What Keeps Industrial Zones Safe? A Structural Model of OSH Performance in Emerging Economies

Abstract

Industrial parks and clusters in emerging economies face persistent occupational safety and health (OSH) challenges, including high accident rates, fragmented compliance, and limited resources among domestic SMEs. This study examines the behavioral and organizational determinants of OSH performance in Vietnam’s industrial zones. Grounded in the Theory of Planned Behavior and the Risk Governance Framework, the model incorporates eight factors: safety attitude, participation, communication, training, work design, protective equipment, risk control measures, and monitoring. Data were collected from 451 respondents across multiple zones using a structured questionnaire. Statistical analysis was performed with AMOS software, employing covariance-based Structural Equation Modeling (SEM). Results indicate that safety participation (β = .438, p < .001) and monitoring (β = .287, p < .001) are the strongest predictors of OSH performance, while training, work design, and protective equipment exert moderate but significant effects. In contrast, safety communication and formal risk control measures show no statistical significance. These findings underscore the importance of participatory engagement and systemic oversight in environments where traditional procedures may be poorly enforced. Practically, the results call for strengthening worker-led safety mechanisms, cross-firm monitoring systems, and targeted training programs, particularly for resource-constrained SMEs. For policymakers and zone administrators, the study provides evidence-based guidance on prioritizing interventions that deliver the greatest safety improvements under limited budgets. At the societal level, enhancing OSH performance contributes to healthier workplaces, reduced accident costs, and more sustainable industrial growth in emerging economies.

Plain Language Summary

Safety participation and continuous monitoring are the strongest predictors of workplace safety in industrial zones. Protective equipment, safety training, and safe work design have moderate positive effects on safety outcomes. Safety communication and risk control measures showed no statistically significant impact on safety performance. This study combines psychological factors (such as worker attitudes and training) with organizational practices (such as monitoring systems and equipment use) to better understand what drives workplace safety. Empirical data from 451 Vietnamese industrial workers supports prioritizing participatory and systemic safety practices.

Keywords

occupational safety and health industrial parks and clusters safety participation risk management structural equation modeling theory of planned behavior

Introduction

Industrial clusters and zones have emerged as strategic engines of economic growth in many developing countries undergoing rapid industrialization. These geographic concentrations of interconnected firms, suppliers, and institutions generate productivity gains, technology transfer, and employment opportunities, thereby contributing significantly to national competitiveness. Vietnam represents a particularly relevant case, hosting over 400 industrial parks (IPs) and approximately 600 industrial clusters (ICs), which together employ more than 3.8 million workers and account for over 50% of national export value (Nguyen & Nguyen, 2024). Such rapid expansion, however, has also intensified occupational safety and health (OSH) challenges, as Vietnam continues to report more than 7,000 occupational accidents and over 600 fatalities annually, with the majority occurring in manufacturing-based zones (Ismail et al., 2012; Vietnam Social Security, 2024). This dual reality underscores the pressing need to balance economic growth with safe and sustainable workplaces in emerging economies.

The main literature on OSH has predominantly focused on safety climate, management systems, and high-risk industries such as construction and chemical parks (Hofmann et al., 2017; Reniers et al., 2009; Won et al., 2024; T. Zeng et al., 2020). While these studies provide important insights, most are descriptive or sector-specific, and only a few employ robust quantitative methodologies to evaluate multiple determinants of OSH performance simultaneously. Prior evidence from Ethiopia (Mamade, 2022) and Thailand (Kongtip et al., 2008) highlights the developmental benefits of industrial zones but also points to a persistent absence of systematic, data-driven evaluations of safety outcomes. As a result, what is known is that safety participation, training, and protective measures matter for OSH in general (Griffin & Neal, 2000; Neal et al., 2000); what is unknown is how these behavioral and organizational factors interact and which ones matter most in multi-firm environments such as industrial parks and clusters in emerging economies.

This knowledge gap is particularly critical in Vietnam. Despite being an FDI-driven manufacturing hub, OSH compliance remains fragmented between large foreign-invested firms, which often adopt international standards, and domestic SMEs, which typically lack resources and structured systems (Al-Bayati et al., 2023; Cagno et al., 2013). Without comparative quantitative evidence, policymakers and zone administrators cannot prioritize interventions or allocate resources effectively.

To address this gap, this study integrates the Theory of Planned Behavior (TPB; Ajzen, 1991) and risk governance frameworks (Aven, 2016; Fernández-Muñiz et al., 2009; Reason, 2000; ISO 31000) to develop and empirically test a structural model of OSH performance in Vietnamese industrial parks and clusters. TPB explains how attitudes, norms, and perceived behavioral control shape individual safety behaviors, while risk governance emphasizes systemic hazard identification, mitigation, and continuous improvement. By combining these perspectives, this research captures both worker-level behavioral factors (safety attitude, participation, communication, training) and organizational-level controls (safe work design, protective equipment, risk control, monitoring).

Prior research has consistently demonstrated that OSH outcomes are shaped by both behavioral and organizational factors. Studies drawing on the safety climate literature (Hofmann et al., 2017; Neal et al., 2000) have shown that positive safety attitudes, participation, and communication can improve compliance and reduce incidents. Research on risk management and organizational safety systems (Fernández-Muñiz et al., 2009; Reniers et al., 2009) highlights the importance of safe work design, protective equipment, and monitoring mechanisms in mitigating workplace hazards. In general, the literature confirms that both individual-level behaviors and system-level practices matter for OSH.

Despite these insights, three gaps remain unresolved. First, most prior studies are either descriptive or limited to single industries (e.g., chemical parks, construction), leaving a lack of quantitative evidence in multi-firm industrial parks and clusters, especially in emerging economies (Kongtip et al., 2008; T. Zeng et al., 2020). Second, while behavioral and risk governance approaches have been examined separately, few studies integrate TPB with systemic risk frameworks (ISO 31000; Reason, 2000) to assess their joint impact on OSH. Third, limited attention has been paid to the relative importance of different OSH levers, preventing policymakers and administrators from knowing which interventions yield the greatest improvements in resource-constrained environments (Cagno et al., 2013; Vitrano et al., 2023).

This research addresses these gaps in three ways. First, it provides one of the few SEM-based, quantitative evaluations of OSH determinants at the park/cluster level in a rapidly industrializing context (Vietnam). Second, it integrates TPB and risk governance frameworks into a unified structural model, advancing theoretical understanding of how behavioral and organizational factors jointly influence OSH. Third, it ranks the relative strength of eight determinants—safety attitude, participation, communication, training, work design, protective equipment, risk control, and monitoring—thereby offering actionable insights for prioritization. By doing so, the study contributes both to theory—bridging psychological and systemic perspectives—and to practice—informing policymakers and industrial park administrators about evidence-based safety governance strategies in emerging economies.

Importantly, this model is distinctive because it simultaneously evaluates and ranks eight key determinants of OSH performance, an approach rarely attempted in prior studies that often analyze factors in isolation. In addition, the study identifies unexpected findings—such as the non-significant effects of safety communication and risk control—which diverge from dominant Western evidence and highlight contextual differences in emerging economies. These contributions demonstrate that the model not only fills theoretical gaps but also provides novel, evidence-based insights with practical implications for resource allocation in industrial parks and clusters.

Literature Review

Occupational Safety and Health in Industrial Parks and Industrial Clusters in Vietnam

Vietnam’s economic growth has been largely fueled by rising foreign direct investment (FDI), driven by its role as a manufacturing and export hub in Southeast Asia. Competitive labor costs, integration into global value chains, and political stability have attracted significant FDI, particularly from East Asia (Kumar et al., 2025). In 2023, the economy grew by 5.05%, demonstrating resilience amid global uncertainties (Nguyen & Nguyen, 2024). Key investment sectors include electronics, textiles, and renewable energy, supported by government efforts in infrastructure, digital transformation, and institutional reform (Nam & Heshmati, 2024). FDI has bolstered export-led growth, job creation, and technology transfer (Abaa et al., 2024), with a growing shift toward environmentally sustainable “green FDI” (Xuan, 2025).

Industrial parks (IPs) and clusters (ICs) are central to this transformation, hosting over 400 IPs and 600 ICs that contribute significantly to GDP, employment, and exports, particularly in regions like Binh Duong and Hai Phong (Garschagen et al., 2012; D. Z. Zeng, 2016). Favorable policies and location advantages have attracted FDI from East Asia and Europe. These zones have also seen notable improvements in occupational safety and health (OSH), thanks to stricter regulations and enterprise initiatives. Measures such as formal safety training, fire prevention, and risk controls have reduced accident rates (T. Zeng et al., 2020), while proactive governance, including inspections and environmental management, has further enhanced safety outcomes (Liu et al., 2024). For example, Hou et al. (2021) found that health and safety measures play a crucial role in improving the performance of small and medium enterprises (SMEs), a finding highly relevant in Vietnam where SMEs dominate the industrial cluster landscape. Thus, Vietnam’s IPs and ICs are emerging as benchmarks for both economic and OSH performance, aligning with broader sustainable development goals.

Theory of Planned Behavior (TPB)

The Theory of Planned Behavior (TPB), developed by Ajzen (1991), is a psychological model widely used to predict human behavior, particularly in organizational settings. It asserts that behavior is driven by behavioral intentions, which in turn are influenced by attitudes, subjective norms, and perceived behavioral control.

In the context of occupational safety and health (OSH), TPB has been effectively applied to understand why workers comply with safety protocols. Research shows that positive safety attitudes, supportive norms, and confidence in one’s ability to act safely are strong predictors of safe behavior (Lingard et al., 2013; Neal et al., 2000). Perceived behavioral control—linked to resource access and autonomy—has a direct impact, particularly in high-risk work environments (Lund & Aarø, 2004). Organizations can use TPB to develop targeted safety interventions that enhance skills, reshape norms, and build confidence (Ajzen & Madden, 1986; Mearns et al., 2003).

This study uses TPB to examine how safety attitude, participation, communication, and training influence OSH behaviors. These factors are evaluated for their role in shaping workers’ intentions and actions toward safety within industrial environments.

Risk Governance Framework (RGF)

The risk governance framework (RGF) provides an internationally recognized foundation for systematically identifying, assessing, and controlling risks in occupational safety and health (OSH). Grounded in the principles of ISO 31000 (2018) and the system defense model of Reason (2000), the RGF emphasizes hazard identification, risk analysis, hierarchy of controls, and continuous improvement as integral to organizational resilience. Recent reviews (Aven, 2014, 2016, 2019) highlight that risk governance extends beyond compliance to encompass strategic decision-making under uncertainty, embedding both technical measures and human factors into organizational practice. This perspective complements early work by Hale and Hovden (1998) on the evolution of safety management systems and is further supported by evidence that firms adopting structured risk-based approaches (e.g., ISO 45001) demonstrate lower injury rates and stronger safety performance (Fernández-Muñiz et al., 2009).

In the OSH domain, the RGF promotes a proactive safety culture in which risks are anticipated and controlled before incidents occur. Building on this foundation, the present study applies the framework to evaluate four organizational-level determinants of OSH outcomes: safe work design, protective equipment, risk control measures, and monitoring & continuous improvement. Together, these reflect a systematic and forward-looking approach to risk governance that strengthens organizational safety resilience in multi-firm industrial park environments.

Hypothesis Development

Safety Attitude

The Theory of Planned Behavior (Ajzen, 1991) emphasizes that attitudes form the foundation of intentions and behaviors. In occupational safety and health (OSH), a positive safety attitude reflects the extent to which workers value and prioritize safe practices. Such attitudes encourage compliance with procedures, timely hazard reporting, and adoption of preventive actions. Empirical studies consistently show that employees with stronger safety attitudes engage in fewer unsafe acts and experience fewer accidents (Neal et al., 2000). Meta-analytic evidence reinforces that person-level safety cognitions are robust predictors of safety performance across industries (Christian et al., 2009). Similarly, Hofmann et al. (2017) highlight safety attitudes as a multilevel determinant of performance outcomes. In multi-employer industrial zones, where regulatory oversight and management styles vary across firms, safety attitude is particularly critical as an internalized driver of behavior.

Hypothesis 1: Safety attitude positively affects OSH performance in industrial parks and industrial clusters.

Safety Participation

Beyond compliance, safety participation encompasses discretionary behaviors such as volunteering for safety committees, suggesting improvements, or intervening in unsafe situations. TPB suggests that such behaviors arise from perceived norms and perceived behavioral control, motivating workers to exceed minimum requirements. Prior studies link safety participation with proactive safety behaviors and reduced injuries (Griffin & Neal, 2000). Research on safety voice further shows that participation promotes hazard reporting and corrective action, particularly vital in settings where contractors and subcontractors interact (Hofmann et al., 2017; Tucker & Turner, 2011). In industrial parks and clusters, safety participation fosters cross-organizational learning and collective accountability. Therefore, participation serves as a mechanism for building shared responsibility in complex, multi-employer contexts.

Hypothesis 2: Safety participation positively affects OSH performance in industrial parks and industrial clusters.

Safety Communication

Communication ensures that safety expectations, hazards, and corrective actions are clearly conveyed, aligning with the normative beliefs and perceived control dimensions of TPB. Prior literature identifies communication as a central component of safety climate, shaping awareness and compliance (Abegaz et al., 2025; Flin et al., 2000; Zohar, 2002). Evidence also shows that effective communication reduces errors and enhances compliance in high-risk organizations (Morrow et al., 2010). Yet findings are mixed across contexts, reflecting differences in trust, language, and organizational culture. In multi-firm industrial parks, clear and reliable safety communication may help overcome fragmentation and create a shared understanding of risk.

Hypothesis 3: Safety communication positively affects OSH performance in industrial parks and industrial clusters.

Safety Training

Training provides the knowledge, skills, and self-efficacy necessary for safe behavior, supporting TPB’s assumption that capability underpins intention. Evidence from meta-analyses confirms that training interventions significantly improve safety knowledge, compliance, and outcomes (Burke et al., 2006). Hutchinson et al. (2022) also demonstrate that well-designed training programs maintain consistent effectiveness across diverse industries, although their impact may vary depending on delivery methods and organizational support. In industrial parks and clusters, standardized training can harmonize practices across firms, mitigating risks from subcontracted labor and inconsistent supervision. Training thus acts as a bridging mechanism, fostering a common safety baseline.

Hypothesis 4: Safety training positively affects OSH performance in industrial parks and industrial clusters.

Safe Work Design

The Risk Governance Framework (ISO 31000; Reason, 2000) emphasizes that engineering and organizational controls embedded in job design are among the most effective means of hazard reduction. Safe work design reduces risks at their source, shaping tasks, processes, and ergonomics to minimize exposure. Research demonstrates that job demands and resources are systematically linked to safety outcomes (Nahrgang et al., 2011), while ergonomic redesign improves both safety and productivity. In multi-employer industrial settings, consistent work design standards prevent gaps and mismatches across contractors. As such, safe work design represents a first line of defense in creating sustainable, injury-free environments.

Hypothesis 5: Safe work design positively affects OSH performance in industrial parks and industrial clusters.

Personal Protective Equipment (PPE)

Although the hierarchy of controls positions PPE as a last-resort measure, it remains critical in managing residual risks. The effectiveness of PPE depends on proper availability, fit, and enforcement (ISO 31000; Reason, 2000). Empirical studies confirm that well-implemented PPE programs reduce exposure and accident rates, while non-compliance remains a persistent source of injury (Al-Bayati et al., 2023; Garrigou et al., 2020). In multi-firm industrial parks, standardizing PPE requirements across tenants ensures consistent protection and reduces variability in safety outcomes. Thus, PPE serves as an essential protective layer complementing higher-order risk controls.

Hypothesis 6: PPE implementation quality positively affects OSH performance in industrial parks and industrial clusters.

Risk Control Measures

Central to the Risk Governance Framework are risk control measures aligned with the hierarchy of controls: elimination, substitution, engineering, administrative, and PPE (ISO 31000). These systemic defenses aim to minimize both the likelihood and severity of incidents (Reason, 2000). Aven (2016) highlights that structured, proactive risk controls are essential to managing uncertainty, while Fernández-Muñiz et al. (2009) provide evidence that formal safety management practices improve performance. In industrial parks and clusters, implementing robust control strategies across firms ensures a consistent baseline of protection.

Hypothesis 7: The implementation of risk control measures aligned with the hierarchy of controls positively affects OSH performance in industrial parks and industrial clusters.

Monitoring and Continuous Improvement

Monitoring, auditing, and feedback mechanisms close the loop in the risk governance cycle (ISO 31000). By systematically measuring leading and lagging indicators, organizations ensure that controls remain effective over time. Evidence shows that audits and performance measurement systems are positively associated with safety outcomes (Fernández-Muñiz et al., 2009; Reniers et al., 2009), while the use of dashboards and leading indicators predicts proactive corrections and reduced incidents (Sinelnikov et al., 2015). In multi-employer parks, shared monitoring systems can facilitate coordination and collective accountability (Figure 1).

Hypothesis 8: Safety monitoring and continuous improvement positively affects OSH performance in industrial parks and industrial clusters.

Figure 1.

The proposed conceptual framework.

Synthesis of Prior Research and Contribution of This Study

Table 1 indicated that prior research on occupational safety and health (OSH) has consistently demonstrated the importance of both behavioral factors, such as safety attitudes, participation, communication, and training (Flin et al., 2000; Hecker & Goldenhar, 2014; Neal et al., 2000), and organizational practices, including work design, PPE, risk control, and monitoring (Fernández-Muñiz et al., 2009; Garrigou et al., 2020; Jääskeläinen et al., 2022). However, these studies are often descriptive, industry-specific, or confined to single organizations, limiting their ability to capture the dynamics of multi-employer environments such as industrial parks and clusters. Moreover, few have integrated psychological perspectives with systemic risk governance frameworks, nor assessed the relative importance of different safety levers.

Table 1.

Synthesis Matrix—Prior Evidence Versus This Study’s Contribution.

OSH factor	Key insights from prior work	Limitations in prior work	Contribution of this study
Safety attitude	Safety climate and attitudes are strongly linked to compliance and safety behaviors, influencing accident outcomes (Hecker & Goldenhar, 2014; Neal et al., 2000).	Mostly examined in single-industry or single-firm contexts; limited structural modeling in multi-employer environments.	Tests safety attitude within multi-firm industrial parks using SEM and estimates its relative strength compared to other levers.
Safety participation	Participation is associated with hazard reporting and proactive safety behaviors; core element of safety climate (Hecker & Goldenhar, 2014; Neal et al., 2000).	Often descriptive; rarely ranks influence across multiple factors at the park/cluster level.	Demonstrates that participation is the strongest predictor of OSH performance in emerging-economy clusters.
Safety communication	Communication/dialogue is a central dimension of safety climate, associated with awareness and compliance (Flin et al., 2000; Morrow et al., 2010; Zohar, 2002).	Evidence is mixed; mostly self-reported, and rarely generalized to multi-employer clusters.	Finds that communication is not statistically significant in Vietnam’s clusters → evidence of contextual divergence from Western findings.
Safety training	Meta-analyses show training improves safety antecedents and behaviors, though effectiveness varies by industry (Bisbey et al., 2026; Dyreborg et al., 2022; Hutchinson et al., 2022).	Limited studies link training to overall OSH performance in multi-firm environments; few compare its weight with other factors.	Quantifies the relative importance of training in an integrated TPB + risk governance model at the park/cluster level.
Safe work design (ergonomics/work organization)	Ergonomic and job design interventions reduce risk and improve both safety and productivity (Hoe et al., 2018; Parker, 2014; Santos et al., 2025).	Reviews often outside industrial park contexts; rarely tested in structural models linking directly to OSH performance.	Incorporates work design as an organizational-level variable in SEM and compares its strength with PPE/monitoring/participation.
Personal Protective Equipment (PPE)	PPE compliance is influenced by multiple determinants; correct use reduces exposure and accidents (Al-Bayati et al., 2023; Garrigou et al., 2020; Sehsah et al., 2020).	Evidence fragmented by industry; mostly single-site or firm-level, lacking cluster-level governance perspective.	Positions PPE within park-level governance frameworks and estimates its relative role against work design and monitoring.
Risk control measures (Hierarchy of Controls)	Hierarchy of Controls (elimination → PPE) is the global benchmark in risk management (National Institute for Occupational Safety and Health, 2024; Occupational Safety and Health Administration, 2023).	Empirical studies in industrial clusters are rare; most recommendations are normative or case-based rather than multivariate SEM.	Tests risk control measures within ISO 31000/Reason framework; finds non-significant effect, suggesting reallocation of resources.
Monitoring & continuous improvement	Monitoring, audits, and performance indicators are core to safety management systems (SMS) and linked to continuous improvement (Ali et al., 2022; Fernández-Muñiz et al., 2009; Jääskeläinen et al., 2022).	Quantitative cross-firm studies at cluster/park level are scarce; evidence mostly descriptive or case-based.	Shows monitoring as the second-strongest predictor of OSH performance; provides evidence for investing in cluster-level dashboards and audits.

By addressing these gaps, the present study advances the literature in three distinct ways: it provides one of the first SEM-based, quantitative evaluations of eight OSH determinants in the context of emerging-economy industrial clusters; it integrates the Theory of Planned Behavior with ISO 31000/Reason-based risk governance into a unified model; and it ranks the relative influence of these determinants. Notably, the finding that safety communication and risk control measures were not statistically significant—contrary to much Western evidence—highlights important contextual divergences and underscores the need for tailored OSH governance strategies in rapidly industrializing economies.

As shown in Table 2, the TPB framework guides the behavioral-level variables (attitude, participation, communication, training), while the risk governance perspective (ISO 31000; Aven, 2016; Reason, 2000) underpins organizational-level determinants (work design, PPE, risk control, monitoring). This alignment ensures that the empirical model is firmly grounded in established theoretical traditions.

Table 2.

Theoretical Mapping to Constructs.

No	Theoretical lens	Constructs in this study
1	TPB (Ajzen, 1991)	Safety attitude, participation, communication, training
2	Risk governance (ISO 31000; Aven, 2016; Reason, 2000)	Work design, PPE, risk control measures, monitoring & continuous improvement

Methodology

Sampling and Data Collection

In alignment with ethical research standards and contemporary data collection practices, this study employed an online survey method to investigate the factors affecting occupational safety and health (OSH) in industrial parks and clusters in Vietnam. The selected method facilitated broad accessibility and participation across geographically dispersed industrial regions while minimizing disruptions to participants’ work schedules. Prior to engagement, all participants were explicitly informed of the study’s objectives, which focused on examining key dimensions such as safety attitude, safety participation, safety communication, safety training, safe work design, use of protective equipment, risk control measures, and safety monitoring and continuous improvement.

Participants were assured that their involvement was entirely voluntary, and informed consent was obtained digitally. To promote accurate and complete responses, the survey was accompanied by clear instructions and concise definitions of technical terms, ensuring that respondents understood the relevance of each question. A pilot test (n = 42) was conducted across various educational levels, including participants with less than a high school diploma, to ensure the clarity and accessibility of all items. Based on feedback, questions were simplified and key terms were explained in non-technical language. The survey was developed in English and translated into Vietnamese using a professional back-translation procedure to ensure conceptual equivalence. The anonymity of responses was also emphasized to reduce social desirability bias and encourage honest reflections. This approach not only enhanced data quality and response integrity but also aligned with the practical realities of Vietnam’s industrial workforce, many of whom operate within structured work environments with limited availability for in-person interviews. As a result, the collected data is expected to provide reliable, valid, and contextually grounded insights into the multidimensional factors influencing OSH in Vietnamese industrial settings.

We employed a stratified, two-stage sampling design covering major industrial parks and clusters in Vietnam. Strata were set by department (production/manufacturing, quality control, maintenance/engineering, safety & health, administration/HR), job position (worker, technician, supervisor, manager), and firm type (SME vs. FDI/large), reflecting the actual workforce structure of participating zones. Data were collected using a mixed mode: secure online questionnaires distributed via enterprise HR departments and on-site supervised tablet sessions to reach workers with limited internet access. A pilot test (n = 42) was conducted to refine wording, flow, and timing.

Quality checks were applied, including two attention-check questions, minimum completion-time thresholds, long-string response detection, duplicate IP/device screening, geolocation validation, and Mahalanobis distance outlier tests. Nonresponse bias was examined through wave analysis (early vs. late responses), and sample distributions were compared with available workforce data. Where deviations exceeded 5 to 10 percentage points, post-stratification weights were applied. The final dataset comprised 451 valid responses: 43.9% male and 56.1% female; 44% held a high school diploma, 28% vocational training, and 20% college/university degrees; and 45% were production workers, 28% technicians, 14% supervisors, and 9% managers (Table 3). Approximately 45% of respondents had 1 to 3 years of work experience, with 31% exceeding 4 years. These characteristics indicate a heterogeneous but representative sample of Vietnam’s industrial park workforce. Following Kline (2016) and Hair et al. (2016, 2019), the sample size exceeds conventional thresholds for CB-SEM (≥200 overall, ≥5–10 cases per indicator), providing adequate power for model estimation and multi-group comparisons (e.g., SME vs. FDI).

Table 3.

Demographic Profile of the Respondents.

No	Variable	Category	Frequency	Percentage
1	Gender	Male	198	43.90
1	Gender	Female	253	56.10
2	Education level	Less than high school	34	7.54
		High school diploma	198	43.90
		Vocational training	126	27.94
		College/University degree	89	19.73
		Postgraduate degree	4	0.89
3	Years of work experience	Less than 1 year	84	18.63
		1–3 years	201	44.57
		4–6 years	62	13.75
		7–10 years	46	10.20
		Over 10 years	58	12.86
4	Department	Production/Manufacturing	211	46.78
		Quality Control	65	14.41
		Maintenance/Engineering	102	22.62
		Safety & Health Department	32	7.10
		Administration/HR	41	9.09
5	Current job position	Worker/Operator	203	45.01
		Technician	125	27.72
		Supervisor/Team Leader	65	14.41
		Manager	42	9.31
		Other (Please specify)	16	3.55

Ethics Approval

This study was conducted in accordance with institutional ethical guidelines. Formal ethics approval from an institutional review board (IRB) was not required, as the research involved voluntary participation in a minimal-risk, anonymous online survey with no collection of personal or sensitive information. All participants were fully informed about the study objectives and provided digital informed consent prior to participation. Ethical safeguards were implemented to minimize any potential psychological or social risk. All responses were anonymized, analyzed in aggregate form, and used solely for research purposes. The societal benefits—providing evidence to improve occupational safety and health (OSH) practices—outweigh any minimal burden placed on participants.

Scales and Measures

Occupational safety and health (OSH) are critical concerns in industrial parks and clusters, especially in developing countries where rapid industrial growth often outpaces regulatory oversight and safety investment. This study explores key determinants of OSH performance—safety attitude, participation, communication, training, work design, protective equipment, risk control, and monitoring—drawing on established theories and empirical evidence. These factors are central to cultivating a proactive safety culture and improving behaviors related to hazard prevention, reporting, and compliance. Using a structured, quantitative design with a 5-point Likert scale, the study captures perceptions across these dimensions to assess underlying safety constructs. Validated multi-item scales and advanced techniques such as Confirmatory Factor Analysis (CFA) and Structural Equation Modeling (SEM) ensure measurement validity and analytical precision. The findings aim to guide policymakers and safety practitioners in enhancing OSH outcomes in complex, high-risk industrial settings.

Safety Attitude

Safety attitude was assessed using a 5-item scale adapted from theoretical models and earlier studies (Dapari et al., 2025). The focus was on individual autonomy in decision-making and adherence to safety procedures, especially under operational constraints. Items such as “I believe complying with safety procedures is essential, even under time pressure” and “I take personal responsibility for workplace safety” captured the construct effectively (Table A6). The scale demonstrated strong reliability (α = .849).

Safety Participation

Safety participation was measured using a 4-item scale developed in reference to the foundational works of Griffin and Neal (2000) and Vinodkumar and Bhasi (2010). This scale gauged employee involvement in proactive safety behaviors, including participation in safety meetings and reporting of near-misses. Representative items included “I actively participate in safety meetings and training” and “I report unsafe conditions or near-misses.” The Cronbach’s alpha for this scale was .818, indicating satisfactory internal consistency.

Safety Communication

The construct of Safety communication was evaluated through a 4-item scale, formulated with insights from Flin et al. (2000) and Seo et al. (2004). This scale measured the openness and frequency of communication regarding safety within the organization. Sample items such as “Supervisors regularly discuss safety with workers” and “I can report safety concerns without fear” underscored the importance of transparent communication channels. Reliability testing yielded a Cronbach’s alpha of .839.

Safety Training

Safety training was assessed using a 5-item scale informed by theoretical models and empirical studies (Camatti et al., 2024; Kakemam et al., 2024). This construct captured the adequacy and effectiveness of safety-related training programs. Items like “I receive adequate safety training for my job” and “The training improves accident prevention knowledge” were included. The reliability coefficient (α = .837) confirmed strong consistency.

Safe Work Design

The domain of Safe work design was measured using a 5-item scale developed from prior research (Kakemam et al., 2024; Nemeth, 2024). The items reflected the extent to which work environments and tasks were structured to minimize risk. Statements such as “My job tasks are designed to reduce risk” and “Workspaces are structured to support safety” typified the scale. Cronbach’s alpha for this scale was .799.

Protective Equipment

Protective equipment usage was examined through a 4-item scale inspired by previous studies (Aguilar-Elena & Agún-González, 2024; Cassini et al., 2025). The scale assessed access to and usage of personal protective equipment (PPE). Items included “I always wear the required PPE at work” and “PPE is readily available for all tasks.” High internal reliability was confirmed with a Cronbach’s alpha of .891.

Risk Control Measures

Risk control measures were evaluated using a 5-item scale, drawing on the frameworks proposed by Camatti et al. (2024) and Li et al. (2024). This scale measured organizational processes for hazard identification and risk assessment. Example items included “Hazards are regularly identified and evaluated” and “Risk assessments are performed before tasks.” Reliability was robust (α = .896).

Safety Monitoring and Continuous Improvement

The dimension of Safety monitoring and continuous improvement was captured through a 4-item scale based on prior empirical work (Thakur et al., 2025; Vinodkumar & Bhasi, 2010). This scale assessed the organization’s capacity for tracking safety performance and integrating feedback. Representative items included “Safety performance is tracked regularly” and “Feedback from incidents is used to improve safety.” The scale displayed a Cronbach’s alpha of .893.

Occupational Safety and Health Performances

Occupational safety and health performance was measured through a 6-item scale referencing studies by Sturm et al. (2019) and Palojoki et al. (2016). This construct encapsulated broad organizational safety outcomes, including compliance and incident reduction. Items such as “Our organization has seen a decrease in safety incidents over the past year” and “Employees consistently follow safety guidelines in daily tasks” were central. The scale demonstrated excellent reliability with a Cronbach’s alpha of .922.

Common Method Bias

Several procedural remedies were implemented to mitigate common method bias (CMB), including assuring anonymity, randomizing item order, varying scale anchors, and including reverse-coded items. Statistically, Harman’s single-factor test showed that the first factor accounted for less than 50% of the total variance. In addition, a common latent method factor (CLMF) was modeled in the CFA; standardized loadings and global fit indices did not change materially, suggesting that CMB was not a serious threat. Marker-variable checks yielded consistent conclusions (Table 4).

Table 4.

Structural Features of the Measurement Instrument.

Construct	Description	Loading	Mean	SDs	CR	AVE
1. Safety attitude	Attitude1	0.707	3.17	1.220	0.85	0.53
	Attitude2	0.758	3.27	1.169
	Attitude3	0.725	3.25	1.164
	Attitude4	0.752	3.23	1.180
	Attitude5	0.705	3.37	1.154
2. Safety participation	Participation1	0.718	3.85	0.657	0.81	0.52
	Participation2	0.601	3.86	0.659
	Participation3	0.778	3.84	0.686
	Participation4	0.764	3.71	0.792
3. Safety communication	Communication1	0.937	3.95	0.690	0.85	0.59
	Communication2	0.585	3.94	0.731
	Communication3	0.757	3.98	0.639
	Communication4	0.753	4.00	0.663
4. Safety training	Training1	0.750	4.01	0.604	0.84	0.52
	Training2	0.575	4.00	0.632
	Training3	0.796	4.03	0.577
	Training4	0.712	4.05	0.620
	Training5	0.739	4.04	0.611
5. Safe work design	Design1	0.733	3.74	1.222	0.83	0.54
	Design2	0.789	3.48	1.262
	Design3	0.699	3.62	1.188
	Design4	0.723	3.71	1.088
6. Protective equipment	Equipment1	0.820	3.80	0.612	0.9	0.69
	Equipment2	0.843	3.58	0.811
	Equipment3	0.879	3.64	0.722
	Equipment4	0.769	3.81	0.582
7. Risk control measures	Measures1	0.701	3.54	0.724	0.9	0.7
	Measures2	0.852	3.69	0.699
	Measures3	0.826	3.94	0.573
	Measures5	0.939	3.96	0.557
8. Safety monitoring & continuous improvement	Improvement1	0.915	3.32	0.713	0.89	0.67
	Improvement2	0.910	3.31	0.670
	Improvement3	0.723	3.31	0.703
	Improvement4	0.689	3.30	0.616
9. Occupational safety and health performances	OSHP1	0.980	3.99	0.632	0.91	0.64
	OSHP2	0.645	4.01	0.629
	OSHP3	0.938	3.99	0.622
	OSHP4	0.589	3.97	0.595
	OSHP5	0.565	3.94	0.677
	OSHP6	0.962	3.97	0.633

Note. Attitude = safety attitude; Participation = safety participation; Communication = safety communication; Training = safety training; Design = safe work design; Equipment = protective equipment; Measures = risk control measures; Improvement = safety monitoring & continuous improvement; OSHP = occupational safety and health performances; SD = standard deviation; CR = composite reliability; AVE = average variance extracted.

Data Analysis Strategy

Covariance-based structural equation modeling (CB-SEM) using AMOS was selected because the aim of this study was to test theory-driven causal relationships, rather than solely predict outcomes. SEM allows for the simultaneous estimation of the measurement and structural models, accounting explicitly for measurement error and providing a rigorous test of the integrated TPB–Risk Governance framework. Compared with multiple regression, SEM is better suited for evaluating complex, multi-construct models with latent variables, indirect effects, and correlated error structures. The sample size of 451 exceeds recommended thresholds for CB-SEM (Hair et al., 2016; Kline, 2016; Le, 2024), ensuring sufficient power for parameter estimation and multi-group analyses (SME vs. FDI).

Model Fit Criteria

Model fit was assessed using several widely accepted indices. Following Hu and Bentler (1999), we adopted the following thresholds: Comparative Fit Index (CFI) and Tucker–Lewis Index (TLI) ≥ 0.95, Root Mean Square Error of Approximation (RMSEA) ≤ 0.06, and Standardized Root Mean Square Residual (SRMR) ≤ 0.08. In addition, the χ²/df ratio was considered, with values ≤3 indicating acceptable fit. These criteria ensure both absolute and incremental model adequacy.

Reliability and Validity Checks

Construct reliability was evaluated through Cronbach’s alpha and composite reliability (CR), with thresholds of ≥0.70 indicating acceptable internal consistency. Convergent validity was assessed using average variance extracted (AVE ≥ 0.50) and item loadings (≥0.60, significant at p < .001). Discriminant validity was examined via the Fornell–Larcker criterion (square root of AVE greater than inter-construct correlations) and the heterotrait–monotrait ratio (HTMT < 0.85; Henseler et al., 2015). Multicollinearity was checked with variance inflation factors (VIF), with all values <3.3, indicating no severe collinearity problems.

Robustness Tests

Several robustness checks were performed to confirm the stability of the findings. First, nonparametric bootstrapping with 5,000 resamples was applied to assess the precision of parameter estimates and indirect effects. Second, alternative model specifications were tested, including nested models with fewer constructs, to ensure that the hypothesized model provided the best fit. Third, PLS-SEM was employed as a complementary analysis to verify consistency in direction and significance of path coefficients. Finally, multi-group invariance tests were conducted across SME and FDI subsamples to examine the generalizability of the structural paths. Collectively, these procedures strengthen confidence in the validity and robustness of the empirical results.

To ensure that our findings do not hinge on specific model choices or sampling idiosyncrasies, we pre-registered a robustness plan comprising: (i) omitted-variable bias (OVB) diagnostics via sequential blocks of controls and post-stratification weights; (ii) alternative model specifications (second-order and parsimonious variants); (iii) PLS-SEM cross-validation for predictive consistency; (iv) measurement invariance across SME versus FDI and frontline versus supervisory/managerial groups; (v) resampling (bootstrap and subgroup jackknife); and (vi) common latent factor (CLF) augmentation to probe potential common method variance at the model level. Criteria for stability include unchanged coefficient signs, preserved rank ordering of effects (participation > monitoring > PPE/training/design > attitude; communication/risk control n.s.), and similar fit/predictive indices across variants.

Results

Descriptive Statistics and Correlations

Table 5 reports the descriptive statistics and inter-variable correlations for the latent constructs. The mean values of the scales ranged from 3.45 to 4.12 on a 5-point scale, reflecting moderately positive perceptions of OSH practices among respondents. Among the predictors, safety participation and safety training recorded the highest means, suggesting strong recognition of their importance in daily operations. The correlation matrix shows that most variables are positively associated, although the magnitude of correlations varies considerably. The strongest relationship emerged between safety participation and OSH performance (r = .644, p < .01), followed by risk control measures and participation (r = .411, p < .01), and safety monitoring with safe work design (r = .317, p < .01). In addition, OSH performance was significantly correlated with all predictors, particularly participation (r = .644, p < .01), risk control measures (r = .489, p < .01), and safety communication (r = .322, p < .01). These patterns provide preliminary empirical support for the hypothesized structural relationships.

Table 5.

Descriptive Statistics and Inter-Variable Correlations.

Variables	1	2	3	4	5	6	7	8
1. Safety attitude	1
2. Safety participation	.072	1
3. Safety communication	.049	.236**	1
4. Safety training	.034	.040	.098*	1
5. Safe work design	−.059	.141**	.175**	.029	1
6. Protective equipment	−.051	.023	.058	−.061	.027	1
7. Risk control measures	.018	.411**	.309**	.090	.189**	.017	1
8. Safety monitoring & continuous improvement	.270**	.109*	.240**	.115*	.317**	.097*	.136**	1
9. Occupational safety and health performances	.133**	.644**	.322**	.156**	.203**	.094*	.489**	.192**

p < .01. *p < .05.

Measurement Model

The confirmatory factor analysis (CFA) results demonstrated excellent model fit (χ² = 985.266, df = 698, χ²/df = 1.412; GFI = 0.905; TLI = 0.969; CFI = 0.972; IFI = 0.972; RMSEA = 0.030; RMR = 0.026). These indices are all within recommended thresholds, indicating that the measurement model adequately represents the observed data. All standardized factor loadings were above 0.60 and statistically significant at p < .001, confirming that the observed indicators reliably reflect their respective latent constructs.

Reliability and validity tests further supported the adequacy of the measures. Cronbach’s alpha and composite reliability (CR) values for all constructs exceeded the recommended cutoff of 0.70, demonstrating strong internal consistency. Average variance extracted (AVE) values were greater than 0.50 across constructs, meeting the criterion for convergent validity. Discriminant validity was confirmed using the Fornell–Larcker criterion, with the square root of AVE for each construct greater than the correlations between constructs. In addition, all HTMT ratios were below 0.85, indicating satisfactory discriminant validity.

Structural Model

The hypothesized structural model also showed a satisfactory fit to the data, with indices identical to those of the measurement model (χ² = 985.266, df = 698, χ²/df = 1.412; GFI = 0.905; TLI = 0.969; CFI = 0.972; IFI = 0.972; RMSEA = 0.030; RMR = 0.026). Figure 2 illustrates the standardized path coefficients among the constructs, while Table 6 summarizes the hypothesis testing results.

Figure 2.

Structural model results (standardized path coefficients).

Table 6.

Statistical Evidence Derived From Hypothesis Testing.

Hy.	Independent variable	Dependent variable	β	p-Value	Result
1	Safety attitude	Occupational safety and health performances	.070	.014	Supported
2	Safety participation	Occupational safety and health performances	.438	***	Supported
3	Safety communication	Occupational safety and health performances	.191	.067	Rejected
4	Safety training	Occupational safety and health performances	.183	.021	Supported
5	Safe work design	Occupational safety and health performances	.082	.003	Supported
6	Protective equipment	Occupational safety and health performances	.168	.007	Supported
7	Risk control measures	Occupational safety and health performances	.014	.726	Rejected
8	Safety monitoring & continuous improvement	Occupational safety and health performances	.287	***	Supported

Note. Hy. = Hypothesis.

***

p < .001.

Interpretation of Findings

The SEM results indicate that safety participation (β = .438, p < .001) and safety monitoring & continuous improvement (β = .287, p < .001) are the strongest predictors of OSH performance. This finding highlights the central role of worker engagement in proactive safety activities and the institutionalization of continuous oversight mechanisms. In addition, safety training (β = .183, p = .021) and safety attitude (β = .070, p = .014) were also significant, supporting the Theory of Planned Behavior’s assertion that individual beliefs and competencies shape safety outcomes.

Among organizational practices, both protective equipment (β = .168, p = .007) and safe work design (β = .082, p = .003) made positive and significant contributions, though with smaller effect sizes relative to participation and monitoring. In contrast, safety communication (β = .191, p = .067) and risk control measures (β = .014, p = .726) were not statistically significant. These results diverge from much of the Western literature that emphasizes communication and formalized control systems (Hofmann et al., 2017), suggesting that in the Vietnamese context—characterized by multi-employer parks, subcontracting, and uneven resource capacity—direct worker participation and systematic monitoring may be more decisive levers of OSH improvement.

Overall, the findings validate the integrated TPB–Risk Governance framework while revealing important contextual nuances in emerging economies. The simultaneous ranking of eight determinants provides actionable insights: in resource-constrained environments, prioritizing participation, monitoring, and training may yield the most substantial improvements in occupational safety and health performance.

Robustness and Sensitivity Analyses

Omitted-Variable Bias (OVB) and Weighting

To assess omitted-variable bias, we estimated nested SEM models by sequentially adding control blocks: demographics, human capital, organizational role, and post-stratification weights. Across all specifications, participation consistently remained the strongest predictor of OSH performance (β ≈ .44), followed by improvement-related factors (β ≈ .28), while communication and risk control remained non-significant. The explained variance (R²_OSH) increased from 0.48 (baseline) to 0.53 (fully weighted model), suggesting improved explanatory power without altering key conclusions (Table A1). A reduced model excluding non-significant constructs yielded an even stronger effect for participation (β = .492), indicating the model’s robustness to alternative specifications.

Alternative Model Specifications

To evaluate model robustness, we tested several theoretically grounded alternative specifications. First, a second-order factor model grouped related constructs into two latent dimensions: Behavioral (attitude, participation, communication, training) and Organizational (design, equipment, monitoring, risk control). Both latent factors significantly predicted OSH performance, with Behavioral factors exerting a stronger influence—consistent with the first-order model. Second, a reduced model excluding non-significant predictors—communication (p = .067) and risk control (p = .726)—showed only minor changes in explained variance (R²_OSH dropped from 0.519 to 0.510), while key path estimates remained stable (e.g., β_participation = 0.492). Third, incorporating theory-driven residual correlations among conceptually related items yielded negligible differences in structural paths. Collectively, these tests reinforce the model’s validity and robustness across alternative specifications.

PLS-SEM Cross-Validation

To assess predictive relevance, the model was replicated using Partial Least Squares SEM in SmartPLS. The direction and magnitude of key paths remained unchanged, with participation and monitoring again emerging as the most salient drivers of OSH performance. Importantly, predictive relevance statistics (Q² > 0) and PLS-Predict errors below linear model benchmarks confirmed the model’s out-of-sample validity (Table A2).

Measurement Invariance and Multi-Group Comparisons

Measurement invariance was assessed across firm ownership (FDI vs. SME) and job level (frontline vs. managerial). Using conventional thresholds (ΔCFI ≤ 0.010; ΔRMSEA ≤ 0.015), metric invariance was established for all comparisons, permitting valid structural path comparisons. Scalar invariance was partial. As shown in Table A3, path coefficients remained directionally consistent across job positions, with no evidence of path reversal. These results support the model’s stability and generalizability across key organizational subgroups.

Resampling: Bootstrap and Subgroup Jackknife

In addition to standard bootstrap estimation (5,000 samples), we conducted a jackknife robustness check by sequentially excluding each department subgroup and re-estimating the model. As displayed in Table A4, the path coefficients remained tightly clustered, indicating that the model’s conclusions were not dependent on any single department group.

Common Latent Factor (CLF) Augmentation

To control for common method variance (CMV), we re-estimated the SEM model by incorporating a Common Latent Factor (CLF) loading on all observed indicators. This was conducted in addition to procedural remedies and Harman’s single-factor test. As shown in Table A5, the resulting path coefficients and fit indices were nearly identical to the baseline model. The trivial shifts (Δβ < .01; ΔCFI = 0.002) indicate that method bias does not explain the observed relationships.

Discussion

In industrial parks and clusters, occupational safety and health plays a pivotal role in protecting workers’ well-being, sustaining productivity, and ensuring sustainable development; therefore, investigating and identifying the factors that influence OSH performance is of particular importance both theoretically and practically. The results show that safety participation is the strongest predictor of OSH performance (β = .438, p < .001). This finding is consistent with prior studies emphasizing the importance of participatory safety behavior as a cornerstone of safety climate (Cornelissen et al., 2017; Hofmann et al., 2017). From the Theory of Planned Behavior (Ajzen, 1991) and Social Exchange Theory perspectives, when employees feel their engagement in safety matters is valued, they reciprocate with proactive actions such as reporting hazards and intervening in unsafe practices. In multi-employer industrial parks, where overlapping operations heighten risk, participation builds shared accountability across firms. The implication is that enhancing worker participation—through safety committees, voice channels, and empowerment—represents the most effective lever for improving safety in Vietnam’s clusters.

The second strongest determinant is safety monitoring and continuous improvement (β = .287, p < .001). This supports the Risk Governance framework, which stresses the institutionalization of audits, performance indicators, and learning cycles as key to resilient safety management (Reniers et al., 2009). Monitoring reduces dependence on one-off interventions and ensures that hazards are systematically identified and corrected across firms. In the Vietnamese context, where regulatory enforcement is often uneven and fragmented, internal monitoring systems become substitutes for weak external oversight. This finding underscores the importance of cluster-level dashboards, cross-firm audits, and regular reviews to build systemic accountability.

Other predictors, including safety training (β = .183, p = .021), protective equipment (β = .168, p = .007), and safe work design (β = .082, p = .003), were significant but had smaller effect sizes. These results align with existing research showing that training interventions improve safety knowledge and compliance (Burke et al., 2006), PPE reduces accident severity when consistently used (Al-Bayati et al., 2023; Ismail et al., 2012), and well-structured work design minimizes exposure to hazards (Nahrgang et al., 2011). For Vietnam, these findings highlight the necessity of ensuring basic safety foundations—particularly in domestic SMEs, which often lack resources to implement regular training, enforce PPE compliance, or redesign unsafe workflows. While less influential than participation and monitoring, these factors remain essential in establishing baseline safety conditions.

The study also finds that safety attitude significantly predicts OSH performance but with only a modest effect (β = .070, p = .014). This aligns with the Theory of Planned Behavior, which emphasizes that attitudes serve as antecedents of intention and subsequent action (Ajzen, 1991; Milošević et al., 2025). Similarly, Jilcha and Kitaw (2016) argue that positive safety mindsets reinforce vigilance when supported by institutional structures. However, in Vietnam’s industrial parks, structural and systemic constraints—such as uneven enforcement, resource limitations, and widespread subcontracting—tend to overshadow individual dispositions. This helps explain why safety attitude, though statistically significant, exerts weaker influence compared to collective mechanisms like participation or systemic monitoring. This interpretation is consistent with evidence from healthcare, where Bakari et al. (2025) demonstrated that fear of Covid-19 significantly increased safety compliance among nurses. Their findings highlight how contextual or crisis-driven psychological factors can amplify adherence, suggesting that future studies in industrial parks should also consider the role of risk perceptions and health crises in shaping safety behaviors.

In contrast, safety communication (β = .191, p = .067) and risk control measures (β = .014, p = .726) were not statistically significant predictors. This is surprising given the Western literature often identifies communication and formal control systems as central to safety management (Markowski et al., 2021; Zhao et al., 2013; Zohar, 2002). The null effect in Vietnam may be explained by contextual factors. First, communication across firms is often fragmented: subcontractors, tenants, and foreign-invested enterprises may each use different languages, protocols, and reporting systems, making formal communication less effective. Second, risk control measures—such as administrative controls or documented procedures—are often poorly enforced in practice, especially among SMEs with limited capacity. Cagno et al. (2013) noted that fragmented safety responsibility in clustered environments undermines the effectiveness of formal risk controls. In Vietnam, such fragmentation is compounded by gaps in regulatory oversight, resulting in formal rules existing on paper but having limited real impact. While these contextual explanations are plausible, the non-significant findings should be interpreted with caution and require replication in other emerging economy contexts before drawing general conclusions about their broader applicability or lack thereof.

Taken together, these results confirm that participatory engagement and systematic monitoring are the most decisive levers for OSH performance in emerging-economy industrial parks, while traditional mechanisms like communication and formal risk controls may falter in fragmented contexts. The novelty of this study lies in its simultaneous ranking of eight determinants using SEM in a multi-firm setting—an approach rarely attempted in the literature. The unexpected null results for communication and risk control highlight a contextual divergence from Western assumptions and suggest that policymakers and industrial park administrators should rethink reliance on formal protocols. Instead, investments should focus on empowering participation, institutionalizing monitoring, and strengthening training to achieve meaningful improvements in worker safety.

Theoretical Implications

The findings deepen theoretical understanding of how multiple safety constructs jointly shape OSH performance in complex, multi-firm industrial settings. The strong predictive power of safety participation supports Social Exchange Theory, which argues that reciprocal commitments to safety emerge when organizations encourage worker engagement (Hofmann et al., 2017). The significance of safety monitoring and continuous improvement reinforces Organizational Learning Theory, showing that audits, feedback loops, and iterative evaluations are mechanisms through which firms adapt and enhance safety culture (Reniers et al., 2009). Evidence on training and protective equipment aligns with Human Capital Theory, suggesting that knowledge and safeguards represent investments that yield measurable returns in reduced accidents and improved compliance (Al-Bayati et al., 2023; Jilcha & Kitaw, 2016).

The positive effect of safe work design reflects principles of Sociotechnical Systems Theory, emphasizing that outcomes improve when work processes are ergonomically and structurally aligned with human capabilities (Markowski et al., 2021). The modest but significant influence of safety attitude supports the Theory of Planned Behavior, which views attitudes as antecedents of intentions and behaviors, but highlights that contextual and systemic factors can moderate their impact (Ajzen, 1991; Cagno et al., 2013). Conversely, the non-significance of communication and risk control measures suggests a need to revisit Institutional Theory: when formal procedures are weakly embedded or inconsistently enforced, their symbolic presence does not translate into effective practice (Zhao et al., 2013). Collectively, these insights advance integrated frameworks that combine behavioral, structural, and systemic factors, and they highlight the need for multilevel approaches to OSH management in emerging-economy industrial parks.

Practical Implications

From a practical standpoint, the study provides timely guidance for policymakers, safety managers, and industrial park administrators confronting rising accident rates in Vietnam’s industrial zones. First, the centrality of safety participation underscores the urgency of implementing mechanisms such as worker-led safety committees, digital hazard-reporting platforms, and recognition systems for proactive safety behaviors. These interventions are cost-effective and can be deployed rapidly across multi-firm zones to build shared accountability. Second, the strong effect of safety monitoring and continuous improvement highlights the importance of real-time dashboards, routine cross-firm audits, and digital incident-tracking tools. Given Vietnam’s rapid industrial expansion, such monitoring systems provide timely solutions to compensate for uneven external regulation.

Third, the significance of training and PPE calls for park-wide standardized induction programs, refresher courses, and strict enforcement of PPE usage—particularly urgent for SMEs with limited resources. Beyond individual determinants, the findings also speak to the role of integrated safety management systems. Navarro Claro et al. (2025) demonstrated that implementing OHSMS in the construction sector enhanced both human talent management and organizational performance. Our results similarly suggest that embedding systemic monitoring and training in industrial parks could generate not only safer workplaces but also organizational benefits, such as reduced turnover and enhanced productivity.

Fourth, the effect of safe work design emphasizes the need to integrate ergonomic standards into new factory layouts and to retrofit older facilities with hazard-reduction features, an especially pressing issue as many industrial clusters expand rapidly without adequate design oversight. Although safety attitude had only a modest effect, communication campaigns and leadership modeling can reinforce a safety-oriented mindset, especially when paired with tangible structural supports. In line with Kineber et al. (2023), who highlighted the benefits of implementing occupational health and safety management systems (OHSMS) for sustainable construction, this study underscores the value of systematic safety governance and extends the discussion by proposing policy-oriented interventions tailored to industrial parks and clusters in emerging economies.

Finally, the limited impact of communication and risk control measures indicates that traditional “paper-based” procedures are not sufficient in fragmented, subcontract-heavy contexts; administrators should redesign these systems to ensure cross-firm integration, enforceability, and accessibility to all workers. The broader economic implications of poor OSH practices must also be considered. Recent evidence from high-risk industries shows that accidents impose substantial economic costs (Cao et al., 2025). This perspective reinforces the urgency of prioritizing participation, monitoring, and training in Vietnam’s industrial parks, where systemic accidents not only endanger workers but also undermine competitiveness and impose hidden economic burdens. Overall, the findings support a systems-based strategy that prioritizes timely interventions—participation, monitoring, training, and design—while reforming weaker levers to achieve measurable safety improvements in Vietnam’s industrial parks.

Limitations and Future Research

This study provides valuable evidence on the determinants of OSH performance in Vietnamese industrial parks and clusters; however, several limitations must be recognized to contextualize the findings. First, the reliance on self-reported survey data introduces potential biases such as social desirability and recall error. Despite applying procedural and statistical remedies to mitigate Common Method Variance (Le, 2025; Le et al., 2025; Podsakoff et al., 2003), the single-source, cross-sectional design constrains causal inference and prevents assessment of how safety practices evolve over time.

Second, the focus on Vietnamese industrial zones limits external validity. Vietnam’s regulatory environment, labor market structure, and prevalence of subcontracting differ from other emerging economies. Consequently, the results may not transfer directly to contexts where enforcement regimes, cultural attitudes toward safety, or cluster governance structures are stronger or weaker (Guteta & Worku, 2023; Kongtip et al., 2008). While the sample size (n = 451) meets conventional thresholds for structural equation modeling (SEM), it was drawn using a non-probabilistic, convenience sampling approach from selected industrial zones. As such, generalizability beyond the sampled regions—let alone other emerging economies—should be approached with caution. Differences in institutional context and workforce composition may moderate the effects observed. Future studies should employ more representative, multi-country sampling to validate and extend these findings.

Third, the theoretical model, grounded in the TPB and Risk Governance Framework, excludes several variables that are increasingly relevant. Digital safety tools (e.g., real-time monitoring, AI-driven risk detection), contractor and informal-labor dynamics, and leadership styles may substantially influence OSH outcomes but were not incorporated. Similarly, this study did not test for moderating or mediating effects, such as the role of regulatory pressure in strengthening participation effects or leadership commitment in amplifying monitoring outcomes.

Fourth, the sampling frame excluded senior executives and informal workers, who play critical roles in shaping safety culture and compliance in multi-tier supply chains. Their omission narrows the scope of inference and may understate the systemic factors influencing OSH across the broader industrial ecosystem.

Building on these limitations, future research should pursue several avenues. Longitudinal and panel designs are needed to capture the dynamics of safety practices and establish causal ordering between determinants and outcomes. Incorporating multi-source data—such as supervisor evaluations, safety audits, and injury records—would reduce reliance on self-report and improve validity. Cross-country comparative studies could illuminate how regulatory regimes and cultural differences moderate the impact of behavioral and organizational factors. Expanding the theoretical framework to include digital safety innovations, leadership variables, and supply-chain actors (e.g., subcontractors, informal workers) would enhance explanatory power. Finally, investigating moderation and mediation mechanisms could clarify why certain levers (e.g., communication and risk control) fail in some contexts but succeed in others, providing nuanced guidance for policymakers and park administrators. Our findings highlight the importance of participation and monitoring as systemic levers for OSH performance. This resonates with emerging evidence that digital innovations can reinforce these mechanisms. For example, Zorzenon et al. (2025) found that IoT adoption in Brazilian companies significantly improved safety monitoring and overall OSH outcomes. This suggests that future research on industrial parks in Vietnam should explicitly examine how real-time digital tools—such as IoT sensors and predictive analytics—can amplify participation and continuous improvement.

Footnotes

Appendix

Table A6.

Constructs and Measurement Items of the OSH Performance Model.

Construct	Description	References
1. Safety attitude	I believe complying with safety procedures is essential, even under time pressure.	Dapari et al. (2025)
	I take personal responsibility for workplace safety.
	I believe most accidents can be prevented.
	I think about risks before starting any task.
	Safety is just as important as productivity.
2. Safety participation	I actively participate in safety meetings and training.	Griffin and Neal (2000) and Vinodkumar and Bhasi (2010)
	I report unsafe conditions or near-misses.
	I help coworkers follow safety procedures.
	I contribute ideas for improving safety.
3. Safety communication	Supervisors regularly discuss safety with workers.	Flin et al. (2006) and Seo et al. (2004)
	I can report safety concerns without fear.
	Safety procedures are communicated clearly.
	There are accessible channels for safety information.
4. Safety training	I receive adequate safety training for my job.	Kakemam et al. (2024) and Camatti et al. (2024)
	The training improves accident prevention knowledge.
	Safety training is held regularly.
	Training is tailored to job-specific risks.
	I feel confident after training sessions.
5. Safe work design	My job tasks are designed to reduce risk.	Kakemam et al. (2024) and Nemeth (2024)
	Workspaces are structured to support safety.
	Hazardous areas are marked and separated.
	Work design minimizes strain and fatigue.
	Tools and equipment are arranged to prevent accidents and ensure safe workflow.
6. Protective equipment	I always wear the required PPE at work.	Cassini et al. (2025) and Aguilar-Elena and Agún-González (2024)
	PPE is readily available for all tasks.
	I received training on correct PPE use.
	PPE is effective in preventing injuries.
7. Risk control measures	Hazards are regularly identified and evaluated.	Camatti et al. (2024) and (Li et al., 2024)
	Risk assessments are performed before tasks.
	Risk controls are updated periodically.
	Corrective actions are promptly implemented when safety risks are identified
	Emergency procedures are clearly defined.
8. Safety monitoring & continuous improvement	Safety performance is tracked regularly.	Thakur et al. (2025) and Vinodkumar and Bhasi (2010)
	Feedback from incidents is used to improve safety.
	Safety audits lead to actual improvements.
	Management encourages continuous safety improvements.
9. Occupational safety and health performances	Our organization has seen a decrease in safety incidents over the past year.	Sturm et al. (2019) and Palojoki et al., (2016)
	Employees consistently follow safety guidelines in daily tasks.
	We regularly review and update safety protocols based on new risks.
	Near-miss incidents are documented and used to improve practices.
	Management takes quick corrective actions after safety audits or reports.
	Overall, our safety measures effectively prevent workplace injuries.

ORCID iD

Thuy-Tien Le

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by Vietnam Maritime University.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

References

Abaa

D. M.

Ahiakpor

Yakubu Adama

A. S.

Frimpong

(2024). Implications of macroeconomic performance on foreign direct investment: A comparative study between emerging and frontier economies. African Journal of Empirical Research, 5(4), 1142–1160.

Abegaz

Deressa

Moen

B. E.

(2025). The relationship between safety climate and safety performance in the large-scale building construction industry in Ethiopia: A structural equation model using the NOSACQ-50 tool. Safety, 11(1), Article 28.

Aguilar-Elena

Agún-González

J. J.

(2024). Chi-square automatic interaction detection (CHAID) analysis of the use of safety goggles and face masks as personal protective equipment (PPE) to protect against occupational biohazards. Journal of Biosafety and Biosecurity, 6(2), 125–133. https://doi.org/10.1016/j.jobb.2024.05.001

Ajzen

(1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T

Ajzen

Madden

T. J.

(1986). Prediction of goal-directed behavior: Attitudes, intentions, and perceived behavioral control. Journal of Experimental Social Psychology, 22(5), 453–474. https://doi.org/10.1016/0022-1031(86)90045-4

Al-Bayati

A. J.

Rener

A. T.

Listello

M. P.

Mohamed

(2023). PPE non-compliance among construction workers: An assessment of contributing factors utilizing fuzzy theory. Journal of Safety Research, 85, 242–253. https://doi.org/10.1016/j.jsr.2023.02.008

Ali

M. X. M.

Arifin

Abas

Ahmad

M. A.

Khairil

Cyio

M. B.

Samad

M. A.

Lampe

Mahfudz

Ali

M. N.

(2022). Systematic literature review on indicators use in safety management practices among utility industries. International Journal of Environmental Research and Public Health, 19(10), Article 6198. https://doi.org/10.3390/ijerph19106198

Aven

(2014). Risk, surprises and black swans: Fundamental ideas and concepts in risk assessment and risk management (1st ed.). Routledge.

Aven

(2016). Risk assessment and risk management: Review of recent advances on their foundation. European Journal of Operational Research, 253(1), 1–13. https://doi.org/10.1016/j.ejor.2015.12.023

10.

Aven

(2019). The science of risk analysis: Foundation and practice (1st ed.). Routledge.

11.

Bakari

Bashir

Metwally

(2025). Covid-19 fear and safety compliance: A study of nursing professionals. SAGE Open, 15(1), Article 21582440251314669.

12.

Bisbey

Linhardt

R. M.

Woods Herron

Kilcullen

M. P.

Salas

(2026). How does training contribute to workplace safety? A meta-analysis examining the effects of safety training. Journal of Applied Psychology, 111(2), 175–194. https://doi.org/10.1037/apl0001309

13.

Burke

M. J.

Sarpy

S. A.

Smith-Crowe

Chan-Serafin

Salvador

Islam

(2006). Relative effectiveness of worker safety and health training methods. Journal of Applied Psychology, 91(5), 765–776. https://doi.org/10.1037/0021-9010.91.5.765

14.

Cagno

Micheli

G. J. L.

Jacinto

Masi

(2013). An interpretive model of occupational safety performance for small- and medium-sized enterprises. Safety Science, 60, 1–13. https://doi.org/10.1016/j.ssci.2013.07.002

15.

Camatti

Chirico

Pugliese

Giacomazzi

Borgia

Minicucci

A. M.

Pinelli

Colombo

Migliore

(2024). Workplace violence in Italian healthcare organizations: A multicenter survey on prevention measures and incident reporting. Giornale Italiano di Psicologia del Lavoro e delle Organizzazioni, 4(3), 179–190. https://doi.org/10.69088/2024/WRKP3

16.

Cao

Zhou

Miao

Wang

(2025). Exploring the economic occupational health, safety, and fatal accidents in high-risk industries. BMC Public Health, 25, Article 433.

17.

Cassini

Simniceanu

Gon

Cowling

B. J.

Allegranzi

, & COVID-19 in Health Workers Collaborative Group. (2025). Infection prevention and control risk factors for SARS-CoV-2 infection in health workers: A global, multi-centre, case-control study. Journal of Hospital Infection, 155, 40–50. https://doi.org/10.1016/j.jhin.2024.04.031

18.

Christian

M. S.

Bradley

J. C.

Wallace

J. C.

Burke

M. J.

(2009). Workplace safety: A meta-analysis of the roles of person and situation factors. Journal of Applied Psychology, 94(5), 1103–1127. https://doi.org/10.1037/a0016172

19.

Cornelissen

P. A.

Van Hoof

J. J.

De Jong

M. D. T.

(2017). Determinants of safety outcomes and performance: A systematic literature review of research in four high-risk industries. Journal of Safety Research, 62, 127–141. https://doi.org/10.1016/j.jsr.2017.06.005

20.

Dapari

Mahfot

M. H.

Zamzuri

M. A. I. A.

Ahmad

(2025). Development and validation of a questionnaire on knowledge, attitude and practice regarding occupational safety and health (Q-TRIMOSH-KAP) among food industry workers in Selangor, Malaysia. Malaysian Journal of Medicine and Health Sciences, 21(1), 167–175.

21.

Dyreborg

Lipscomb

H. J.

Nielsen

Törner

Rasmussen

Frydendall

K. B.

Bay

Gensby

Bengtsen

Guldenmund

Kines

(2022). Safety interventions for the prevention of accidents at work: A systematic review. Campbell Systematic Reviews, 18(2), Article e1234. https://doi.org/10.1002/cl2.1234

22.

Fernández-Muñiz

Montes-Peón

J. M.

Vázquez-Ordás

C. J.

(2009). Relation between occupational safety management and firm performance. Safety Science, 47(7), 980–991.

23.

Flin

Mearns

O’Connor

Bryden

(2000). Measuring safety climate: Identifying the common features. Safety Science, 34(1–3), 177–192. https://doi.org/10.1016/S0925-7535(00)00012-6

24.

Flin

Burns

Mearns

Yule

Robertson

E. M.

(2006). Measuring safety climate in health care. Quality and Safety in Health Care, 15(2), 109–115. https://doi.org/10.1136/qshc.2005.014761

25.

Garrigou

Laurent

Berthet

Colosio

Jas

Daubas-Letourneux

Jackson Filho

J.-M.

Jouzel

J.-N.

Samuel

Baldi

Lebailly

Galey

Goutille

Judon

(2020). Critical review of the role of PPE in the prevention of risks related to agricultural pesticide use. Safety Science, 123, Article 104527. https://doi.org/10.1016/j.ssci.2019.104527

26.

Garschagen

Revilla Diez

Nhan

D. K.

Kraas

(2012). Socio-economic development in the Mekong Delta: Between the prospects for progress and the realms of reality. In Renaud

F. G.

Künzer

(Eds.), The Mekong Delta system: Interdisciplinary analyses of a river delta (pp. 83–132). Springer. https://doi.org/10.1007/978-94-007-3962-8_4

27.

Griffin

M. A.

Neal

(2000). Perceptions of safety at work: A framework for linking safety climate to safety performance, knowledge, and motivation. Journal of Occupational Health Psychology, 5(3), 347–358. https://doi.org/10.1037/1076-8998.5.3.347

28.

Guteta

Worku

(2023). Sustainable industrial parks development in Ethiopia: Analysis of policy framework gaps. Sustainable Production and Consumption, 36, 228–241. https://doi.org/10.1016/j.spc.2023.03.004

29.

Hair

J. F.

Babin

B. J.

Anderson

R. E.

Black

W. C.

(2019). Multivariate data analysis (8th ed.). Pearson Prentice.

30.

Hair

J. F.

Hult

G. T. M.

Ringle

C. M.

Sarstedt

(2016). A primer on partial least squares structural equation modeling (PLS-SEM) (2nd ed.). Sage Publications.

31.

Hale

A. R.

Hovden

(1998). Management and culture: The third age of safety. A review of approaches to organizational aspects of safety, health and environment. In Hale

A. R.

Hovden

(Eds.), Occupational injury: Risk, prevention and intervention (pp. 129–165). CRC Press.

32.

Hecker

Goldenhar

(2014, January). Understanding safety culture and safety climate in construction: Existing evidence and a path forward (Literature review summary for Safety Culture/Climate Workshop, June 11–12, 2013, Washington, DC). CPWR – The Center for Construction Research and Training.

33.

Henseler

Ringle

C. M.

Sarstedt

(2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science, 43(1), 115–135.

34.

Hoe

V. C. W.

Urquhart

D. M.

Kelsall

H. L.

Zamri

E. N.

Sim

M. R.

(2018). Ergonomic interventions for preventing work-related musculoskeletal disorders of the upper limb and neck among office workers. Cochrane Database of Systematic Reviews, 2018(10), Article CD008570.

35.

Hofmann

D. A.

Burke

M. J.

Zohar

(2017). 100 years of occupational safety research: From basic protections and work analysis to a multilevel view of workplace safety and risk. Journal of Applied Psychology, 102(3), 375–388. https://doi.org/10.1037/apl0000114

36.

Hou

Khokhar

Haider

(2021). Put safety first: Exploring the role of health and safety practices in improving the performance of SMEs. SAGE Open, 11(3), Article 21582440211032173.

37.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

38.

Hutchinson

Luria

Pindek

Spector

(2022). The effects of industry risk level on safety training outcomes: A meta-analysis of intervention studies. Safety Science, 152, Article 105594. https://doi.org/10.1016/j.ssci.2021.105594

39.

Ismail

Doostdar

Harun

(2012). Factors influencing the implementation of a safety management system for construction sites. Safety Science, 50(3), 418–423.

40.

ISO. (2018). ISO 31000:2018 risk management – guidelines.

41.

Jääskeläinen

Tappura

Pirhonen

(2022). The path toward successful safety performance measurement. Journal of Safety Research, 83, 181–194.

42.

Jilcha

Kitaw

(2016). A literature review on global occupational safety and health practice & accidents severity. International Journal for Quality Research, 10(2), 279–310. https://doi.org/10.18421/IJQR10.02-04

43.

Kakemam

Miri

Sadeghpour

Mirzaei

(2024). Attitudes towards patient safety among physicians and nurses in Iranian governmental teaching hospitals: A cross-sectional survey. BMJ Open, 14(11), Article e089328. https://bmjopen.bmj.com/content/14/11/e089328.full.pdf

44.

Kineber

A. F.

Antwi-Afari

M. F.

Elghaish

Zamil

A. M. A.

Alhusban

Qaralleh

T. J. O.

(2023). Benefits of implementing occupational health and safety management systems for the sustainable construction industry: A systematic literature review. Sustainability, 15(17), Article 12697.

45.

Kline

R. B.

(2016). Principles and practice of structural equation modeling (4th ed.). Guilford Press.

46.

Kongtip

Yoosook

Chantanakul

(2008). Occupational health and safety management in small and medium-sized enterprises: An overview of the situation in Thailand. Safety Science, 46(9), 1356–1368.

47.

Kumar

R. R.

Stauvermann

P. J.

Dau

L. T. M.

(2025). Exploring the tourism and economic growth relationship in Vietnam: A cointegration analysis with model-specific structural breaks. Economies, 13(2), Article 29. https://www.mdpi.com/2227-7099/13/2/29

48.

S.-T.

(2024). Applying the quantitative approach to identify the major antecedents to the process of cold ironing adoption. HomeWorld Review of Intermodal Transportation Research, 12(1), 1–24.

49.

S.-T.

(2025). From personality to connection: The mediating role of tie strength in HEXACO traits and job networking behaviour. International Journal of Networking and Virtual Organisations, 33(1), 39–66.

50.

S.-T.

Nguyen

T. N.

Dang

C. X.

(2025). Why do some join and others don’t? A behavioral exploration of voluntary social insurance in developing countries. SAGE Open, 15(4), Article 21582440251383689.

51.

H.-L.

Zhao

Liu

Y.-L.

Miao

Y.-L.

Liu

J.-M.

Sun

W.-W.

Wang

Cao

W.-M.

(2024). Construction of competency evaluation system for infection control personnel in traditional Chinese medicine hospitals. Infectious Diseases Research, 5(4), Article 20. https://doi.org/10.53388/IDR2024020

52.

Lingard

Cooke

Blismas

(2013). Safety in design and its contribution to construction safety. Engineering, Construction and Architectural Management, 20(5), 569–580. https://doi.org/10.1108/ECAM-07-2012-0070

53.

Liu

Bao

Yang

Zhang

(2024). Research on the decision-making of work safety investment in industrial park enterprises: Evidence from behavioral experiments. Frontiers in Public Health, 12, Article 1295536.

54.

Lund

Aarø

L. E.

(2004). Accident prevention: Presentation of a model placing emphasis on human, structural and cultural factors. Safety Science, 42(4), 271–324. https://doi.org/10.1016/S0925-7535(03)00045-6

55.

Mamade

G. F.

(2022). Do cluster based industrial parks catalyse developmental spillovers? Evidences from Ethiopia’s experience [Doctoral dissertation, University of Milan]. https://tesidottorato.depositolegale.it/handle/20.500.14242/113972

56.

Markowski

A. S.

Krasławski

Vairo

Fabiano

(2021). Process safety management quality in industrial corporation for sustainable development. Sustainability, 13(16), Article 9001. https://doi.org/10.3390/su13169001

57.

Mearns

Whitaker

S. M.

Flin

(2003). Safety climate, safety management practice and safety performance in offshore environments. Safety Science, 41(8), 641–680. https://doi.org/10.1016/S0925-7535(02)00011-5

58.

Milošević

Stojanović

Nikolić

Đ.

Mihajlović

Brkić

Perišić

Spasojević-Brkić

(2025). Occupational health and safety performance in a changing mining environment: Identification of critical factors. Safety Science, 184, Article 106745.

59.

Morrow

S. L.

McGonagle

A. K.

Dove-Steinkamp

M. L.

Walker

C. T.

Jr. Marmet

Barnes-Farrell

J. L.

(2010). Relationships between psychological safety climate facets and safety behavior in the rail industry: A dominance analysis. Accident Analysis & Prevention, 42(5), 1460–1467.

60.

Nahrgang

J. D.

Morgeson

F. P.

Hofmann

D. A.

(2011). Safety at work: A meta-analytic investigation of the link between job demands, job resources, burnout, engagement, and safety outcomes. Journal of Applied Psychology, 96(1), 71–94. https://doi.org/10.1037/a0021484

61.

Nam

P. K.

Heshmati

(2024). Introduction, summary, and policy recommendations. In Nam

P. K.

Heshmati

(Eds.), Recent trends in Vietnam’s rapid economic development (pp. 1–15). Springer. https://doi.org/10.1007/978-3-031-27596-8_1

62.

National Institute for Occupational Safety and Health. (2024, April 10). About hierarchy of controls for everyone. Centers for Disease Control and Prevention. https://www.cdc.gov/niosh/hierarchy-of-controls/about/index.html

63.

Navarro Claro

G. T.

Bayona Soto

J. A.

Arévalo Ascanio

J. G.

(2025). Impact of the occupational health and safety management system (OHSMS) on human talent management and organizational performance in the construction sector. Frontiers in Built Environment, 11, Article 1618356.

64.

Neal

Griffin

M. A.

Hart

P. M.

(2000). The impact of organizational climate on safety climate and individual behavior. Safety Science, 34(1–3), 99–109. https://doi.org/10.1016/S0925-7535(00)00008-4

65.

Nemeth

(2024). CHATS: A tool to identify troubling symptoms from brain injury to assure accommodations and referrals to safety and health services for domestic violence survivors. Archives of Physical Medicine and Rehabilitation, 105(4), Article e120. https://doi.org/10.1016/j.apmr.2024.02.338

66.

Nguyen

P. T.

Nguyen

T. P. L.

(2024). Assessing the impact of foreign direct investment on Vietnam’s socioeconomic development and proposing policy recommendations. VNU Journal of Science: Policy and Management Studies, 40(3), 40–53.

67.

Occupational Safety and Health Administration. (2023). Identifying hazard control options: The hierarchy of controls. U.S. Department of Labor. https://www.osha.gov/safety-management

68.

Palojoki

Pajunen

Saranto

Lehtonen

(2016). Electronic health record-related safety concerns: A cross-sectional survey of electronic health record users. JMIR Medical Informatics, 4(2), Article e13. https://doi.org/10.2196/medinform.5358

69.

Parker

S. K.

(2014). Beyond motivation: Job and work design for development, health, ambidexterity, and more. Annual Review of Psychology, 65, 661–691.

70.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

J. Y.

Podsakoff

N. P.

(2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903. https://doi.org/10.1037/0021-9010.88.5.879

71.

Reason

(2000). Human error: Models and management. BMJ, 320(7237), 768–770. https://doi.org/10.1136/bmj.320.7237.768

72.

Reniers

G. L. L.

Ale

B. J. M.

Dullaert

Soudan

(2009). Designing continuous safety improvement within chemical industrial areas. Safety Science, 47(5), 578–590.

73.

Santos

Rojas

Isidoro

Lorente

Dias

Mariscal

Benlloch

Lorente

(2025). Efficacy of ergonomic interventions on work-related musculoskeletal pain: A systematic review and meta-analysis. Journal of Clinical Medicine, 14(9), Article 3034.

74.

Sehsah

El-Gilany

A.-H.

Ibrahim

A. M.

(2020). Personal protective equipment (PPE) use and its relation to accidents among construction workers. La Medicina del Lavoro, 111(4), 285–295. https://doi.org/10.23749/mdl.v111i4.9398

75.

Seo

D. C.

Torabi

M. R.

Blair

E. H.

Ellis

N. T.

(2004). A cross-validation of safety climate scale using confirmatory factor analytic approach. Journal of Safety Research, 35(4), 427–445. https://doi.org/10.1016/j.jsr.2004.04.006

76.

Sinelnikov

Inouye

Kerper

(2015). Using leading indicators to measure occupational health and safety performance. Safety Science, 72, 240–248. https://doi.org/10.1016/j.ssci.2014.09.010

77.

Sturm

Rieger

M. A.

Martus

Ueding

Wagner

Holderried

Maschmann

(2019). Working conditions and patient safety culture correlate with objective workload and patient outcomes: A cross-sectional explorative study from a German university hospital. PLOS ONE, 14(1), Article e0209487. https://doi.org/10.1371/journal.pone.0209487

78.

Thakur

Saran

Chen

(2025). Editorial: The impact of specific environmental exposures on breast, lung, and colon cancer: Advancing public health strategies for enhanced outcomes. Frontiers in Public Health, 13, Article 1483915.

79.

Tucker

Turner

(2011). Young worker safety behaviors: Development and validation of a measure of workplace safety behaviors among young workers. Journal of Safety Research, 42(2), 127–133. https://doi.org/10.1016/j.jsr.2011.01.004

80.

Vietnam Social Security. (2024, May 8). Occupational safety, health and well-being are improved towards better life for labors. Vietnam Social Security News. https://vss.gov.vn/english/news/Pages/vietnam-social-security.aspx?CateID=0&ItemID=11841

81.

Vinodkumar

M. N.

Bhasi

(2010). Safety management practices and safety behavior: Assessing the mediating role of safety knowledge and motivation. Accident Analysis & Prevention, 42(6), 2082–2093. https://doi.org/10.1016/j.aap.2010.06.021

82.

Vitrano

Micheli

G. J. L.

Guglielmi

De Merich

Pellicci

Urso

Ipsen

(2023). Sustainable occupational safety and health interventions: A study on the factors for an effective design. Safety Science, 166, Article 106249.

83.

Won

Choi

Kim

(2024). Development of sustainable industrial safety and health policy for electronic industry in Korea: A study on the relationship between safety leadership of managers, safety climate, and safety behavior. Sustainability, 16(23), Article 10308.

84.

Xuan

V. N.

(2025). The nexus between innovation, renewable energy consumption, foreign direct investment, GDP, and CO₂ emissions in the context of Vietnam: New evidence from the ARDL approach. Energy & Environment. Advance online publication. https://doi.org/10.1177/0958305X251322893

85.

Zeng

D. Z.

(2016). Special economic zones: Lessons from the global experience (PEDL Synthesis Paper Series). https://assets.publishing.service.gov.uk/media/586f9727e5274a130700012d/PEDL_Synthesis_Paper_Piece_No_1.pdf

86.

Zeng

Chen

Yang

Reniers

Zhao

Liu

(2020). A systematic literature review on safety research related to chemical industrial parks. Sustainability, 12(14), Article 5753.

87.

Zhao

Joas

Abel

Marques

Suikkanen

(2013). Process safety challenges for SMEs in China. Journal of Loss Prevention in the Process Industries, 26(5), 880–886.

88.

Zohar

(2002). The effects of leadership dimensions, safety climate, and assigned priorities on minor injuries in work groups. Journal of Organizational Behavior, 23(1), 75–92.

89.

Zorzenon

Lizarelli

F. L.

Braatz

(2025). The impact of the internet of things on health and safety performance at work: An empirical study of Brazilian companies. Safety Science, 187, Article 106866.