Front. Public HealthFrontiers in Public HealthFront. Public Health2296-2565Frontiers Media S.A.10.3389/fpubh.2025.1611106Public HealthOriginal ResearchSustainable safety practices and hazard management in the oil and gas industry: an HSE perspectiveJamilZara1*NordinShahrina1MirajMohammad2*†AlqahtaniMazen3†ShaikRiyaz Ahamad4†AkhterShamim5Nizam IshaAhmad Shahrul11Department of Management and Social Sciences, Institute of Smart and Sustainable Living, University Teknologi PETRONAS, Bandar Seri Iskandar, Malaysia2Department of Physical Therapy and Health Rehabilitation, College of Applied Medical Sciences, Majmaah University, Al Majmaah, Saudi Arabia3Occupational Health and Safety, Diriyah Company, Ad Diriyah, Saudi Arabia4Department of Family and Community Medicine, College of Medicine Majmaah University, Majmaah, Saudi Arabia5Department of Information System Management, Stanton University, Garden Grove, CA, United States
Edited by: Justin Yang, Boston University, United States
Reviewed by: Nicola Mucci, University of Florence, Italy
Vasyl Lozynskyi, Dnipro University of Technology, Ukraine
Atul Kumar Singh, Dayananda Sagar College of Engineering, India
*Correspondence: Zara Jamil, zara.yprofest@gmail.comMohammad Miraj, m.molla@mu.edu.sa
†ORCID: Mohammad Miraj, orcid.org/0000-0001-8590-0531
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Despite advancements in operational technologies, the oil and gas (O&G) industry continues to face safety lapses due to persistent challenges in occupational health and safety management (OHSM), hazard identification and risk assessment (HIRA), and safety communication (SC). This study aims to examine how OHSM, HIRA, and SC influence safety knowledge (SK) and safety performance (SP), with safety culture (SCULT) mediating these relationships. A novel framework, integrating technical, procedural, and cultural dimensions is proposed and empirically tested. Grounded in social exchange theory (SET), this framework is distinct in its dual focus on system-level safety practices and cultural mechanisms in high-risk environments. Malaysia’s downstream O&G sector was chosen due to its operational complexity, multicultural workforce, and elevated accident rates, making it a critical context for evaluating sustainable safety interventions. Data was collected from 350 employees from the operational department using stratified sampling across two major national oil companies PGB and MLNG. Partial least squares structural equation modeling (PLS-SEM) was employed to validate the model, demonstrating strong reliability and predictive relevance (SRMR = 0.064, AVE > 0.5). Notably, SC exhibited weak direct effects on SK and SP, but strong indirect effects via SCULT, suggesting that communication-based safety interventions are only effective when trust and cultural alignment are present. Behavioral outcomes such as proactive safety participation and cognitive outcomes such as hazard recognition were both positively influenced by a strong SCULT. This research offers practical strategies for industry stakeholders, including the adoption of a near-miss reporting system, behavior-based safety (BBS) training programs, and culturally adaptive communication audits. Policymakers are encouraged to embed cultural indicators within national safety audit frameworks and promote leadership accountability across organizational levels. The findings emphasize that achieving sustainable safety outcomes require more than structural compliance with culturally integrated safety systems.
hazard identification and risk assessmentoccupational health and safetysafety communicationsafety performancesustainable safety practicessafety knowledgesafety cultureUniversiti Teknologi PETRONAS10.13039/501100005710Universiti Teknologi PETRONAS10.13039/501100005710section-at-acceptanceOccupational Health and SafetyIntroduction
Sustainability, within high-risk industries is no longer confined to environmental stewardship and economic efficiency. It now necessitates equal emphasis on the social dimensions that are particularly related to the health and safety of workers. In the Malaysian oil and gas (O&G) sector, where operational complexity and hazard exposure are prevalent, the integration of occupational health and safety (OHS) practices into sustainability agendas has become imperative. Traditionally, the industry has emphasized economic metrics such as lifecycle cost and return on investment, alongside environmental targets like emissions control and resource efficiency (1). However, recent literature acknowledges that sustainable industrial development is unattainable without ensuring the physical, mental, and emotional well-being of the workforce (2, 3).
The downstream segment of Malaysia’s O&G industry which encompasses refining, processing, and distribution is especially prone to safety lapses due to the hazardous nature of its operations, aging infrastructure, and high workforce density. Although major legislative frameworks such as the 1984 Petroleum Act and 1994 Occupational Safety and Health Act (OSHA) have established safety standards, the sector continues to experience incident rates. For example, Malaysia recorded a 125% increase in O&G worker fatalities between 2009 and 2015, a trend that significantly contrasts with countries like the United States, where O&G-related fatalities account for approximately 18% of all industrial deaths (4). Hong Kong O&G sector has been cited as responsible for nearly 75% of worker fatalities, underscoring the severity of risks inherent to the industry. These trends suggest critical limitations in the effectiveness of conventional OSHMS in comprehensively addressing complex operational hazards. Moreover, macroeconomic stressors such as global financial downturns, COVID-19-related budget constraints, and energy market volatility have led many organizations to deprioritize safety investments in favor of cost containment. As a result, safety training programs, supervisory oversight, and communication systems are often underfunded or fragmented, especially in emerging economies. This reinforces the urgent need for resilient and culturally embedded safety frameworks capable of withstanding such external pressures. (1, 3, 5).
The selection of Malaysia’s downstream O&G sector for this study is deliberate. This segment represents a high-density, multi-ethnic, and operationally intense environment, often subject to both internal organizational challenges and external regulatory and economic fluctuations. Its strategic relevance in Malaysia’s national energy landscape, coupled with a rising concern over accident rates, makes it an ideal context for examining sustainable safety practices. Recent studies have pointed to the growing importance of safety culture (SCULT) and communication in shaping safety outcomes, yet several challenges persist. Safety communication (SC), although widely acknowledged as critical, lacks standardization in definition and implementation across regulatory bodies and organizational levels (6–8). This ambiguity hinders consistent enforcement and impairs the feedback loop required for responsive hazard management. Similarly, the concept of SCULT though recognized as a basis of sustainable safety practices remains difficult to operationalize, especially when disaggregated from organizational climate and leadership engagement. The industry continues to struggle with inadequate hazard identification and risk assessment (HIRA) practices, fragmented communication structures, and a disconnect between formalized safety procedures and employee behaviors on the ground (9–11).
While various studies have independently examined the roles of OHSM, HIRA, and SC in improving safety performance (SP). Yet, their combined influence and interaction, especially through the mediating mechanism of SCULT, are not well understood. Furthermore, limited empirical evidence exists on how these constructs affect safety knowledge (SK), a critical component for employee preparedness, hazard recognition, and safety decision-making. This is particularly relevant in Malaysian downstream O&G sites. These are sites where hierarchical communication patterns, multinational workforce dynamics, and complex site operations can dilute safety priorities and hinder knowledge retention.
To address this knowledge gap, the present study proposes and empirically tests a structural model that captures the interrelationships among OHSM, HIRA, SC, SCULT, SK, and SP. By elucidating these relationships, the study aims to provide valuable insights for targeted interventions that enhance sustainable safety practices (SSP) and prevent accidents. Grounded in social exchange theory (SET), this research posits that a reciprocal relationship exists between employees and their organization: when organizations invest in transparent communication, participatory safety practices, and effective HIRA protocols, employees are more likely to respond with increased safety compliance, knowledge sharing, and proactive behavior. SET provides a theoretical lens through which the motivational and behavioral aspects of workplace safety can be better understood, particularly in relation to trust, fairness, and mutual accountability.
This study’s primary objective is to explore the multidimensional dynamics of sustainable safety practices in the Malaysian downstream O&G industry. Using partial least squares structural equation modeling (PLS-SEM) and survey data from 350 workers from operational department, this research:
Investigate how OHSM, HIRA, and SC affect SCULT.
Examine the role of SCULT in mediating their impact on SP and SK.
Identifies barriers and enablers of effective SC and hazard management.
Provides actionable insights for developing targeted interventions to enhance sustainable safety outcomes.
This study aims to contribute to the growing body of occupational health and safety research by offering a robust, evidence-based framework that supports policy formulation, capacity building, and operational excellence in high-risk industrial settings. It also sets a precedent for evaluating SCULT not as a standalone construct, but as a dynamic mediator that reflects the depth and quality of organizational safety practices.
These research understanding can be applied not just in Malaysia’s downstream O&G industry, but also in other high-risk sectors globally. It can also support the development of practical strategies to improve safety within O&G companies. These strategies can then be disseminated and adapted for use in other industries facing similar challenges. Thus, the proposed framework directly contributes to achieving a more sustainable O&G industry. By promoting a safety-conscious workforce and preventing accidents, the proposed approach can minimize environmental risks and ensure the well-being of workers essential elements for long-term industry sustainability.
The rest of the paper is organized as follows: Section 2 provides the background to identify the knowledge gap and develop hypotheses. Section 3 details the research method used, including the quantitative approach and data analysis techniques. Section 4 presents the study’s findings and explores their significance for safety practices in the O&G industry. Finally, Section 5 concludes the paper by summarizing the key takeaways, limitations, and future directions.
Background
Despite advances in occupational safety regulation and industry standards, the Malaysian O&G sector, particularly its downstream sector continuously faces persistent safety challenges. Characterized by high-risk processing environments, aging infrastructure, and a multi-ethnic workforce, this sector remains vulnerable to both procedural lapses and systemic communication breakdowns. Recent data reveal that while global occupational fatality rates have improved (Figure 1), Malaysia’s downstream O&G operations lag, with incident frequencies surpassing regional benchmarks (12). Human error, often symptomatic of deeper organizational issues, remain a dominant cause exacerbated by weak SCULT, limited safety knowledge, and inconsistent communication practices (13, 14).
Comparative analysis of different countries by occupational injury and fatality rates (71).
Bar and line chart titled "Analysis of Occupational Injuries and Fatalities" compares data across nine countries. Gray bars represent non-fatal injuries per 100,000 workers, the green area signifies inspectors per 10,000 employed persons, and the blue line depicts fatalities per 100,000 workers. Countries shown are Australia, Canada, China, Japan, Malaysia, New Zealand, Philippines, Singapore, Thailand, and the USA. The chart illustrates varying levels of injuries and fatalities, with China having the highest injuries and Singapore the lowest.
Malaysia’s regulatory landscape is shaped by the 1984 Petroleum Act, the 1967 Factory and Machinery Act, and the 1994 OSHA. However, the sector still struggles with effective enforcement, especially in downstream operations, where onshore and offshore challenges differ significantly. According to Nawaz et al. (1) and Sanchez-Gomez et al. (5), broader macroeconomic forces such as financial downturns, pandemic-driven budget constraints, and energy market volatility further compromise organizational commitment to safety investments and training. These external pressures often lead to reprioritization of safety expenditures in favor of operational throughput. In response to these multifactorial risks, there is growing academic and industry consensus on the need to move beyond compliance-based approaches toward systems that embed safety into organizational culture. Studies across different regions reinforce that cultivating a robust SCULT built on trust, leadership visibility, and participatory communication is essential for improving safety outcomes in high-risk industries (14–18). In parallel, SC is increasingly recognized not merely as a function of information dissemination, but as a social and behavioral construct shaped by trust, clarity, timeliness, and feedback loops. CS and employee involvement in safety dialogs substantially enhance hazard recognition and risk mitigation (19, 20). However, SC in many Malaysian O&G settings remains hierarchical and compliance-driven, undermining its effectiveness in engaging frontline workers.
Hazard Identification and Risk Assessment (HIRA) continues to serve as a foundational practice for mitigating site-level risks. Yet, emerging research recommends integrating HIRA into broader organizational learning systems to ensure that risk data is not only recorded but also translated into preventive actions and workforce knowledge (21, 22). The absence of such integration often results in repeated near-miss events and minimal learning from past incidents. Additionally, limited employee involvement in HIRA processes impedes ownership, situational awareness, and safety-related decision-making (23).
This study integrates three core constructs OHSM, SC, and HIRA within a unified safety performance model mediated by SCULT. Rooted in SET, the framework views safety as a reciprocal process: when organizations invest in inclusive safety systems, transparent communication, and trust-building practices, employees reciprocate through knowledge sharing, compliance, and proactive safety behavior (24, 25). Notably, previous studies have examined OHSM, HIRA, and SC in isolation, failing to account for their synergistic effects within a cultural framework. Moreover, the mediating role of SCULT in converting these systemic inputs into improved Safety Knowledge (SK) and Safety Performance (SP) remains underexplored, particularly in the Malaysian downstream O&G context. As such, this study aims to fill that gap by empirically testing a structural model that links these constructs, while integrating contextual realities such as hierarchical communication, economic uncertainty, and workforce diversity. This research contributes to both theory and practice. It provides empirical evidence supporting the role of cultural enablers in safety systems and informs targeted interventions for improving hazard control, communication structures, and workforce engagement in high-risk environments.
Conceptual framework and hypotheses formation
Contemporary literature increasingly highlights the pivotal role of integrated safety management practices in enhancing organizational performance and resilience, particularly in high-risk industries such as oil and gas. SSP has emerged as a strategic imperative, not only for regulatory compliance but also for ensuring operational continuity, environmental stewardship, and workforce well-being. Key components of SSP include structured OHSM, comprehensive HIRA, effective SC, and a strong SCULT. Together, these elements contribute to improve SK among employees and better SP across operations (26, 27).
While previous research has explored these components individually, there is a dearth of studies that empirically examine their interrelationships within a unified framework, particularly in the downstream O&G context. Moreover, few models integrate the mediating role of SCULT or address how these constructs jointly influence knowledge transfer and safety behavior on the ground. Drawing on the principles of SET, this study posits that workplace safety is a reciprocal process, wherein management’s commitment to transparent communication, proactive risk management, and participatory safety systems enhances employee trust, knowledge retention, and compliance with safety protocols. This framework is grounded in SET which suggests that when organizations invest in fair, participatory, and transparent practices, employees reciprocate through enhanced engagement, trust, and compliance. Within the safety domain, this translates into proactive safety behavior, willingness to share safety concerns, and improved hazard responsiveness.
The conceptual framework underpinning this study is adapted from the logic of the Balanced Scorecard (BSC). BSC is a well-established performance measurement system adopted by over 70% of global industries (28, 29). While the BSC traditionally focuses on financial, customer, internal process, and learning perspectives, this study adapts its structure to suit the petrochemical and O&G sectors by emphasizing health and safety as critical performance dimensions. Figure 2 illustrates the proposed framework, which maps the relationships between OHSM, SC, HIRA, SCULT, SK, and SP.
Structural model with hypothesis.
A flowchart depicting relationships between variables in safety management. Independent variables include Occupational Health and Safety Management, Safety Communication, and Hazards Identification and Risk Assessment. The dependent variables are Safety Performance and Safety Knowledge. Safety Culture acts as a mediating variable. Arrows represent hypothesized paths, labeled H1 to H11, indicating connections between variables.Occupational health and safety management system
OHSM refers to the set of organizational strategies, policies, procedures, and leadership behaviors aimed at promoting workplace health and safety. In high-risk industries, its implementation requires more than compliance. It demands cultural alignment, leadership commitment, and continuous workforce engagement. Despite its recognized value, OHSM implementation often suffers from fragmented leadership attention, constrained resources, and the prioritization of production targets over safety concerns (30). These barriers are especially pronounced in downstream O&G operations, where fast-paced production and complex processing environments can marginalize safety initiatives.
Evidence suggests that effective OHSM enhances employee perceptions of organizational safety commitment, which in turn improves safety behavior and reduces accident rates (31). This study adopts a holistic view of OHSM, emphasizing its alignment with proactive HIRA practices, access to protective equipment, real-time training, and inclusive safety participation. When safety policies are consistently implemented and communicated, they strengthen SCULT, advance knowledge dissemination, and reinforce safety behavior (32). Thus, the hypotheses are:
Hypothesis 3: OHSM positively influences SCULT.
Hypothesis 3: OHSM has a direct positive impact on SP.
Hypothesis 3: OHSM positively enhances SK.
Safety communication
SC is the mechanism through which safety-related information, expectations, and concerns are exchanged within an organization. It plays a critical role in encouraging a shared understanding of risks, aligning safety expectations, and facilitating immediate responses to emerging hazards. Effective SC is characterized by clarity, timeliness, openness, and bidirectional flow. Studies show that upward communication, where employees feel safe voicing concerns, significantly enhances hazard awareness and contributes to injury prevention (33, 34). SC also serves as an indicator of leadership commitment. When leaders actively participate in safety dialogs and model best practices, they legitimize safety priorities and instill confidence across teams (35). Conversely, attributing safety breakdowns solely to worker error, without addressing managerial communication failures, undermines employee trust and weakens SCULT.
In the Malaysian O&G context, nuanced findings have emerged. Even in environments with a generally positive SCULT, operational performance pressures and job security fears may lead employees to deprioritize safety concerns in favor of meeting production targets (36, 37). This stresses the need for structured, consistent SC practices that address both formal and informal communication pathways. Thus, the study’s hypotheses for SC are:
Hypothesis 4: SC positively affects SCULT.
Hypothesis 5: SC directly improves SP.
Hypothesis 6: SC positively influences SK.
Hazard identification and risk assessment
HIRA is a foundational process within OHSMS that systematically identifies, evaluates, and prioritizes potential hazards associated with work activities. Its objective is to develop control measures that mitigate health, safety, and environmental risks before they materialize. Effective HIRA not only reduces operational vulnerability but also cultivates a culture of vigilance and shared responsibility (25). A well-structured HIRA process should be integrated into the local operating management system (LOMS), with clear documentation of procedures, assignment of roles, and training protocols. Active involvement of personnel in hazard assessment promotes ownership and enhances safety awareness. Moreover, regular review of HIRA output ensures alignment with evolving operational conditions and workforce competencies. The value of HIRA extends beyond procedural compliance; it supports SCULT development, improves SP, and builds institutional knowledge that strengthens long-term resilience (24). The proposed hypotheses for HIRA are:
Hypothesis 7: HIRA positively affects SCULT.
Hypothesis 8: HIRA significantly improves SP.
Hypothesis 9: HIRA positively enhances SK
Safety culture
SCULT encompasses the collective values, beliefs, and behaviors that determine how safety is prioritized, communicated, and enacted within an organization. It reflects the depth of leadership commitment to safety, the clarity of policies, and the alignment of individual behavior with organizational goals (38). Empirical evidence associates strong SCULT with lower incident rates, increased employee engagement, and higher compliance with safety procedures (39). As a multidimensional construct, SCULT not only influences SP and SK directly but also mediates the relationship between managerial practices and safety outcomes. Turner et al. emphasized that SCULT emerges from the integration of systems, symbols, and behaviors that minimize risks. Regulatory integration of SCULT, as argued by Antonsen et al. (40), can enhance both compliance and performance at institutional levels.
In high-risk sectors like O&G, cultivating SCULT requires more than symbolic gestures; it demands continuous investment in safety leadership, transparent SC, and inclusive safety participation (33, 41). The mediating role of SCULT is particularly relevant in environments where trust and communication gaps may distort the intended effects of safety interventions. Accordingly, this study hypothesizes both direct and mediating roles of SCULT as:
Hypothesis 10: SCULT has a positive impact on SP.
Hypothesis 11: SCULT has a positive impact on SK.
Hypothesis 12: SCULT mediates the relationship between OHSM and SP.
Hypothesis 13: SCULT mediates the relationship between OHSM and SK.
Hypothesis 14: SCULT mediates the relationship between SC and SP.
Hypothesis 15: SCULT mediates the relationship between SC and SK.
Hypothesis 16: SCULT mediates the relationship between HIRA and SP.
Hypothesis 17: SCULT mediates the relationship between HIRA and SK.
Safety knowledge
SK refers to employees’ understanding of safety procedures, hazard recognition, response protocols, and their ability to apply this knowledge in practice. It is a vital determinant of effective safety behavior and is closely linked to training effectiveness, communication quality, and leadership involvement (42). Accurate and accessible SK fosters confidence, encourages hazard reporting, and improves decision-making in high-pressure situations (43). Despite its importance, knowledge gaps persist in many O&G settings, particularly among contract workers and shift-based personnel (44). This study posits that SC and SCULT significantly shape the depth and application of SK across operational levels.
Safety performance
SP reflects the extent to which organizations achieve desired safety outcomes, such as reduced injury rates, incident-free operations, and proactive safety behavior. It encompasses both leading indicators (e.g., participation, audits, and training compliance) and lagging indicators (e.g., accident rates, near misses) (45). A high level of SP indicates that safety systems are not only in place but are effectively internalized and practiced by the workforce.
Effective SP requires alignment between technical systems, behavioral interventions, and organizational culture. Employee involvement in safety planning and implementation has been shown to significantly enhance SP outcomes (46). Hence, SP is not only a function of management oversight but a collective measure of organizational safety maturity.
Research methods
This section outlines the research methodology adopted to investigate the structural relationships between OHSM, SC, HIRA, SCULT, SK, and SP within Malaysia’s downstream O&G industry. A quantitative approach was employed, underpinned by the positivist paradigm, to validate a theoretical model using structural equation modeling techniques. The research design serves as the comprehensive plan of action, encompassing the formulation of research questions, data collection, analysis, and the subsequent discussion and justification (47). This section briefly describes the research method used for this study and Figure 3 presents the research flow chart.
Study’s research flow-chart.
Flowchart illustrating the research approach for safety enhancement. It starts with identifying critical factors, followed by a pilot study, a main survey, and PLS-SEM development, leading to structural and measurement models. Structural model results in bootstrap and collinearity analysis, while measurement model checks convergent and discriminant validity. These contribute to the overall structure model of research approaches.Research design
This study employed a cross-sectional research design, following a quantitative approach. Data was collected at a single point in time, from July to October 2023. This design allowed the study to investigate the relationships between OHSM, SCULT, SC, HIRA, SK, and SP. The study will discuss the details of research methodology in subsequent sections. The selection of the positivist paradigm was guided by its focus on empirically grounded concepts. This approach enabled to gain a comprehensive understanding of the current state of safety practices within the studied population (9, 48).
Questionnaire design
The research instrument was developed based on an extensive literature review and adapted from validated measurement scales to ensure construct validity. The questionnaire was divided into seven sections: demographics, OHSM, SC, HIRA, SCULT, SK, and SP. A five-point Likert scale ranging from 1 (“strongly disagree”) to 5 (“strongly agree”) was employed to capture respondent perceptions with appropriate granularity. To establish content and face validity, initial drafts of the instrument were reviewed by academic experts in occupational health and safety and industry practitioners from the Malaysian O&G sector. Later, minor revisions were made based on clarity, contextual fit, and language appropriateness. Subsequently, interviews were conducted with four downstream O&G professionals to confirm item relevance and comprehension, further strengthening construct reliability.
This study investigated the relationship between OHSM, SCULT, and employee safety outcomes. Therefore, to capture these concepts, a questionnaire was developed based on a thorough review of relevant literature. Items for SC were drawn from (49), ensuring they accurately reflect the exchange of safety information and procedures between supervisors and employees. OHSM, which refers to the overall safety atmosphere created by management safety practices, was measured using items adapted from Guzman et al. (50) and Ammari et al. (51). To assess SCULT, the questionnaire incorporated items based on the work of Dahl & Kongsvik (52) and Newnam & Goode (53), which explore employee contentment with hazard identification and risk assessment processes. Finally, the construct of SK and SP was operationalized using items informed by research from Ertürk et al. (35); Bray & Williams (54); and Ehiaguina (55). These references address employee dedication to safety practices and procedures, ensuring the questionnaire effectively captures this crucial aspect of SCULT.
Sampling and data collection
A stratified purposive sampling technique was employed to ensure proportional representation across job roles (technicians, HSE officers, engineers, and managerial staff) and departments (safety, operations, maintenance). The study was conducted in collaboration with safety departments at PETRONAS Gas Berhad (PGB) and Malaysia LNG (MLNG). Out of 1,400 distributed questionnaires, 350 were completed and validated, yielding a response rate of 25%. The survey was administered via a hybrid approach. Online questionnaires were distributed through official company email lists and paper-based, in-person questionnaires were completed during toolbox meetings and HSE training sessions for employees without frequent digital access. This multi-modal approach (occurred between July and October 2023) ensured inclusivity across various working environments, including shift-based and remote operational units. The final sample was deemed adequate for partial least squares structural equation modeling (PLS-SEM), as recommended by Hair et al. (56) for models with multiple latent constructs.
Pilot study
A pilot study was conducted prior to full-scale deployment to assess the instrument’s reliability and feasibility. The pilot responses were analyzed for internal consistency, response clarity, and survey duration. Based on the pilot findings ambiguities in terminology were resolved, question sequence was adjusted to improve flow, and a Cronbach’s alpha of above 0.7 was confirmed for all constructs, indicating acceptable reliability.
Analytical approach
The study employed partial least square structural equation modeling (PLS-SEM) and PLSpredict to assess out-of-sample prediction for the model due to the robustness in handling complex models with multiple latent constructs, flexibility with smaller sample sizes, and emphasis on prediction and exploratory modeling. SEM is often used in studies to examine and model workplace safety and ergonomics (57, 58). This study utilizes SEM to analyze the complex relationships between the variables in question. SEM is particularly well-suited for investigating safety-related factors within the Malaysian downstream O&G industry. It goes beyond simply identifying patterns in the data by combining elements of multilevel regression and component analysis. This allows for a more nuanced assessment of how well the proposed model fits the actual data and quantifies the relative importance of each factor (59). In SEM, survey data is interpreted by establishing connections between underlying (latent) variables and the observable measures used to collect data (9). To ensure the stability of the analysis, a sample size of 350 respondents was deemed appropriate for this study (56). Furthermore, bootstrapping was employed within the Smart PLS software to assess the model’s robustness and analyze data from the main questionnaire. This technique provides a valuable tool for examining the functionality of the relationships identified through SEM.
Result
This section presents the results of the data analysis conducted using PLS-SEM. The analysis was performed in two major stages: (1) evaluation of the measurement model and (2) assessment of the structural model, including hypothesis testing. A total of 350 valid responses were included in the analysis, representing various roles across Malaysia’s downstream oil and gas (O&G) operations.
Construct metrics: descriptive and validity analysis
Focusing on O&G downstream accidents in Malaysia and SC’s role in preventing them, this study surveyed 1,400 individuals however 350 individuals responded (87.5% response rate), which yielded five main components: OHSM (18.25%), SC (9.93%), SCULT (7.87%), HIRA (6.56%), SK (6.57%), and SP (3.34%). This explains 38% of the variance, suggesting several factors contributing to safety.
Demographic profile
The sample consisted of 326 male respondents and 24 females, which aligns with the male-dominated nature of the field. The 68 respondents were aged between 20 and 30, 197 were aged between 31 and 40, 31 were aged between 41 and 50, and 54 respondents were over 51, indicating a relatively youthful and energetic workforce in the PETRONAS O&G industry. Among the participants, 39 were unmarried, while 311 were married (presented in Figure 4). The research participants generally agreed upon the survey items, as indicated by the high mean scores averaging 4 for each construct. The standard deviation (SD) values ranged from 0.420 (for SK1) to 0.803 (for SC8), as reported in Table 1.
Demographic profile.
Demographic profile diagram displaying various attributes: Gender/Marital Status (84% male, 80% married), Age (56.3% are 31-40 years), Academic Level (34% have a first degree), Employment (75% permanent), Experience (20% with 1-4 years), Level in Organization (56% non-executive), Others (77% regular hours job, 98% Malaysian), and Incident (94% never happen).
Reliability and validity test outcomes of constructs.
Safety phase
Assigned code
Mean
SD
Initial loadings
Modified
Cronbach alpha
Composite reliability
AVE
Occupational health and safety management (OHSM)
OHSM 1
4.46
0.553
0.813
0.856
0.813
0.864
0.519
OHSM 2
4.37
0.634
0.798
0.810
OHSM 3
4.38
0.707
0.773
0.804
OHSM 4
4.12
0.777
0.608
0.672
OHSM 5
4.41
0.727
0.789
0.83
OHSM 6
4.33
0.641
0.603
0.666
OHSM 7
4.31
0.72
0.449
Deleted
OHSM 8
3.24
1.305
0.533
Deleted
Hazard identification and risk assessment (HIRA)
HIRA 1
4.50
0.544
0.801
0.809
0.726
0.829
0.550
HIRA 2
4.19
0.813
0.645
0.655
HIRA 3
4.33
0.694
0.796
0.802
HIRA 4
4.32
0.668
0.678
0.686
HIRA 5
4.31
0.995
0.789
0.835
Safety Culture (SCULT)
SCULT 1
4.29
0.713
0.812
0.866
0.885
0.878
0.522
SCULT 2
3.92
0.981
0.686
0.771
SCULT 3
4.27
0.695
0.849
0.894
SCULT 4
4.49
0.665
0.788
0.822
SCULT 5
0.512
659
Safety Communication (SC)
SC 1
4.21
0.692
0.699
0.831
0.887
0.912
0.759
SC 2
4.22
0.742
0.802
0.814
SC 3
3.93
0.767
0.791
0.804
SC 4
3.97
0.878
0.799
0.805
SC 5
4.24
0.717
0.723
0.757
SC 6
4.11
0.803
0.551
Deleted
Safety Knowledge (SK)
SK1
4.8
0.420
0.812
0.822
0.800
0.861
0.555
SK2
4.55
0.605
0.771
0.805
SK3
4.48
0.863
0.533
0.655
SK4
4.62
0.535
0.715
0.734
SK5
4.43
0.598
0.665
0.696
Safety Performance (SP)
SP1
4.43
0.598
0.745
0.788
0.903
0.927
0.719
SP2
4.52
0.544
0.819
0.858
SP3
4.24
0.7
0.812
0.817
SP4
4.38
0.841
0.874
0.877
SP5
4.46
0.907
0.833
0.854
SP6
4.53
0.681
0.422
Deleted
SP7
0.110
Deleted
Model assessment
Analyzing the constructs (OHSM and SK/ SP) revealed that numerous items demonstrated statistically significant loading values exceeding 0.5, indicating their strong contribution. However, a closer examination identified five items that exhibit both insignificant contributions and loading values below 0.5, constituting less than 20% of total items, hence excluded. Furthermore, the variance inflation factor (VIF) for the remaining items persisted below 3.3, confirming the absence of multicollinearity issues within the model and that other factors do not inflate the constructs. Finally, subsequent validity and reliability tests confirmed the robustness and validity of the first-order construct.
All measured items in this analysis established internal consistency and reliability criteria. Specifically, all items possessed average variance extracted (AVE) values exceeding 0.5, Cronbach’s alpha surpassing 0.7, and composite reliability exceeding 0.8. These statistics, presented in Table 2, demonstrate that the items consistently measure the intended construct and are free from random error. Notably, while all loading values surpass 0.7, except for item SK5 (0.696), this single instance below the threshold is considered acceptable based on established criteria (56). This indicates that even with this slightly lower loading, the overall internal consistency of the formative construct remains satisfactory.
Discriminant validity heterotrait-monotrait ratio (HTMT).
Construct
HIRA
OHSM
SP
SCULT
SC
SK
HIRA
OHSM
0.801
SP
0.846
0.845
SCULT
0.674
0.65
0.721
SC
0.634
0.596
0.647
0.853
SK
0.835
0.698
0.697
0.534
0.816
To ensure the entire model’s validity and robustness, this study employed comprehensive assessments focusing on discriminant validity and model fit.
Discriminant Validity: This assessment aimed to confirm that the latent constructs within the model are distinct and not simply measuring the same underlying phenomenon. To achieve this, the study examined the correlations between the latent constructs and compared them to the square root of AVE for each construct. If the correlations remain lower than the square root of AVE, it indicates that the constructs are sufficiently distinct. Additionally, the study employed the Heterotrait-Monotrait (HTMT) ratio procedure, which further confirms discriminant validity when HTMT values are less than 1. The Fornell-Larcker criterion (based on correlations and AVE) and the HTMT procedure confirmed discriminant validity in this model. Table 2 highlights the advantages of HTMT over the traditional Fornell-Larcker criterion (presented in Table 3), demonstrating its superiority in discerning truly distinct constructs. The first stage of analysis of formative construct is presented in Table 4.
Discriminant validity-Fornell-Lacker criterion.
Construct
HIRA
OHSM
SP
SCULT
SC
SK
HIRA
0.761
OHSM
0.654
0.716
SP
0.729
0.686
0.768
SCULT
0.566
0.522
0.616
0.714
SC
0.563
0.507
0.577
0.756
0.846
SK
0.727
0.591
0.622
0.466
0.48
0.795
Latent variables description.
Construct
Items
Outer weight
STDEV
T-value
p-value
Outer loading
VIF
Occupational Health and Safety Management (OHSM)
OHSM 1
0.296
0.020
14.914
0.001
0.856
2.366
OHSM 2
0.291
0.025
11.854
0.001
0.810
1.998
OHSM 3
0.201
0.021
9.356
0.001
0.738
1.434
OHSM 4
0.160
0.022
7.228
0.001
0.631
1.463
OHSM 5
0.213
0.024
8.709
0.001
0.830
1.464
OHSM 6
0.203
0.021
9.652
0.001
0.717
1.441
OHSM 7
0.205
0.016
11.722
0.001
0.756
1.452
OHSM 8
Dropped
Safety Culture (SCULT)
SCULT 1
0.246
0.016
5.777
0.001
0.866
1.577
SCULT 2
0.247
0.011
22.84
0.001
0.894
3.450
SCULT 3
0.231
0.021
18.334
0.001
0.822
2.278
SCULT 4
0.222
0.012
14.136
0.001
0.729
1.482
SCULT 5
0.163
0.013
16.559
0.001
0.771
1.802
Safety Communication (SC)
SC 1
0.237
0.013
18.803
0.001
0.740
2.598
SC 2
0.191
0.016
12.266
0.001
0.788
2.704
SC 3
0.201
0.014
14.854
0.001
0.859
2.056
SC 4
0.210
0.013
14.795
0.001
0.984
2.060
SC 5
0.192
0.017
14.771
0.001
0.874
1.802
SC 6
Dropped
Safety Knowledge (SK)
SK1
0.416
0.019
3.844
0.001
0.854
2.036
SK2
0.292
0.023
11.298
0.001
0.778
1.463
SK3
0.348
0.016
5.777
0.877
1.858
SK4
0.375
0.011
22.844
0.001
0.826
1.692
SK5
0.282
0.014
23.002
0.001
0.747
1.407
Safety Performance (SP)
SP1
0.239
0.016
15.236
0.001
0.815
2.388
SP2
0.259
0.015
16.870
0.001
0.836
2.882
SP3
0.183
0.014
13.274
0.001
0.853
3.511
SP4
0.220
0.011
20.282
0.001
0.891
1.196
SP5
0.226
0.014
16.344
0.001
0.834
1.099
SP6
Dropped
SP7
Dropped
Hazard Identification and Risk Assessment (HIRA)
HIRA 1
0.416
0.026
16.280
0.001
0.809
1.463
HIRA 2
0.292
0.028
10.557
0.001
0.655
1.239
HIRA 3
0.348
0.019
18.035
0.001
0.802
1.611
HIRA 4
0.282
0.025
11.144
0.001
0.86
1.360
HIRA 5
0.261
0.020
14.914
0.001
0.812
2.366
Model Fit: This assessment ensures that the proposed model adequately represents the observed data. This study employed the standardized-root-mean-square-residual (SRMSR) as a key indicator of model fit. An SRMSR value below 0.08 generally suggests a good fit. In this case, the obtained value of 0.078 further confirms that the model accurately reflects the relationship between the constructs and variables. R-squared (R2) values for each endogenous construct also provide insights into the model’s explanatory power. Acceptable R2 values in this analysis indicate that the model effectively explains a significant portion of the variance in the dependent variables.
Hypothesis Testing: Finally, we rigorously tested the 17 proposed hypotheses within the model using a bootstrapping procedure with 5,000 samples. This technique strengthens the reliability of the results by simulating repeated samples and assessing the stability of the findings. The bootstrapping analysis confirmed 11 hypotheses have a significant impact while two hypothesis were rejected, with detailed results in Table 5. Additionally, consistent PLS (PLSc) was employed throughout the analysis to ensure consistent estimation of path coefficients and indicator loadings, particularly relevant for models with non-recursive structures, as outlined by Hair et al. (56).
Result of hypothesis testing.
Path coefficient
β
STDEV
T Statistics
p-values
H1 OHSM→SCULT
0.343
0.059
5.829
0.001
H2 OHSM→SK
0.125
0.06
2.093
0.001
H3 OHSM→SP
0.125
0.064
1.946
0.001
H4 SC→SCULT
0.111
0.059
1.891
0.059
H5 SC→SK
0.053
0.066
0.81
0.418
H6 SC→SP
0.096
0.065
1.468
0.142
H7 HIRA→SCULT
0.424
0.053
7.931
0.001
H8 HIRA→SK
0.361
0.068
5.269
0.001
H9 HIRA→SP
0.347
0.07
4.933
0.001
H10 SCULT→SK
0.203
0.054
3.732
0.001
H11 SCULT→SP
0.272
0.058
4.645
0.001
H12 OHSM→SCULT→SK
0.13
0.035
3.672
0.001
H13 OHSM→SCULT→SP
0.099
0.031
3.184
0.001
H14 SC→SCULT→SK
0.625
0.036
17.371
0.001
H15 SC→SCULT→SK
0.826
0.019
43.981
0.001
H16 HIRA→SCULT→SK
0.301
0.013
23.456
0.001
H17 HIRA→SCULT→SK
0.255
0.015
13.002
0.001
R2
Q2
SCULT
0.571
0.354
SK
0.613
0.189
SP
0.417
0.229
OHSM
0.387
0.354
OHSM
The results strongly support the positive influence of OHSM on all three key outcomes. In H1 OHSM demonstrated a significant impact on SCULT, revealing a significant association (H1: β = 0.343, t = 5.829, p < 0.001), SK (H2: β = 0.125, t = 2.093, p < 0.001), and safety performance (H3: β = 0.125, t = 1.946, p < 0.001). These findings reinforce the previous assertions that effective OHSM systems create a structured safety environments that strengthens organizational learning, reduces hazard exposure, and ensures behavioral compliance (43). Recognizing its significance, Alcantara et al. (60) assert that OHSM is vital for promoting SK/performance and ensuring employees receive essential education and information. OHSM’s effect on SCULT aligns with the principles of SET, indicating that when organizations visibly prioritize safety, employees are more likely to engage in reciprocal safety behaviors. This reflects the cultural embedding of safety protocols into daily operations, which translates into improved awareness and performance.
SC
The results reveal a more nuanced role of SC. While SC had a marginally significant effect on SCULT (H4: β = 0.111, t = 1.891, p = 0.059), its direct relationships with SK (H5: β = 0.053, t = 0.810, p = 0.418) and SP (H6: β = 0.096, t = 1.468, p = 0.142) were statistically nonsignificant. This suggests that SC, when practiced as a unidirectional or compliance-driven tool, may lack the capacity to influence knowledge retention and behavioral change independently. However, its marginal impact on SCULT hints at a latent influence, emphasizing the importance of communication quality, openness, and consistency. These findings resonate with earlier literature, including Fernández-Muñiz et al. (61); Neal & Griffin (62); and Xia et al. (63), which advocates that SC should be embedded in both regulatory systems and organizational norms to improve hazard awareness and behavioral commitment.
Hira
HIRA emerged as a strong predictor of all three safety outcomes. It significantly influenced SCULT (H7: β = 0.424, t = 7.931, p < 0.001), SK (H8: β = 0.361, t = 5.269, p < 0.001), and SP (H9: β = 0.347, t = 4.933, p < 0.001). These results validate the central role of structured risk assessments in high-risk industries like oil and gas. The findings indicate that when HIRA is implemented with clear accountability and ongoing feedback mechanisms, it strengthens situational awareness and creates a culture of shared responsibility. This aligns with research by Iqbal et al. (25), which emphasizes the capacity of HIRA to transform reactive safety postures into proactive hazard prevention strategies.
Safety culture
Safety culture (SCULT) demonstrated a robust influence on both SK and SP. The results confirm H10 (β = 0.203, t = 3.732, p < 0.001) and H11 (β = 0.272, t = 4.645, p < 0.001), indicating that SCULT is not only a product of organizational systems but also a key driver of behavioral and cognitive safety outcomes.
Furthermore, mediation analysis revealed that SCULT plays a significant intermediary role across multiple paths:
H12: OHSM → SCULT → SK (β = 0.130, t = 3.672, p < 0.001)
H13: OHSM → SCULT → SP (β = 0.099, t = 3.184, p < 0.001)
H14: SC → SCULT → SK (β = 0.625, t = 17.371, p < 0.001)
H15: SC → SCULT → SP (β = 0.826, t = 43.981, p < 0.001)
H16: HIRA → SCULT → SK (β = 0.301, t = 23.456, p < 0.001)
H17: HIRA → SCULT → SP (β = 0.255, t = 13.002, p < 0.001)
These mediation results validate the central theoretical proposition of the study: that SCULT serves as the organizational climate through which formal systems (OHSM, SC, HIRA) manifest their influence on safety outcomes. In accordance with SET, it reflects the reciprocity between management actions and employee safety engagement. These findings align with the work of Zhao et al. (64) and Wang & Bielicki (65), emphasizing the significant impact of SCULT on hazard recognition and prevention in organizations. Thus, maintaining up-to-date knowledge through effective practices is a crucial control tool for accident and injury prevention.
Mediating latent effects
This section explores the indirect effects of OHSM, SC, and HIRA on SP and SK through the mediating role of SCULT. The analysis, presented in Tables 5, 6, reveals that SCULT significantly mediates multiple structural relationships, reinforcing its pivotal function in facilitating sustainable safety outcomes in the downstream oil and gas (O&G) sector.
The mediating effect of SCULT between OHSM and both SP and SK is statistically significant. Specifically, the indirect path from OHSM to SK via SCULT (H12: β = 0.130, t = 3.672, p < 0.001) and from OHSM to SP via SCULT (H13: β = 0.099, t = 3.184, p < 0.001) confirms that cultural alignment is essential for translating safety management systems into practical safety gains. While OHSM had its own direct effects on both SK and SP, the added significance of its indirect effects underlines the reinforcing role of a safety-oriented work climate.
Similarly, HIRA demonstrated strong mediated effects on both SK and SP through SCULT. The results confirm the mediation paths (H16: β = 0.301, t = 23.456, p < 0.001; H17: β = 0.255, t = 13.002, p < 0.001), indicating that structured risk assessment protocols do not act in isolation but are most effective when embedded in a culture of trust, shared responsibility, and continuous learning. These findings reinforce earlier research suggesting that HIRA’s success is amplified by strong SCULT and responsive leadership (25).
Notably, while the direct impact of SC on SK and performance was statistically insignificant (H5 and H6), its indirect effects through SCULT were among the strongest in the model. SC exhibited a robust mediated impact on SK (H14: β = 0.625, t = 17.371, p < 0.001) and an even stronger indirect influence on SP (H15: β = 0.826, t = 43.981, p < 0.001). These findings imply that communication in the O&G context may be ineffective unless it is trusted, culturally grounded, and participatory. The results align with studies emphasizing the need to move from compliance-based SC models to interactive, socially embedded communication systems (66).
In contrast, total effects for SC on SP and SK (including both direct and indirect pathways) remain statistically marginal or non-significant when cultural mediation is weak or absent. For example, SC → SP (T = 0.909, p = 0.363) and SC → SK (T = 1.719, p = 0.086) without SCULT mediation fail to reach conventional thresholds. These results suggest that SC alone is insufficient in driving behavioral change unless accompanied by cultural transformation.
Overall, these findings affirm that SCULT is not merely a background factor, but a central latent construct that enables or inhibits the effectiveness of organizational safety systems. Organizations must therefore cultivate SCULT proactively if they wish to realize the full benefits of technical interventions such as OHSM, SC, and HIRA. This requires aligning communication practices, leadership engagement, and hazard management strategies within a coherent cultural framework to maximize impact on SK and performance (66).
Discussion
This study offers a multidimensional view of how safety practices and outcomes are shaped within high-risk operational environments. Drawing on SET, the results confirm that formal safety systems (OHSM, HIRA, and SC) only translate into meaningful SK and SP when mediated through a strong SCULT. One of the most revealing findings is the behavior of SC. Although SC is commonly positioned in the literature as a pivotal enabler of safety outcomes, the empirical findings demonstrate that SC does not exert a statistically significant direct effect on either Safety Knowledge (SK) or Safety Performance (SP), as reflected in the results for H5 and H6. This challenges traditional assumptions that view communication as an isolated lever of behavioral change. In complex, hierarchical environments such as Malaysia’s downstream O&G sector, where communication often follows a top-down format, SC alone may lack the relational depth, contextual relevance, or psychological safety necessary to directly influence frontline behavior or cognitive safety awareness (33, 39).
However, this does not negate the strategic value of SC. Rather, its true influence is realized when embedded within a strong SCULT. The significant indirect effects of SC mediated through SCULT (H14, H15) highlight that communication becomes effective when grounded in a culture of mutual trust, fairness, and openness. This finding reinforces the premise of SET, which asserts that individuals are more likely to internalize safety messaging and reciprocate safe behavior when they perceive organizational sincerity and fairness. Thus, SC functions not as a standalone mechanism, but as a relational facilitator whose efficacy is contingent on the cultural ecosystem in which it operates. Furthermore, SCULT’s mediating role was not limited to SC alone but it significantly influenced the transmission of OHSM and HIRA effects as well. This affirms SCULT as the trust mechanism that transmits institutional intent into employee engagement (67, 68). For OHSM, the results reveal both direct and indirect pathways toward SK and SP, highlighting the importance of structured systems (e.g., hazard reporting, training, procedural clarity) as foundational, but not sufficient on their own. The effectiveness of OHSM systems is amplified when employees internalize them through a supportive culture that promotes inclusion and accountability.
HIRA emerged as the strongest direct predictor across all dependent variables. Its impact on SK and SP highlights the proactive potential of participatory risk assessment in building situational awareness and behavioral discipline. Employees who are involved in identifying hazards tend to demonstrate stronger cognitive retention and safer decision-making. Moreover, HIRA’s mediated effects through SCULT indicate that the effectiveness of hazard assessments improves when embedded within a trusting culture that fosters shared ownership. This finding aligns with previous research that advocates for dialogic, inclusive, and psychologically safe communication channels in high-risk sectors (13, 66). In essence, it is not just the presence of communication that matters, but its quality, tone, and contextual trustworthiness that determine its effectiveness. The findings substantiate SET in the context of occupational safety. The results illustrate that when organizations fulfill their implicit “contract” with employees by offering safety systems, risk mitigation, and open communication, employees reciprocate with higher engagement, knowledge sharing, and compliance. This reciprocity is particularly evident in the mediating role of SCULT, which functions as a relational bridge between institutional inputs and employee outputs.
Collectively, these findings offer a sector-specific extension of SET, illustrating how both formal structures (OHSM and HIRA) and soft variables (SC and SCULT) must interact to deliver tangible safety outcomes (69). From a practical standpoint, the results advocate investing in SCULT as a strategic asset. Leadership visibility in safety initiatives, transparent communication channels, real-time feedback loops, and workforce empowerment are not peripheral practices but central levers for improving both safety behavior and knowledge (70). For managers, this means evolving from compliance-oriented leadership to relational safety leadership. For policymakers, regulatory frameworks such as Malaysia’s OSHA should consider incorporating leading indicators like cultural cohesion, communication satisfaction, and psychological safety, moving beyond retrospective incident tracking. In summary, the study challenges a one-size-fits-all approach to safety and emphasizes that technical systems only yield results when reinforced by a culture of trust, mutual obligation, and engagement. Safety outcomes are thus best sustained through the co-evolution of structure and culture, formal systems and informal norms.
Conclusion
This study offers a comprehensive examination of sustainable safety practices in Malaysia’s downstream O&G industry by exploring the interconnected roles of OHSM, HIRA, and SC, with SCULT acting as a mediating construct. Through an empirical investigation involving 350 respondents across PETRONAS-affiliated downstream sites, the research employs PLS-SEM to validate a multidimensional safety framework grounded in SET. The findings reveal that while OHSM and HIRA significantly and directly influence both SK and SP, the direct effects of SC were not statistically supported. This challenges longstanding assumptions in the safety literature that position communication as a direct determinant of safety outcomes. Instead, the analysis confirms that SC demonstrates substantial influence only when transmitted through a trust-anchored and participatory SCULT. SCULT not only mediates SC’s effects but also enhances the impact of OHSM and HIRA acting as a relational bridge that translates structural inputs into behavioral and cognitive outcomes.
The study advances theoretical understanding by empirically demonstrating that cultural embedding is essential for converting managerial systems into sustained workforce engagement. From a practical perspective, it offers industry-specific recommendations, including the development of real-time digital hazard reporting tools, behavior-based safety (BBS) training, and leadership development programs that prioritize communication satisfaction and cultural alignment. These tools can reinforce proactive hazard recognition, build trust in communication processes, and elevate accountability across all organizational levels. In terms of contextual relevance, the proposed framework is particularly well-suited for high-density, multicultural, and operationally complex environments such as those found in the downstream oil and gas sector. It accommodates variable communication needs and hierarchical structures, making it a flexible yet robust model for cross-functional safety improvement. While not predictive in a strict statistical sense, its applied utility provides a foundation for ongoing diagnostics, risk assessment refinement, and cultural monitoring.
Finally, the study lays the groundwork for future exploration. Longitudinal and mixed-methods research could assess how safety culture evolves under shifting organizational and economic conditions. Additionally, expanding the framework to incorporate sustainability indicators such as environmental performance and corporate social responsibility would provide a more holistic understanding of sustainable safety management in volatile and high-risk industries.
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
The studies involving humans were approved by Universiti Teknologi PETRONAS Ethics Committee. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
The author(s) declare that financial support was received for the research and/or publication of this article. The authors extend the appreciation to the Deanship of Postgraduate Studies and Scientific Research at Majmaah University for funding this research work through the project number R-2025-1907.
The authors would like to thank the Centre for Graduate Studies (CGS), Universiti Teknologi PETRONAS and Deanship of Postgraduate Studies and Scientific Research at Majmaah University for the study.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declare that no Gen AI was used in the creation of this manuscript.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
ReferencesNawazWLinkePKoҫM. Safety and sustainability nexus: a review and appraisal. J Clean Prod. (2019) 216:74–87. doi: 10.1016/j.jclepro.2019.01.167MunnyAAAliSMKabirGMoktadirMARahmanTMahtabZ. Enablers of social sustainability in the supply chain: an example of footwear industry from an emerging economy. Sustain Prod Consum. (2019) 20:230–42. doi: 10.1016/j.spc.2019.07.003ChenCReniersG. Chemical industry in China: the current status, safety problems, and pathways for future sustainable development. Saf Sci. (2020) 128:104741. doi: 10.1016/j.ssci.2020.104741RibasJRArceMESohlerFASuárez-GarcíaA. Multi-criteria risk assessment: case study of a large hydroelectric project. J Clean Prod. (2019) 227:237–47. doi: 10.1016/j.jclepro.2019.04.043Sanchez-GomezMGiorgiGFinstadGLAlessioFAriza-MontesAArcangeliG. Economic stress at work: its impact over absenteeism and innovation. Int J Environ Res Public Health. (2021) 18:5265. doi: 10.3390/ijerph18105265, PMID: 34063404ByeRJAalbergALRøyrvikJOD. What we talk about when we talk about HSE and culture–a mapping and analysis of the academic discourses. Saf Sci. (2020) 129:104846. doi: 10.1016/j.ssci.2020.104846BirmaGJAgajaSANduJC. Evaluation of safety culture in institutional chemical analytical Laboratories in Oghara and Warri, Delta state, Nigeria. Open J. Saf. Sci. Technol. (2022) 12:17–29. doi: 10.4236/ojsst.2022.121002,KilcullenMPBisbeyTMOttosenMJTsaoKSalasEThomasEJ. The safer culture framework: an application to healthcare based on a multi-industry review of safety culture literature. Hum Factors. (2022) 64:207–27. doi: 10.1177/00187208211060891, PMID: 35068229OforiEKAramSASaalidongBMGyimahJNiyonzimaPMintahC. Exploring new antecedent metrics for safety performance in Ghana’s oil and gas industry using partial least squares structural equation modelling (PLS-SEM). Res Policy. (2023) 81:103368. doi: 10.1016/j.resourpol.2023.103368KiyantajRRamezaniAAmrolahiN. The correlation relationship between concepts of learning organization and safety culture in Iran's nuclear and radiological industries. J Health Saf Work. (2022) 12:160–75.ZaraJNordinSMIshaASN. Influence of communication determinants on safety commitment in a high-risk workplace: a systematic literature review of four communication dimensions. Front Public Health. (2023) 11:1225995.HamidANoor AzmiMAminudinEJayaRZakariaRZawawiA. Causes of fatal construction accidents in Malaysia. IOP Conf Ser Earth Environ Sci. (2019) 220:012044. doi: 10.1088/1755-1315/220/1/012044NealAGriffinMAHartPM. The impact of organizational climate on safety climate and individual behavior. Saf Sci. (2000) 34:99–109. doi: 10.1016/S0925-7535(00)00008-4SadeghiRGoerlandtFJS. Validation of system safety hazard analysis in safety-critical industries: an interview study with industry practitioners. Saf Sci. (2023) 161:106084.BoughabaAHassaneCRoukiaOJS. Safety culture assessment in petrochemical industry: a comparative study of two Algerian plants. Saf Health Work. (2014) 5:60–5. doi: 10.1016/j.shaw.2014.03.005, PMID: 25180135Al-MekhlafiA-BAKanwalNAlhajjMNIshaASNBaarimahAOJS. Trends in safety culture research: a scopus analysis. Safety. (2025) 11:33. doi: 10.3390/safety11020033AigbedionEAyorindeOBAdebisiBJ. Leadership and organizational change in high-risk industries: a model for managing engineering change in oil and gas operations. Gulf Journal of Advance Business Research. (2025) 3:425–55.JamilZNordinSMIshaASN. Enhancing SCT in high-risk work environments: a comprehensive framework for effective communication. ASEAN J Appl Lang. (2023) 2:49–66.ModaHMMahmoudFOKozhiparambathLAbdelrahim RAJIJoOS. The impact of leadership on workplace safety climate and employees’ well-being with particular focus on transformational leadership. Ergonomics. (2025) 31:205–13.TaMTDKimT-EGausdalAH. Leadership styles and safety performance in high-risk industries: A systematic review. Saf Reliab. (2022) 41:10–44. doi: 10.1080/09617353.2022.2035627KemajlZStojanceMGzimILediML. Comprehensive analysis of the mining accident forecasting and risk assessment methodologies: case study–Stanterg mine. Min Miner Depos. (2024) 18:11–7. doi: 10.33271/mining18.02.011BorjaliluN. Risk assessment and machine learning models In: IvascuLPislaruMAlexaL, editors. The future of risk management. Rijeka: IntechOpen (2024)HennyHBudiAHSAdityaMSumaryatnoDAMariawatiAS. Work hazards at workstations in the leather tanning industry using the hazard identification and risk assessment (HIRA) method. J. Ind. Serv. (2025) 10:355–60.PurohitDPSiddiquiNNandanAYadavBP. Hazard identification and risk assessment in construction industry. Int J Appl Eng Res. (2018) 13:7639–67.IqbalMIIsaacOAl RajawyIKhuthbuddinSAmeenA. Hazard identification and risk assessment with controls (Hirac) in oil industry–a proposed approach. Materials Today: Proceedings. (2021) 44:4898–902.BevilacquaMBottaniECiarapicaFECostantinoFDi DonatoLFerraroA. Digital twin reference model development to prevent operators’ risk in process plants. Sustain For. (2020) 12:1088. doi: 10.3390/su12031088GopangMANebhwaniMKhatriAMarriHB. An assessment of occupational health and safety measures and performance of SMEs: an empirical investigation. Saf Sci. (2017) 93:127–33. doi: 10.1016/j.ssci.2016.11.024BiazzoSGarengoPBiazzoSGarengoP. Measuring business performances: the balanced scorecard model In: Performance measurement with the balanced scorecard: A practical approach to implementation within SMEs. Springer Briefs in Business, Springer. (2012). 1–22. Available at: https://books.google.com.my/books/about/Performance_Measurement_with_the_Balance.html?id=NhXHxSqxmncC&redir_esc=yAsiaeiKBontisN. Using a balanced scorecard to manage corporate social responsibility. Knowl Process Manag. (2019) 26:371–9. doi: 10.1002/kpm.1616SmithPMOudykJPotterGMustardC. Labour market attachment, workplace infection control procedures and mental health: a cross-sectional survey of Canadian non-healthcare workers during the COVID-19 pandemic. Annals of work exposures and health. (2021) 65:266–76. doi: 10.1093/annweh/wxaa119, PMID: 33313670YanarBKosnyASmithPM. Occupational health and safety vulnerability of recent immigrants and refugees. Int J Environ Res Public Health. (2018) 15:2004. doi: 10.3390/ijerph15092004, PMID: 30223449Organization WH. COVID-19: Occupational health and safety for health workers: Interim guidance, 2 February 2021World Health Organization (2021). Available at: https://iris.who.int/bitstream/handle/10665/339151/WHO-2019-nCoV-HCW_advice-2021.1-eng.pdf?sequence=1JohariJTanFYAdnanZ. Demystifying the empirical link between safety climate, safety communication, work environment, and unsafe behaviour at work. Jurnal Pengurusan. (2017) 50:35–43.BakasDKostisPPetrakisP. Culture and labour productivity: an empirical investigation. Econ Model. (2020) 85:233–43. doi: 10.1016/j.econmod.2019.05.020ErtürkAVan den BroeckHVerbriggheJ. Self-other agreement on transformational leadership and subordinates’ assessment of supervisor’s performance: mediating role of leader-member exchange. Leadersh Org Dev J. (2018) 39:291–308. doi: 10.1108/LODJ-02-2016-0048LyndonAJohnsonMCBinghamDNapolitanoPGJosephGMaxfieldDG. Transforming communication and safety culture in intrapartum care: a multi-organization blueprint. J Obstet Gynecol Neonatal Nurs. (2015) 44:341–9. doi: 10.1111/1552-6909.12575, PMID: 25851413KimNKRahimNFAIranmaneshMForoughiB. The role of the safety climate in the successful implementation of safety management systems. Saf Sci. (2019) 118:48–56. doi: 10.1016/j.ssci.2019.05.008Al MazroueiMAKhalidKDavidsonRAbdallahS. Impact of organizational culture and perceived process safety in the UAE oil and gas industry. Qual Rep. (2019) 24:3215–38.GaoYFanYWangJLiXPeiJ. The mediating role of safety management practices in process safety culture in the Chinese oil industry. J Loss Prev Process Ind. (2019) 57:223–30. doi: 10.1016/j.jlp.2018.11.017AntonsenSNilsenMAlmklovPG. Regulating the intangible. Searching for safety culture in the Norwegian petroleum industry. Saf Sci. (2017) 92:232–40. doi: 10.1016/j.ssci.2016.10.013SmithSAPatmosAPittsMJ. Communication and teleworking: a study of communication channel satisfaction, personality, and job satisfaction for teleworking employees. Int J Bus Commun. (2018) 55:44–68. doi: 10.1177/2329488415589101LeeY-HLuT-EYangCCChangG. A multilevel approach on empowering leadership and safety behavior in the medical industry: the mediating effects of knowledge sharing and safety climate. Saf Sci. (2019) 117:1–9. doi: 10.1016/j.ssci.2019.03.022LiuSNkrumahENKAkotoLSGyabengENkrumahE. The state of occupational health and safety management frameworks (OHSMF) and occupational injuries and accidents in the Ghanaian oil and gas industry: assessing the mediating role of safety knowledge. Biomed Res Int. (2020) V1:6354895.AsadMMHassanRBSherwaniFAamirMSoomroQMSohuS. Design and development of a novel knowledge-based decision support system for industrial safety management at drilling process: HAZFO expert 1.0. J Eng Des Technol. (2019) 17:705–18. doi: 10.1108/JEDT-09-2018-0167NajiGMAIshaASNAl-MekhlafiA-BASharafaddinOAjmalM. Implementation of leading and lagging indicators to improve safety performance in the upstream oil and gas industry. J Crit Rev. (2020) 7:265–9.SaediAMajidAAIsaZ. Relationships between safety climate and safety participation in the petroleum industry: a structural equation modeling approach. Saf Sci. (2020) 121:240–8.GlinerJAMorganGALeechNL. Research methods in applied settings: An integrated approach to design and analysis. New York: Routledge (2011).SaundersMToseyP. The layers of research design. Winter: Rapport. (2013):58–9. Available at: https://www.csd.uoc.gr/~hy109/resources/layers.pdfVinodkumarMBhasiM. Safety management practices and safety behaviour: assessing the mediating role of safety knowledge and motivation. Accid Anal Prev. (2010) 42:2082–93. doi: 10.1016/j.aap.2010.06.021, PMID: 20728666GuzmanJRecocoGAPadronesJMIgnacioJJ. Evaluating workplace safety in the oil and gas industry during the COVID-19 pandemic using occupational health and safety vulnerability measure and partial least square structural equation modelling. Clean Eng Technol. (2022) 6:100378. doi: 10.1016/j.clet.2021.100378, PMID: 35013732AmmariGAlkurdiBAlshuridehAAlrowwadA. Investigating the impact of communication satisfaction on organizational commitment: a practical approach to increase employees’ loyalty. Int J Mark Stud. (2017) 9:113–33.DahlØKongsvikT. Safety climate and mindful safety practices in the oil and gas industry. J Saf Res. (2018) 64:29–36. doi: 10.1016/j.jsr.2017.12.009, PMID: 29636167NewnamSGoodeN. Communication in the workplace: defining the conversations of supervisors. J Saf Res. (2019) 70:19–23. doi: 10.1016/j.jsr.2019.04.009, PMID: 31847994BrayNJWilliamsL. A quantitative study on organisational commitment and communication satisfaction of professional staff at a master’s institution in the United States. J High Educ Policy Manag. (2017) 39:487–502. doi: 10.1080/1360080X.2017.1354757EhiaguinaE. Development of evidence based factors to enhance safety behaviour in oil and gas industry in the Niger Delta region of Nigeria. Manchester: Manchester Metropolitan University (2022).HairJFRisherJJSarstedtMRingleCM. When to use and how to report the results of PLS-SEM. Eur Bus Rev. (2019) 31:2–24. doi: 10.1108/EBR-11-2018-0203SantosLHaddadAdos SantosIL. Process safety leading indicators in oil storage and pipelines: building a panel of indicators. Chem Eng Trans. (2019) 77:73–8.KaynakRTokluATElciMTokluIT. Effects of occupational health and safety practices on organizational commitment, work alienation, and job performance: using the PLS-SEM approach. Int J Bus Manag. (2016) 11:146–66. doi: 10.5539/ijbm.v11n5p146RamayahTCheahJChuahFTingHMemonMA. Partial least squares structural equation modeling (PLS-SEM) using smartPLS 3.0. An updated guide and practical guide to statistical analysis. Pearson. (2018).AlcantaraJWhettenAAlcantaraJ. Towards a safety culture in chiropractic: the use of the safety, communication, operational reliability, and engagement (SCORE) questionnaire. Complement Ther Clin Pract. (2021) 42:101266. doi: 10.1016/j.ctcp.2020.101266, PMID: 33276224Fernández-MuñizBMontes-PeónJMVázquez-OrdásCJ. The role of safety leadership and working conditions in safety performance in process industries. J Loss Prev Process Ind. (2017) 50:403–15. doi: 10.1016/j.jlp.2017.11.001NealAGriffinMA. A study of the lagged relationships among safety climate, safety motivation, safety behavior, and accidents at the individual and group levels. J Appl Psychol. (2006) 91:946–53. doi: 10.1037/0021-9010.91.4.946, PMID: 16834517XiaNGriffinMAWangXLiuXWangD. Is there agreement between worker self and supervisor assessment of worker safety performance? An examination in the construction industry. J Saf Res. (2018) 65:29–37. doi: 10.1016/j.jsr.2018.03.001, PMID: 29776527ZhaoDTianFSunXZhangD. The effects of entrepreneurship on the enterprises’ sustainable innovation capability in the digital era: the role of organizational commitment, person–organization value fit, and perceived organizational support. Sustain For. (2021) 13:6156. doi: 10.3390/su13116156WangYBielickiJM. Acclimation and the response of hourly electricity loads to meteorological variables. Energy. (2018) 142:473–85. doi: 10.1016/j.energy.2017.10.037KongsvikTGjøsundGViklandKM. HSE culture in the petroleum industry: lost in translation?Saf Sci. (2016) 81:81–9. doi: 10.1016/j.ssci.2015.04.019NajiGMAIshaASNAlazzaniASaleemMSAlzoraikiM. Assessing the mediating role of safety communication between safety culture and employees safety performance. Front Public Health. (2022) 10:840281. doi: 10.3389/fpubh.2022.840281, PMID: 35359765MarininaOTsvetkovaAVasilevYKomendantovaNParfenovaA. Evaluating the downstream development strategy of oil companies: the case of Rosneft. Resources. (2022) 11:4. doi: 10.3390/resources11010004AkinwaleOE. Employee voice: speaking up in organisation as a correlate of employee productivity in oil and gas industry: an empirical investigation from Nigeria. Serb J Manage. (2019) 14:97–121. doi: 10.5937/sjm14-15308BarbazyukEMyachinaKRyakhovRDubrovskayaS. An assessment of the potential effect of oil-and-gas 6–10 kV overhead power lines on bird mortality in the western Orenburg region steppe. IOP Conf Ser Earth Environ Sci. (2021) 817:012012.ILOSTAT. (2024). Statistics on safety and health at work website: International Labour Organization. Available online at: https://ilostat.ilo.org/topics/safety-and-health-at-work/ (Accessed on 11 January 2024)