigate whether the occupational and individual aspects perceived by the workers and their safety performance as perceived by their supervisors could be jointly used to identify the most relevant safety interventions.

Worker’s poor safety performance critically increases the risk of accidents/injuries in high-risk workplaces such as underground mines and construction [7]. The safety performance of a worker defined as an individual’s actions and behavior, when performing allocated jobs to promote the individual’s health and safety, that of co-workers and the public, as well as safety environment and organization [8 –10]. It was reported that safety outcome and safety performance were associated with workplace characteristics and circumstances, safety climate and culture, and employee characteristics [11 –14]. Furthermore, safety attitude has a direct influence on safety performance, but it also exerts an influence through safety climate. As a result, safety climate acts as a mediator between safety attitude and safety performance, and thus it plays a crucial role in predicting the level of safety performance [15]. Emotional stability is positively associated with safety performance [16].

The identification and prioritization of the most relevant safety interventions based on occupational and individual risk factors may improve workers’ safety performance. Such an approach may provide timely interventions as coal mining organizations generally work in a resource-constrained environment when it comes to safety. Moreover, both workers and their supervisors may more likely support the interventions, and it will be more efficient. Prioritization of safety interventions is important because limited resources (capital, equipment, personnel, and policy implementation) are available to mine management to improve safety. Hence, it is important to formulate a simple and practical approach for prioritizing safety interventions based on potential occupational and individual risk factors perceived by the workers and their supervisors.

It should be noted that a very few studies attempted to provide a framework for mine management to improve workers’ safety by formulating, selecting, and prioritizing safety interventions. There are some studies that reported effective safety interventions [17 –20], and a few studies developed methodologies to prioritize safety interventions [21, 22].

The present study focuses on the formulation, selection, and prioritization of relevant safety interventions based on workers’ occupational and individual factors to improve safety performance of coal miners in case study mines using a practical approach. The Ministry of Human Resource and Development of the Government of India and Research and Development Cell of the Indian Institute of Technology Kharagpur supported the study.

2 Material and methods

2.1 Subjects

This study was conducted on three underground coal mines in the eastern part of India. These three coal mines employed a total of 2900 workers at the time of data collection. The information about workers with injury was taken from the injury register maintained by the safety departments of mines. In order to establish safety interventions, this study focused on a sample of 202 workers randomly selected from a list of 780 underground miners who experienced an unintentional occupational accident, causing bodily damage, and at least one workday loss and compensation in the last five years. Sample size was calculated for finite population [23].

2.2 Data Collection

A validated questionnaire called worker’s response device (WRD) was used to assess workers’ perceived risk factors [24], which included poor working conditions, poor safety environment, poor job satisfaction, high job stress, poor emotional stability, poor job involvement, and impulsiveness. Workers’ experience was also recorded. A total number of eight risk factors were considered in this study. The WRD was filled by interviewing the workers; each interview took approximately 45 minutes. The interview was conducted privately; as a result, supervisors could not influence the workers’ responses. Another validated standardized questionnaire called the supervisor’s response device (SRD) was used to assess the worker’s safety performance perceived by the supervisors (24). The SRD was administered to the immediate supervisors of workers. Each item of WRD and SRD had three possible responses: yes, cannot say, and no, which were assigned the values 3, 2, and 1, respectively. For the items negatively formulated (indicated by “*”), the responses were assigned the values 1, 2, and 3, respectively (Appendix-A). The interview team consisted of researchers and personnel from academia. They possess extensive experience in conducting survey-based interviews and are involved in various studies. They also have working experience in the mining sector and received training before the interviews. The questionnaires were translated into local languages (Hindi and Bengali) to increase the workers’ understanding. The team provided necessary clarifications to the participant without exerting any influence on their responses. Previous experiences led us to use three-point scales instead of five-point or seven-point scales because using these scales verbally during interviews was found to be challenging, while three-point scales ensured that the subjects provide clear and accurate responses.

2.3 Statistical analysis

Pearson correlation coefficient was used to find out the correlation between workers’ safety performance and risk factors. Exploratory factor analysis was employed to find out the underlying factors. The principal axis factoring extraction method and varimax rotation were used in exploratory factor analysis. Further, multiple linear regression was used to investigate the relationship between underlying factors and safety performance. All statistical significances were reported using a two-sided p value of <0.05. SPSS version 25 was used for all statistical analyses.

2.4 Methodology for safety intervention development and prioritization

A simplified approach to formulate, select and prioritize safety interventions is presented in this section. The stepwise approach is given below:

Pearson correlation coefficient was used to assess the correlation between workers’ safety performance and risk factors; also, the correlation among risk factors. It was found that risk factors are highly correlated. So, exploratory factor analysis was conducted to find underlying factors, followed by multiple linear regression to determine significant underlying factor/factors. After that, significant underlying factor/factors were considered for safety intervention development and prioritization.

First, each risk factor was decomposed into different needed improvements, based on questionnaire items (Appendix-A). For each needed improvement, several necessary interventions were formulated using brainstorming and discussion with experts from industry and academia.

For each needed improvement, one necessary intervention was selected out of several interventions formulated earlier, based on experts’ opinions. Experts from industry and academia were asked to rate the intervention on a scale of 1 to 7, where 1 represents least effective and 7 represents most effective. Based on the average expert opinion score (EO), one necessary intervention was selected for each needed improvement.

These selected necessary interventions were prioritized based on how efficient the intervention is in achieving the corresponding needed improvement (expert opinion score of each necessary intervention) and how important the corresponding needed improvement is in improving worker’s safety performance (importance rating of needed improvement). Because implementation of an efficient intervention (high expert opinion score) could effectively achieve the needed improvement as opposed to a less-efficient intervention, which might partially improve the issue. Also, if a needed improvement with high importance (strongly associated with safety performance/ needed improvement with high importance rating) is achieved, it will improve safety performance to a higher degree than a needed improvement with less importance. The stepwise methodology for calculating the importance rating of each needed improvement, the effectiveness of each necessary intervention, and subsequent priority weights of interventions is given in the following steps.

For each needed improvement, the Average Score (AS_i) was calculated, which is the average value of the questionnaire items score falling under the specific needed improvements. AS_i ranged from 1 to 3 (Annexure-A), as all responses to each questionnaire item range between 1 to 3.

“s_j” denotes the response to an individual questionnaire item by a single worker. If there are “n” questionnaire items that fall under a particular needed improvement, the average score for each needed improvement for an individual worker can be calculated as the sum of all “s_j” values divided by “n”. $Average score for k^{th} worker = \frac{\sum_{j = 1}^{n} s_{j}}{n}$

To determine the average score (ASi) for that specific needed improvement, accounting for all the workers denoted by m (m = 202 in this study), the following expression was employed. ${AS}_{i} = \frac{\sum_{k = 1}^{m} {(\frac{\sum_{j = 1}^{n} s_{j}}{n})}_{k}}{m}$

So, a higher value of ASi (closer to 3) denotes that more workers were affected by that specific issue that needed improvement.

To calculate the importance rating (R_i) of needed improvement, AS_i is multiplied by the absolute regression coefficient of the underlying factor |β_i|, and factor loading of the risk factor on that underlying factor (λ_i). $R_{i} = {AS}_{i} \times | β_{i} | \times λ_{i}$

Factor loading quantifies the extent to which a risk factor is related to an underlying factor, and the regression coefficient represents the strength of the association between the underlying factor and worker’s safety performance. So, factor loading multiplied with regression coefficient represents the strength of association between a risk factor and worker’s safety performance or its effect on elevating worker’s safety performance.

Normalized importance rating is calculated using following expression. $r_{i} = \frac{R_{i}}{\sum R_{i}}$

Normalized impact rating denotes the relative importance of the needed improvement, which shows the effect of the needed improvement on worker’s safety performance compared to the other needed improvements.

To calculate the priority weight of necessary interventions (W_i), normalized importance rating (r_i) is multiplied by expert opinion score (EO_i). Expert opinion score represents the practical effectiveness of the intervention to achieve the corresponding improvement and, in turn, improve worker’s safety performance. $W_{i} = r_{i} \times {EO}_{i}$

To calculate normalized priority weight (ω_i) following expression was used. $ω_{i} = \frac{W_{i}}{\sum W_{i}}$

Based on these normalized priority weights, necessary interventions were ranked. Normalized priority weight denotes the relative effectiveness of each necessary intervention and rank denotes the priority it should be given in implementation through resource allocation.

3 Results

The correlation matrix shown in Table 1 reveals that worker’s safety performance was significantly and negatively correlated with poor working condition, poor safety environment, poor emotional stability, and impulsiveness (Pearson correlation coefficients between –0.293 and –0.144, p < 0.05). It can also be noted that there are high correlations between these risk factors (Pearson correlation coefficient up to 0.650, mostly with p < 0.01).

Table 1
Correlation matrix (Pearson Correlation Coefficient)

Safety performance Poor Working condition Poor safety environment Poor job satisfaction High job stress Poor emotional stability Poor job involvement impulsiveness Experience

Safety performance 1

Poor Working Condition –0.145* 1

Poor safety environment –0.293 0.434 1

Poor job satisfaction –0.136 0.315 0.490 1

High job stress –0.083 0.366 0.465 0.650 1

Poor emotional stability –0.195 0.231 0.326 0.325 0.334 1

Poor job involvement –0.013 0.129* 0.285 0.444 0.317 0.431 1

Impulsiveness –0.144* 0.143* 0.287 0.417 0.311 0.425 0.402** 1

Experience 0.096 –0.025 –0.035 –0.100 –0.073 0.181* 0.212* –0.069 1

	Safety performance	Poor Working condition	Poor safety environment	Poor job satisfaction	High job stress	Poor emotional stability	Poor job involvement	impulsiveness	Experience
Safety performance	1
Poor Working Condition	–0.145*	1
Poor safety environment	–0.293**	0.434**	1
Poor job satisfaction	–0.136	0.315**	0.490**	1
High job stress	–0.083	0.366**	0.465**	0.650**	1
Poor emotional stability	–0.195**	0.231**	0.326**	0.325**	0.334**	1
Poor job involvement	–0.013	0.129*	0.285**	0.444**	0.317**	0.431**	1
Impulsiveness	–0.144*	0.143*	0.287**	0.417**	0.311**	0.425**	0.402**	1
Experience	0.096	–0.025	–0.035	–0.100	–0.073	0.181*	0.212*	–0.069	1

The suitability of data for exploratory factor analysis was satisfied. Kaiser-Meyer-Olkin (KMO) criterion was found to be 0.782, and it was above the recommended values of 0.50–0.60 [25 –27]. Further, Bartlett’s test [28] revealed that the correlation matrix was not an identity matrix, and it can be inferred that there was a significant correlation between the risk factors (chi-square = 422.5, df = 28, p < 0.001). Table 2 shows the results of the exploratory factor analysis. Based on Kaiser’s criterion (Eigen values > 1) and scree plot [29, 30], we found two main factors (called factors 1 and 2) with eigen values greater than 1. Poor working conditions, poor safety environment, poor job satisfaction, and high job stress were highly loaded on factor 1, while poor emotional stability, poor job involvement, and impulsiveness were highly loaded on factor 2. These two main factors explained 55.8% of the total variance. Therefore, all the risk factors can be clubbed into two main dimensions. Factor 1 represented the occupational factor domain, and factor 2 represented the individual factor domain.

Table 2

Exploratory factor analysis for risk factors and associations of worker’s safety performance with the two main factors: regression coefficient (β) and standard error (SE)

	Factor 1	Factor 2	Communalities
Worker’s perceived risk
factors
Poor working conditions	0.486	0.078	0.236
Poor safety environment	0.641	0.104	0.430
Poor job satisfaction	0.767	0.235	0.636
High job stress	0.750	0.141	0.578
Poor emotional stability	0.345	0.614	0.438
Poor job involvement	0.316	0.580	0.540
Impulsiveness	0.399	0.428	0.312
Experience	–0.174	0.269	0.147
% Variance explained	38.878	15.988
Eigenvalue	3.190	1.279
Factors	β (SE)	p	VIF ^b
Factor 1	–2.143 (0.628)	0.003**	1.034
Factor 2	–0.127 (0.824)	0.877	1.034
Constant	39.812	<0.001	–

*p < 0.05, **p < 0.01, ***p < 0.001. ^bVIF: Variation Inflation Factor.

Table 2 shows that based on multiple linear regression models, only factor 1 was found to be significantly associated with safety performance. So, only the occupational factor domain was considered for the design of safety interventions. As four risk factors were clubbed under factor 1 (poor working conditions, poor safety environment, poor job satisfaction, and high job stress), they were used to formulate the interventions focusing on the occupational dimension.

Table 3 shows the hierarchies of needed improvements and corresponding necessary interventions. Table 4 shows their normalized importance ratings and normalized priority weights, respectively. Table 5 shows the interventions identified and ordered according to their normalized priority weight.

Table 3

Needed improvements and necessary safety intervention

Underlying significant factor	Variables	Needed improvements	Necessary safety interventions
Factor 1 (Occupational factor \|β\| = 2.413)	Poor working condition (λ= 0.486)	Improve lighting	Make proper layout for lighting as per regulations
		Improve ventilation	Review ventilation network periodically and make changes as per requirement
		Improve water logging condition	Install and maintain proper drainage and pumping system
	Poor safety environment (λ= 0.641)	Improve managements involvement vis-à-vis safety	Linking performance related to safety with career progression and incentives
		Improve workers involvement vis-à-vis safety	Imparting safety training
		Make workers suitable for their jobs	Provide safety training and assess workers safety awareness
		Eliminate hazard from workplace	Develop or formulate risk management plan
		Improve Safe working procedures (SOPs)	Proper compliance of safety inspection regarding SOPs
	Poor job satisfaction (λ= 0.767)	Ensure proper career progression for workers	Formulate promotion policy and implement it
		Ensure workers get appreciation for their good work	Sensitize the management about the issue
		Motivate workers	Provide perks and benefits suitable to their jobs
		Ensure workers behave cordially with their co workers	Develop team spirit among workers
		Ensure workers have freedom of decision making	Encouragement from management regarding on-the-job decision making
	High job stress (λ= 0.750)	Impart confidence to workers for doing hard and dangerous work.	Provide job specific training
		Ensure adequate salary for workers	Provide all benefits as mandated by government
		Reduce performance pressure on workers	Give workers achievable targets

λ: factor loadings.

Table 4

Calculation of normalized importance rating of needed improvements and normalized priority weights of necessary safety interventions

Needed improvements	\|β_i\|: Absolute value of Regression coefficient	λ_i: Factor loadings	ASi: Average Score	R_i: Importance rating [ R _i = β_i × λ_i × AS_i]	r_i: Normalized Importance rating [r_i = R _i/∑ R _i]
Improve lighting		0.486	2.22	2.29	5.12
Improve ventilation		0.486	2.57	2.65	5.93
Improve water logging condition		0.486	2.18	2.25	5.02
Improve managements involvement vis-à-vis safety		0.641	1.86	2.55	5.69
Improve workers involvement vis-à-vis safety		0.641	2.13	2.91	6.50
Make workers suitable for their jobs		0.641	2.46	3.36	7.51
Eliminate hazard from workplace		0.641	1.69	2.31	5.16
Improve Safe working procedures (SOPs)	2.143	0.641	1.97	2.70	6.02
Ensure proper career progression for workers		0.767	2.13	3.49	7.78
Ensure workers get appreciation for their good work		0.767	1.69	2.76	6.17
Motivate workers		0.767	2.08	3.40	7.60
Ensure workers behave cordially with their co workers		0.767	1.25	2.03	4.54
Ensure workers have freedom of decision making		0.767	1.36	2.22	4.97
Impart confidence to workers for doing hard and dangerous work.		0.750	1.94	3.10	6.93
Ensure adequate salary for workers		0.750	2.22	3.55	7.94
Reduce performance pressure on workers		0.750	1.96	3.15	7.03
Necessary safety interventions		EO_i: Average expert opinion score	W_i: Priority weight [w_i = r_i × EO_i]	ω_i: Normalized priority weight [ω_i = W_i/∑W_i]
Make proper layout for lighting as per regulations		5.81	29.74	5.12
Review ventilation network periodically and make changes as per requirement		6.18	36.64	6.31
Install and maintain proper drainage and pumping system		5.90	29.61	5.09
Linking performance related to safety with career progression and incentives		5.54	31.52	5.42
Imparting safety training		5.63	36.59	6.30
Provide safety training and assess workers safety awareness		5.72	42.95	7.39
Develop or formulate risk management plan		5.81	29.97	5.16
Proper compliance of safety inspection regarding SOPs		5.81	34.97	6.02
Formulate promotion policy and implement it		5.45	42.40	7.30
Sensitize the management about the issue		5.90	36.40	6.26
Provide perks and benefits suitable to their jobs		5.81	44.15	7.60
Develop team spirit among workers		5.00	22.70	3.89
Encouragement from management regarding on-the-job decision making		5.72	28.42	4.89
Provide job specific training		6.36	44.07	7.58
Provide all benefits as mandated by government		6.27	49.78	8.57
Give workers achievable targets		5.81	40.84	7.03

Table 5

Priority ranks of necessary safety interventions

Priority rank	Necessary safety interventions	Normalized Priority weight (ω_i)
1	Provide all benefits as mandated by government	8.57
2	Provide perks and benefits suitable to their jobs	7.60
3	Provide job specific training	7.58
4	Provide safety training and assess workers safety awareness	7.39
5	Formulate promotion policy and implement it	7.30
6	Give workers achievable targets	7.03
7	Review ventilation network periodically and make changes as per requirement	6.31
8	Imparting safety training	6.30
9	Sensitize the management about the issue	6.26
10	Proper compliance of safety inspection regarding SOPs	6.02
11	Linking performance related to safety with career progression and incentives	5.42
12	Develop or formulate risk management plan	5.16
13	Make proper layout for lighting as per regulations	5.12
14	Install and maintain proper drainage and pumping system	5.09
15	Encouragement from management regarding on-the-job decision making	4.89
16	Develop team spirit among workers	3.90

4 Discussion

This study proposed an approach that helped to determine the most effective interventions and prioritize them to promote safety performance of underground coal mine workers using both workers’ self-assessed occupational risk factors and workers’ safety performance assessed by supervisors. Xia et al. reported that the workers’ safety performance assessment by the workers themselves was less valid than that by their supervisors, although the difference may be attenuated by the supervisor’s lack of experience [7]. In our study, the assessment of the supervisors may rather be correct as all of them had more than 20 years of experience.

The present study reported that only the occupational dimension was associated with safety performance. So, a simplified approach was used to formulate the framework to improve the safety performance of workers only considering the occupational dimension. Multivariate analysis revealed that the occupational dimension was the only factor significantly associated with safety performance. So, to improve safety performance, issues related to occupational factor need to be addressed. Moreover, occupational factor was a combination of four risk factors: poor working conditions, poor safety environment, poor job satisfaction, and high job stress. So, the aforementioned risk factors were further disintegrated to formulate, select, and prioritize safety interventions. Our approach provided a wide range of safety interventions related to the occupational dimension, which were according to their priority order:

‘Provide all benefits as mandated by government’: Workers should get fair compensation as mandated by the government for the work they do. If workers are not properly compensated, it might give rise to mental and financial stress, and the person working under stress is more likely to make mistakes and experience occupational injuries. Rautiainen et al. suggested that financial incentives can reduce occupational injuries [31]. According to Shirali et al., it was found that the amount of money that companies pay to each worker for safety has a strong impact on the overall safety outcomes [32].

‘Provide perks and benefits suitable to their jobs’: Workers should get compensated according to the degree of difficulties and hardness of the job, or else they might not be satisfied with the job and will be more prone to experience occupational injuries [33].

‘Provide job-specific training’: Job-specific training is essential in order to prepare workers to perform arduous tasks and cope with high-risk jobs.

‘Provide safety training and assess workers’ safety awareness’: In order to make workers suitable for their job, safety training should be imparted to workers and monitored during the job to determine the effectiveness of the training. It has also been observed that safety training is essential to reduce occupational accidents and injuries [21, 34]. National Institute of Occupational Safety and Health (NIOSH) published reports on the effectiveness of safety training, and reports suggested that safety training vastly reduces occupational accidents and injuries [35, 36]. However, the effectiveness of different training methods and methodology requires separate investigation.

‘Formulate promotion policy and implement it’: getting timely promotion boosts workers’ morale and induces feelings of job satisfaction.

‘Give workers achievable targets’: Workers should be given targets that can be achieved in a timely manner without physically straining the workers.

‘Review ventilation network periodically and make changes as per requirement’: As ventilation capacity is determined following the stipulations, poorly ventilated working faces are primarily results of poor network design and leaks in the airway. These issues should be addressed through regular reviews of the ventilation network.

‘Imparting safety training’: Imparting safety training is critical in improving the involvement of workers in affairs of safety.

‘Sensitize the management about the issue’: Getting appreciation from management is crucial for having job satisfaction, and sometimes management remains aloof from this issue. Therefore, management should be made aware regularly of the issue. Hopkins et al. found that praise plays a role in ensuring safe behavior [37].

‘Proper compliance of safety inspection regarding SOPs’: Continuous improvement of safe working procedures is also essential as new technology and machinery are being introduced regularly in underground mines. Experts have suggested that compliance with safety inspection criteria vis-à-vis safe working procedures is a suitable method to improve safe working procedures continuously.

‘Linking performance related to safety with career progression and incentives’ and ‘Develop or formulate risk management plan’: These two interventions were suggested to improve the safety environment. The monetary incentive is a practical motivation to follow safety protocols. A risk management plan issues clear instructions, which help immensely during an adverse situation, as individuals can follow their roles to minimize the risk.

‘Make proper layout for lighting as per regulations’ and ‘Install and maintain proper drainage and pumping system’: Proper lighting and proper drainage system are essential in the underground coal mine in order to have a better working condition, and thus improve safety performance.

‘Encouragement from management regarding on-the-job decision making’: Freedom of decision making impart the feeling of job satisfaction in workers.

‘Develop team spirit among workers’: Though it has the lowest priority, it is imperative to build team spirit among workers because, in a hazardous workplace like an underground coal mine, workers look out for each other, train, and inspire each other in matters of safety.

4.1 Strengths and limitations

One of the strengths of this study is that the data appeared to be well suited for exploratory factor analysis, as indicated by the KMO and Bartlett’s test results. Additionally, the use of interviews with workers as a data collection method is a widely accepted and efficient tool in the field of occupational health and injury research [38, 39]. The researchers took measures to ensure data quality, such as conducting interviews in private to prevent supervisor influence on responses. The long work experience of the experts involved in the study, some with over 25 years in the mining sector, adds to the credibility of the findings. Furthermore, the inclusion of workers who had experienced injuries provides more valid and informative insights compared to those who had not experienced injuries. The use of an indirect questioning procedure and efforts to minimize recall bias further strengthen the study.

One potential weakness of the study is the possibility of recall bias, although it is expected to be minimal as the workers had been employed in their respective workplaces for an extended period and are knowledgeable about their job and occupational hazards. Additionally, the workers were informed that their responses would not be divulged to their supervisors and mine management. While efforts were made to collect data in a double-blind manner to increase validity, other sources of bias may still exist. Furthermore, present study only included individual and occupational factors, future research may include other factors such as workplace safety-related policies, leadership, communication, safety climate, personal protective equipment (PPE), individual employees’ vulnerability or susceptibility to environmental hazards and other relevant factors. Overall, these limitations should be taken into consideration when interpreting the study’s findings.

5 Conclusion

The present study among underground coal mine workers showed that workers’ safety performance as perceived by their supervisors can be an important indicator for the state of safety in the organizations. It showed that a wide range of occupational risk factors (including poor working conditions, poor safety environment, poor job satisfaction, high job stress) play a crucial role in determining workers’ safety performance. To effectively enhance workers’ safety performance by mitigating occupational risk factors, our systematic approach helped to identify and prioritize the most impactful safety interventions. As a result, the following interventions were determined, listed in order of priority: provide workers’ fair compensation, job-specific/safety training, promotion policy, achievable targets, periodic review of ventilation network and changes, sensitize the management, relevant perks/benefits, safety training/awareness, workplace lighting, associate safety performance to promotion, proper draining and pumping system, on job decision making, and develop team spirit.

Effective safety enhancement relies on the careful allocation of finite organizational resources. The framework proposed in this study empowers mine management to prioritize and select safety interventions, and maximizing resource utilization. By utilizing this framework, organization can improve safety performance and promote a safer workplace.

Ethics statement

This research exclusively involved interviews with human participants and did not include any procedure involving experiments on human subjects. Nonetheless, it received approval from the Research and Development Cell at the Indian Institute of Technology Kharagpur (no ethical approval number assigned).

Informed consent

Prior to conducting the study, the participants’ consent was obtained. They were informed about the aim, objective and methods and consequently publication of the study. Their identities were safeguarded throughout the data processing and all records were securely disposed of upon study completion. Consequently, there is no way to discern the subjects’ identities. Additionally, neither the participants’ names nor mines’ names and location are referenced anywhere in the manuscript. Therefore, the privacy of participants was not infringed.

Conflict of interest

The authors have no competing interests to declare that are relevant to the content of this article.

Footnotes

Acknowledgments

The support from the management, supervisors, staff, and workers of the mines, and experts from the mining industry and academia is greatly appreciated.

Funding

None to report.

Appendix

Table 6

Variable	Needed Improvements	Questionnaire items
Poor Working Condition	Improve lighting	Is there sufficient arrangement of lights at convenient places in the mine?*
		Is inconvenience caused due to bad light at the work places?
	Improve ventilation	Does one feel hot at workplaces?
		Is sweating a common phenomenon at the time of work?
		Is there a feeling of humidity below ground?
		Is sufficient amount of air available at the time of work?*
		Is there good ventilation in the mine (absence of noxious gases, dust etc.)?*
	Improve water logging condition	Are the roadways watery causing slips and fall accidents?
		Is pumping of water done properly in the mine?*
Poor Safety Environment	Improve managements involvement vis-à-vis safety	Is taking risks, short-cuts or unsafe behaviour acceptable to management in your workplace?
		Do supervisors ignore risk-taking behaviour in your workplace?
		Do supervisors actively discourage unsafe behaviour?*
		Do you receive positive feedback from supervision for following safe work practices?*
		Are safety issues you raise with our supervisor or foreman adequately dealt with?*
		Are you discouraged from reporting or bringing safety issues to the attention of management?
		Have you been penalized for not performing a task that you considered unsafe?
		Is your employer serious about safety and health?*
		Are safety bulletin and safety incident reports made available for you to read?*
	improve workers involvement vis-à-vis safety	Have you observed any risk taking by other employees at your mine?
	Make workers suitable for their jobs	Have you been adequately trained to perform your assigned tasks underground?*
	Eliminate hazard from workplace	Are hazards eliminated promptly and effectively in your workplace?*
	Improve Safe working procedures (SOPs)	Are the procedures used underground the same as the ones written in the latest regulation?
	Ensure workers get appreciation for their good work	Does your officer appreciate your competency and quality? *
		Do other people give respect to your work?*
		Do you think the workers get right behaviour as they expect from their officers?*
		Do you think this work is a source of respect for a man like you?*
	Motivate workers	Are you satisfied with the condition of your workplace?*
		Do you think you have got right job for yourself?*
		Do you think your work is difficult and arduous?
		If you are given a chance would you like to leave this company and join in other Company?
	Ensure workers behave cordially with their co workers	Do your co-workers helpful and co-ordeal to you?*
	Ensure workers have freedom of decision making	Are you allowed to take decision independently regarding your job?*
High Job Stress	Impart confidence to workers for doing hard and dangerous work.	Do you have to work hard in mining job?
		Is the mining work boring and monotonous?
		Is your work full of danger?
		Is your work full of danger and difficult?
		Do you feel your service is a burden to you?
	Ensure adequate salary for workers	Do you have the salary justified with your labour?
Poor Job Satisfaction	Ensure proper career progression for workers	Are the promotions to workers given with justice and honesty?*
	Reduce performance pressure on workers	Do you have to work in hurry due to excess load of work?
		Do you feel you have the responsibility for the development of the mine?
		Are you happy with the working situation of the mine?

References

World Statistics: International Labour Organisation [Available from: https://www.ilo.org/moscow/areas-of-work/occupational-safety-and-health/WCMS 249278/lang–en/index.htm

MSHA. Mine Safety and Health Administration (MSHA): United States Department of Labor; 2022 [Available from: https://www.msha.gov/data-reports/fatality-reports/search.

Department of Mines IRaS. Safety Performance in the Western Australian Mineral Industry, Accident and Injury Statistics 2017-2018. East Perth; 2019.

DGMS. Standard Note 01.01.2020. Dhanbad: Directorate General of Mines Safety; 2020.

Takala

, Hämäläinen

, Saarela

, Yun

, Manickam

, Jin

, et al. Global estimates of the burden of injury and illness at work in 2012. Journal of Occupational and Environmental Hygiene. 2014;11(5):326–37.

Hine

, Lewko

, Blanco

. Alignment to workplace safety principles: an application to mining. Journal of Safety Research. 1999;30(3):173–85.

Xia

, Griffin

, Wang

, Liu

, Wang

. Is there agreement between worker self and supervisor assessment of worker safety performance? An examination in the construction industry. Journal of Safety Research. 2018;65:29–37.

Griffin

, Neal

. Perceptions of safety at work: a framework for linking safety climate to safety performance, knowledge, and motivation. Journal of Occupational Health Psychology. 2000;5(3):347.

, Taksa

, Jia

. Influence of management practices on safety performance: The case of mining sector in China. Safety Science. 2020;132:104947.

10.

Burke

, Sarpy

, Tesluk

, Smith-crowe

. General safety performance: A test of a grounded theoretical model. Personnel Psychology. 2002;55(2):429–57.

11.

Cornelissen

, Van Hoof

, De Jong

. Determinants of safety outcomes and performance: A systematic literature review of research in four high-risk industries. Journal of Safety Research. 2017;62:127–41.

12.

Ajith

, Ghosh

, Jansz

. Contributing effects of individual characteristics, behavioural and job-related factors on occurrence of mining-related injuries: A systematic review. Work. 2022;71(1):87–117.

13.

Schwatka

, Rosecrance

. Safety climate and safety behaviors in the construction industry: The importance of co-workers commitment to safety. Work. 2016;54(2):401–13.

14.

Senapati

, Bhattacherjee

, Chatterjee

. Causal relationship of some personal and impersonal variates to occupational injuries at continuous miner worksites in underground coal mines. Safety Science. 2022;146:105562.

15.

Karimpour

, Jahani

, Shirali

. The relationship between safety attitude and safety performance in the petrochemical industry: The mediating role of safety climate. Work. 2021;68(2):507–17.

16.

Chu

, Fu

, Liu

. Organization is also a “life form”: Organizational-level personality, job satisfaction, and safety performance of high-speed rail operators. Accident Analysis & Prevention. 2019;125:217–23.

17.

Mullan

, Smith

, Sainsbury

, Allom

, Paterson

, Lopez

A-L

. Active behaviour change safety interventions in the construction industry: A systematic review. Safety Science. 2015;79:139–48.

18.

Verbeek

, Ivanov

. Essential occupational safety and health interventions for low-and middle-income countries: an overview of the evidence. Safety and health at work. 2013;4(2):77–83.

19.

Sámano-Ríos

, Ijaz

, Ruotsalainen

, Breslin

, Gummesson

, Verbeek

. Occupational safety and health interventions to protect young workers from hazardous work–A scoping review. Safety Science. 2019;113:389–403.

20.

Oyewole

, Haight

, Freivalds

, Cannon

, Rothrock

. Statistical evaluation and analysis of safety intervention in the determination of an effective resource allocation strategy. Journal of Loss Prevention in the Process Industries. 2010;23(5):585–93.

21.

Verma

, Maiti

, Boustras

. Analysis of categorical incident data and design for safety interventions using axiomatic design framework. Safety Science. 2020;123:104557.

22.

Babur

, Cevikcan

, Durmusoglu

. Axiomatic Design for Lean-oriented Occupational Health and Safety systems: An application in shipbuilding industry. Computers & Industrial Engineering. 2016;100:88–109.

23.

Kothari

. Sampling fundamentals. In: Research Methodology. 3rd ed. New Delhi: New Age International; 2004. p. 152-183.

24.

Ghosh

, Bhattacherjee

, Chau

. Relationships of working conditions and individual characteristics to occupational injuries: a case-control study in coal miners. Journal of Occupational Health. 2004;46(6):470–80.

25.

Field

. Discovering statistics using IBM SPSS statistics: sage; 2013.

26.

Hartas

. Educational research and inquiry: Qualitative and quantitative approaches: Bloomsbury Publishing; 2015.

27.

Tabachnick

, Fidell

, Ullman

. Using multivariate statistics: Pearson Boston, MA; 2007.

28.

Bartlett

. A note on the multiplying factors for various χ 2 approximations. Journal of the Royal Statistical Society Series B (Methodological). 1954:296–8.

29.

Cattell

. The scree test for the number of factors. Multivariate Behavioral Research. 1966;1(2):245–76.

30.

Kaiser

. The application of electronic computers to factor analysis. Educational and Psychological Measurement. 1960;20(1):141–51.

31.

Rautiainen

, Lehtola

, Day

, Schonstein

, Suutarinen

, Salminen

, et al. Interventions for preventing injuries in the agricultural industry. Cochrane Database of Systematic Reviews. 2008(1).

32.

Shirali

, Salehi

, Savari

, Ahmadiangali

. Investigating the effectiveness of safety costs on productivity and quality enhancement by means of a quantitative approach. Safety Science. 2018;103:316–22.

33.

Kudo

, Belzer

. The association between truck driver compensation and safety performance. Safety Science. 2019;120:447–55.

34.

Ofori

, Aram

, Saalidong

, Gyimah

, Niyonzima

, Mintah

, et al. Exploring new antecedent metrics for safety performance in Ghana’s oil and gas industry using partial least squares structural equation modelling (PLS-SEM). Resources Policy. 2023;81:103368.

35.

Cohen

, Colligan

, Sinclair

, Newman

, Schuler

. Assessing occupational safety and health training. Cincinnati, OH: National Institutes of Health. 1998:1–174.

36.

Robson

, Stephenson

, Schulte

, Amick

, Chan

, Bielecky

, et al. A systematic review of the effectiveness of training & education for the protection of workers. Toronto: Institute for Work & Health. 2010;2127.

37.

Hopkins

, Conard

, Dangel

, Fitch

, Smith

, Anger

. Behavioral technology for reducing occupational exposures to styrene. Journal of Applied Behavior Analysis. 1986;19(1):3–11.

38.

Ghosh

, Bhattacherjee

. Predictors of occupational injuries among coal miners: causal analysis. Mining Technology. 2007;116(1):16–24.

39.

Senapati

, Bhattacherjee

, Chau

. Associations of job-related hazards and personal factors with occupational injuries at continuous miner worksites in underground coal mines: a matched case-control study in Indian coal mine workers. Industrial Health. 2020;58(4):306–17.