Safety and Health Maturity Levels as Indicators of Accident Reduction in Indonesia’s Nickel Mining Sector

Introduction

Nickel has been designated as a critical mineral by the Indonesian Government, due to its role in supporting the energy transition and the electric vehicle battery industry. As global demand increases, Indonesia’s nickel production also continues to increase, accompanied by increasingly complex operational risks, considering that mining is an industry that involves the use of complicated and complex machines, equipment, and processes as well as various kinds of worker activities that take place in a dynamic and challenging environment.¹ Nickel mining operations are also very dynamic, which can cause most hazards to go undetected.² In addition, workers in the nickel mining industry are often exposed to dust and nickel compounds that can cause various diseases, including chronic bronchitis, asthma, and hearing loss.³ However, data from 2024 show that despite production growth, there has been no significant improvement in Occupational and Health (OHS) outcomes: 51 nickel mining accidents were recorded, including 12 fatal incidents. In addition, there were 8 deaths due to work-related illness.⁴ To address these risks, the MEMR enacted Regulation No. 10.K/MB.01/DJB.T/2023, requiring all mining companies to assess their Mining Safety and Health Performance Maturity Level (MSHPML) which is integrated with the Mining Safety Management System. This requirement is tied to the approval of the Work Plan and Budget. The MSHPML assessment is conceptually grounded in the Safety Maturity Model, a progressive framework that evaluates how deeply safety culture is embedded within an organization’s operational systems. Based on previous studies, the OHS Maturity As-sessment will provide benefits to companies that implement it. Organizations that assess the Safety Maturity Level can identify weaknesses in operational safety culture and practices that lead to targeted interventions that improve overall safety performance.⁵ Research conducted by Foster (2013) also states that the Safety Maturity Model is a strategic framework used to assess and improve safety practices in high-risk industries, especially mining industry/Furthermore.⁶ Research about the nature of safety culture in 2000 concluded that continuous measurement and improvement of Safety Maturity can also lead to sustainable practices that integrate safety into the organizational struc-ture, thereby ensuring long-term success.⁷

Meanwhile, after one year of the regulation being enforced, based on data from the MEMR, in 2024 there were still 42 mining accidents in nickel mining companies, including 7 accidents resulting in workers’ death.⁸ In addition, in 2024 there were 8 Death Cases due to Worker Illnesses in nickel mining companies. This highlights a dis-crepancy between the intended regulatory outcomes and actual field performance. Based on this background, a study is needed to examine the relationship between MSHPML and OHS performance lagging indicators in nickel mining companies in Indonesia in order to increase the effectiveness of its implementation to prevent OHS cases. While the maturity-based approach to safety management has been widely applied in high-risk sectors such as energy, construction, and healthcare, there remains a lack of quantitative studies assessing the effectiveness of MSHPML in improving both OHS outcomes, especially in the context of Indonesia’s nickel mining industry. Previous related research was conducted at the Mining industry in 2019 who examined the relationship between safety culture maturity levels and safety performance in the mining industry in Ghana.⁹ It was found that mines with lower incidence rates consistently had higher safety culture maturity scores for the elements than mines with higher incidence rates. In Construction Industry revealed that in Construction Industries, organizations operating at a higher level of safety culture maturity demonstrate fewer incidents, as maturity strengthens proactive hazard identification and control practices.¹⁰ Meanwhile, re-vealed that higher levels of maturity correlate with reduced accident rates and improved process safety performance.¹¹

Therefore, this study addresses that gap by analyzing the relationship between MSHPML as a predictor of key OHS outcomes. The findings aim to provide empirical evidence that maturity-based safety systems not only reduce OHS cases but also offer a strategic foundation for enhancing national OHS governance and performance evaluation frameworks. The hypothesis of this study is that the higher the MSHPML score, the better the achievement of the nickel mining company’s OHS performance lagging indicators. This is based on the Safety Maturity theory developed by Hudson (2007), Parker et al. (2006), and Foster (2013), which states that increasing organizational maturity in OHS aspects will have an impact on reducing incidents and increasing OHS integration in daily work practices.^5,6,12

Methods

This study uses a quantitative approach with a cross-sectional design to measure the relationship between the MSHPML and the OHS lagging indicators in nickel mining companies in Indonesia. The purpose of this approach is to identify associative rela-tionships based on current year data (2024). The population in this study were all holders of nickel mining companies in Indonesia, which have a 2024–2026 Work Plan and Budget approved by the MEMR. The inclusion criteria were nickel mining companies that had reported MSHPML data and OHS lagging indicators for 2024 through the formal mechanism determined by the MEMR. The sample was calculated using the Finite Population Correction formula, so that a total of 77 companies were obtained that met the inclusion criteria as final respondents, spread across the provinces of Maluku, North Maluku, Papua, West Papua, South Sulawesi, Central Sulawesi, Southeast Sulawesi. The unit of analysis in this study is the nickel mining company as an operational entity. The observation unit is the MSHPML score and the OHS Lagging Indicators data from each company based on the 2024 reporting.

The Dependent Variables in this study were Occupational Safety Performance In-dicators, including the Number of Mining Accidents, Accident Frequency Rate (FR), and Accident Severity Rate (SR), as well as Occupational Health Performance Indicators, including: Number of Occupational Diseases, Number of Death Cases due to Worker Illnesses, Morbidity Frequency Rate (MFR), and Absence Severity Rate (ASR). Mean-while, the Independent Variable is the MSHPML, which consists of 4 main indicators: Worker Participation, Responsibility of Working Unit Leaders, Analysis and Statistics of Mining Safety Cases, and Risk Control Implementation Efforts. Each indicator is given a Basic, Reactive, Planned, Proactive, or Resilient category in accordance with the government regulation. The total MSHPML score is calculated as the sum of all indicator scores, and is also analyzed categorically. The definition of the Dependent and Independent Variables refers to laws and regulations, namely the MEMR Regulation Number 1827 K/30/MEM/2018 and Director General of Minerals and Coal Decree Number 10.K/MB.01/DJB.T/2023.

The study used data from the MEMR, which has granted Preliminary Research Permit and Initial Data Collection to Universitas Indonesia through the Letter of the Director of Engineering and Environment number B-3957/MB.07/DBT.KP/2025. The validity and reliability of the data were maintained through a data verification and validation process carried out together with the Mine Inspector of MEMR. The data was taken from the official reporting system that had gone through an internal evaluation process by the MEMR.

The data filtering stage was carried out for a total of 258 initial data. The exclusion criteria used were companies with duplicate data, not in accordance with the MSHPML assessment format, or not meeting statutory provisions. For these data, data duplication was removed (total data deleted: 9), data was removed from outside the Inclusion Criteria (total data deleted: 97), data was removed from companies that did not meet the assessment standards according to statutory regulations according to data from the MEMR (total data deleted: 75), so that the total sample size was 77 nickel mining companies. The data filtering and verification process in this study was directly assisted by the Mine Inspector of the Mine Inspector of MEMR to ensure data validity.

In this study, Data Processing and Analysis were carried out using a quantitative approach through three levels of analysis: univariate, bivariate, and multivariate, equipped with a series of assumption tests to ensure the validity of the statistical model used. Univariate analysis was carried out with descriptive analysis (proportion, average, mean, standard deviation and confident interval). In addition, Bivariate and Multivari-ate Analysis were carried out to see the relationship between MSHPML and OHS Lagging indicators (Number of Mining Accidents, Occupational Disease Death Cases, Occupational Diseases, FR, SR, MFR, ASR). Poisson Regression was conducted for the number of accidents and illnesses, Spearman Correlation for ordinal relationships, and Linear Regression for continuous variables such as FR, SR, MFR, and ASR. Normality, multicollinearity, and heteroscedasticity assumption tests were conducted on the linear model before final interpretation. This method allows empirical evaluation of the effectiveness of MSHPML policies from the aspect of occupational safety and health outcomes quantitatively, with a multi-level analysis approach that combines managerial evaluation and field outcomes.

Results

This study analyzed 77 nickel mining companies in Indonesia, the majority of which are located in Southeast Sulawesi, Central Sulawesi, and North Maluku. Most companies are classified as medium-scale, based on their production capacity outlined in the 2024–2026 Work Plan and Budget.

Based on Table 1, most companies have not yet reached the minimum maturity standard as regulated by the government. Specifically, 36.4% of companies are categorized at the Basic Level and 44.2% at the Reactive Level. Only 18.2% of companies have reached the Planned or Proactive levels, and none have achieved the Resilient Level.

Building on this distribution, further breakdown by individual indicators reveals varied levels of maturity across key dimensions. As shown in Table 2, the Worker Participation indicator records the lowest achievement, with 32 companies (41.6%) still at the Basic level and 27 companies (35.1%) at the Reactive level. Only 23.4% of companies have reached the desired Planned or Proactive level in this domain. Similar patterns were observed in the Working Unit Leadership indicator, with 43% at Basic and 35% at Reactive levels. In contrast, the Analysis and Statistics indicator show a stronger performance, with 68.8% of companies at the Reactive level and only 14.3% still in the Basic category. The Risk Control Implementation Efforts indicator shows a more even distribution: 27.3% Basic, 42.9% Reactive, and 30% Planned or Proactive. Table 2 presents the descriptive statistics for the Occupational Health and Safety (OHS) lagging indicators. The average FR was 0.050 (95% CI: 0.029–0.070), and the average SR was 69.225 (95% CI: 39.17–99.28). For health-related indicators, the MFR averaged 142.236, while the ASR averaged 170.442, both with high standard deviations indicating data variability.

The relationship between MSHPML and the number of mining accidents was tested using Poisson Regression ( Table 3). All four MSHPML indicators—Worker Participation, Working Unit Leadership, Analysis and Statistics, and Risk Control Implementation Efforts showed significant negative coefficients (p < 0.05), confirming that higher maturity levels are associated with fewer accident cases.

Spearman correlation analysis ( Table 4) further supports this finding. All MSHPML indicators show a significant negative correlation with the number of accidents (ρ ranging from -0.333 to -0.376, p < 0.01). However, no significant correlations were found between the indicators and the number of occupational disease cases or deaths due to worker illnesses.

Pearson correlation ( Table 5) revealed a moderate, statistically significant negative relationship between the overall MSHPML score and both FR (r = -0.475, p = 0.001) and SR (r = -0.394, p < 0.001). These findings were confirmed by simple linear regression ( Table 7), which showed that MSHPML scores significantly predict reductions in FR and SR (p = 0.001), with R² values of 0.226 and 0.156 respectively.

However, the relationship between MSHPML and health indicators (MFR and ASR) was not significant. Pearson correlation results ( Table 6) and regression models ( Table 7 and Table 8) yielded p-values > 0.05 and low R² values (<0.03), indicating weak or no association. This suggests that the MSHPML framework may not yet adequately address occupational health outcomes.

Notably, multiple linear regression analysis ( Table 8) showed that only the Risk Control Implementation Efforts indicator had a significant effect on SR (p = 0.046), reinforcing its importance in reducing the severity of incidents. Other indicators, including Worker Participation and Working Unit Leadership, did not demonstrate a statistically significant impact in multivariate models.

Discussion

The results of this study indicate that the majority of nickel mining companies in Indonesia (80.60%) have not yet achieved the minimum standard of the MSHPML, specifically the Planned Level, as mandated by government regulation. Among the four assessed MSHPML indicators, Worker Participation consistently shows the lowest level of achievement. This suggests persistent challenges in involving workers in OHS programs. Such low engagement may be rooted in organizational cultures that do not actively support or incentivize employee involvement. An unsupportive organizational culture that does not support active employee participation can be a major barrier in workplace health promotion programs.¹³ This condition poses a serious risk for fu-ture OHS incidents, particularly in the mining industry where previous research has established that individual behavior and human factors are critical components in shaping safety culture and reducing accident likelihood.¹⁴

In addition, the study found that 78% of company management has not achieved the Planned Level for the Working Unit Leadership indicator. This suggests that managerial commitment and involvement in OHS efforts remain largely compliance-driven. This superficial form of engagement may have downstream effects on worker participation, as managerial leadership plays a vital role in cultivating a psychologically safe and collaborative environment. Organization leaders have a critical role to play in creating an open, collaborative, and non-punitive environment that enables workers to be the strongest link in responding to risks.¹⁵

This study found that the MSHPML, an index reflecting the maturity of safety and health management systems, is significantly associated with a reduction in mining accidents, as well as with lower FR and SR. These findings suggest that higher MSHPML scores are associated with the decrease in accident frequency and severity rates. These findings offer empirical validation of the central hypothesis, which demonstrated that improvements in safety management maturity and organizational safety culture are strongly linked to better safety performance in high-risk industries.^16–18

Beyond reducing accident frequency, an increase in MSHPML is also associated with a reduction in accident severity. This may reflect the enhanced capacity of mitigative safety barriers in more mature systems that are typically more developed in organizations with higher safety maturity. The presence of robust mitigative controls, such as emergency response procedures, containment systems, and escalation prevention protocols, can significantly reduce the impact of accidents when they do occur. As emphasized by Ricci et al. (2024), emergency response is a procedural safety barrier of paramount importance for the mitigation and the prevention of escalation.

The results of Poisson Regression and Spearman Correlation analysis further confirm a strong and negative association between MSHPML and key safety indicators. Notably, the indicators of Analysis & Statistics, Risk Control Implementation Efforts, and Worker Participation demonstrated the strongest correlations. These findings re-nforce the critical role of proactive data management, systematic risk control, and workforce engagement in enhancing organizational safety maturity and reducing both the frequency and severity of incidents in the mining sector.

Multivariate analysis confirmed that the Risk Control Implementation Efforts indicator was the most significant predictor of reduced accident frequency and severity. This highlights the critical role of proactive hazard mitigation in improving safety performance. The findings consistently demonstrate that integrating maturity evaluations into safety culture assessments enables companies to sustain lower incident rates.¹⁹ This effect is largely due to the presence of stronger systemic controls and more effective behavior management. Furthermore, the results align with the emphasis on proactive risk control as central to reducing incidents.¹²

On the other hand, this study found no significant relationship between the MSHPML and occupational health indicators, including the number of reported occupational disease cases, MFR, and ASR. These results suggest that the current MSHPML framework may not sufficiently reflect or incorporate the domain of Occupational Health, as its components remain more heavily oriented toward safety outcomes. Several explanations may account for this disconnect. First, occupational diseases are frequently underreported or undetected, particularly those with long latency periods. Underreporting of occupational diseases is common due to factors such as delayed onset, multifactorial etiology, and limited clinical suspicion.²⁰ Employers may intentionally underreport occupational illnesses due to concerns over legal liability or increased insurance costs.²¹ Second, the lack of integration of health-specific indicators into the MSHPML framework may also limit its ability to capture variations in performance. While safety practices are often visible and immediately measurable, health-related outcomes typically require longitudinal observation, medical diagnosis, and robust surveillance systems, which are often lacking in mining operations. These findings underscore the need for regulatory authorities to re-evaluate and expand the MSHPML framework to better integrate occupational health considerations.

This study also has limitations in the nature of the cross-sectional design, so it cannot conclude a causal relationship directly. In addition, the use of secondary data from the reporting system may contain reporting bias. The MSHPML scores of nickel mining companies was obtained by the MEMR through reports based on self-assessments. As is known, self-report measures have however been found to be susceptible to social desirability bias and thus threaten the validity of the study.²² However, this study contributes to empirical testing of the effectiveness of the MSHPML model adopted in national policies. This is also the first study to integrate MSHPML assessments with quantitative Occupational Health lagging indicators in the nickel mining subsector.

Future research is recommended to adopt a qualitative approach to complement the findings of this study regarding the relationship between MSHPML and OHS lagging indicators in nickel mining companies in Indonesia. A qualitative inquiry may provide deeper insight into the contextual and organizational factors behind the numerical trends. Moreover, a longitudinal research design is also recommended to observe the impact of changes in MSHPML scores over time on key safety indicators such as FR and SR. Such an approach would allow researchers to assess whether improvements in maturity levels produce sustained and measurable impacts on OHS performance.

Ethical considerations

This research did not involve any human participants or animals. Ethical review and approval were waived for this study due to the exclusive use of secondary, aggregated, and de-identified company performance data obtained from a governmental regulatory body (MEMR), which does not constitute human subjects research.