How big data analytics improves hospital environmental performance through supply chain innovation, decision quality, and risk taking

Abstract

Despite big data analytics (BDA) capabilities have been increasingly recognized for their potential to improve sustainability, the underlying mechanisms by which BDA capabilities influence hospital environmental performance in the context of healthcare supply chains are not well understood. This paper aims to bridge this significant empirical void by examining the mediating effect of supply chain innovation, decision-making quality and risk-taking on the links between BDA capabilities and environmental performance for Chinese hospitals. Based on Stimulus-Organism-Response theory, the theoretical model depicts BDA capability as a key stimulus factor affecting the hospital sustainability outcomes. This research employed a quantitative research method, and a structured survey instrument was administered to 653 healthcare providers from various hospitals. The participants were recruited using a random sampling method to achieve broad representation. Variables in the survey include measures of big data analytics capability, supply chain innovation, quality of decision-making, risk-taking, and environmental performance. AMOS was used for Structural Equation Modeling (SEM) analysis to test the proposed relationships and mediation effects among the variables in due diligence. Empirical results support a positive relationship between hospitals’ BDA capability and environmental performance, indicating that it is statistically significant. Crucially, this link is to some degree mediated by supply chain innovation, quality of decision-making and risk taking behaviour. In particular, hospitals with high analytical capability were more innovative in their supply chain production, had better decision-making structures, and showed a tendency to be more risk takers, leading to good environmental outcomes. This research limns the manner in which improving BDA capabilities can systematically contribute to hospital sustainability in innovative, informed, and strategically bold supply chain management practices; thereby new theoretical and practical aspects are provided. These results not only enrich current theoretical constructs but also give insights to healthcare managers and policy makers for harnessing the big data analytics to promote environmental sustainability and implement a real change in the healthcare performance.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-16541-0.

Introduction

In real-world applications, big data frequently fails to generate valuable insights due to inherent challenges associated with handling and interpreting large, diverse, and complex datasets¹. The complexity stems from dimensions including volume (too much data for traditional data management systems to store and process), variety (of data structures and sources; from structured, semi-structured, and unstructured sources; making integration and consistency a challenge), and velocity (speed that data accumulates making real-time or near-real-time processing capacity a significant deficit for many organizations)². Moreover, make the truthfulness and reliability of large data sets is proved to be challenging because of the uncertainty of noise, errors, inconsistency and their absent in data gathering and data integration process. The complexity of big data also encompasses issues related to privacy, security, and compliance with regulatory requirements, further constraining the ability of organizations to leverage big data effectively³. These multifaceted challenges collectively undermine the potential for big data analytics to consistently deliver actionable insights, unless organizations develop advanced capabilities specifically tailored to manage, process, and analyse such complex data effectively. Overcoming these obstacles necessitates robust big data analytics capabilities, which encompass more than merely adapting existing analytical algorithms; significant resources and expertise are required⁴.

However, big data analytics capabilities play a crucial role across multiple dimensions of organizational performance, particularly in enhancing environmental performance⁵. These capabilities enable organizations to effectively process and interpret vast datasets, thus uncovering critical insights that drive informed decision-making. Specifically, in the environmental dimension, big data analytics facilitates improved resource allocation, waste reduction, and the adoption of sustainable practices by providing accurate and timely information on consumption patterns, emissions, and environmental impacts^6,7. Consequently, organizations equipped with robust big data analytics capabilities are better positioned to achieve sustainability goals, optimize their environmental performance, and maintain competitiveness in rapidly evolving market conditions⁸.

In the context of hospitals, environmental performance refers to interorganizational processes designed to simultaneously achieve improved outcomes across environmental, social, and economic dimensions, thereby promoting holistic sustainability⁹. It can be achieved through things like proper waste practices, smarter use of resources, less emissions and adopting sustainability into everything hospitals and their supply chains do. The environmental dimension of sustainability performance remains a significant research gap, demanding further comprehensive and critical investigation¹⁰. Increasingly interconnected environmental concerns produce substantial volumes of data, often referred to as big data, which require advanced analytical capabilities for meaningful interpretation¹¹. Acquiring big data analytics abilities allows hospitals to transform numerous data into useful information to improve decision-makings and environmental performance¹². Furthermore, integrating digital technologies to hospital operational processes can enable real-time access to data and puts in place structured communication mechanisms that further support environmental sustainability practices¹³.

Research gap

An integrated review of existing literature indicates that various theoretical perspectives have been utilized to investigate big data analytics capability, such as resource-based view¹⁴, open innovation theory¹⁵, and stakeholder theory¹⁶. Moreover, existing literature has explored the relationships between big data analytics and environmental performance^12,17, as well as the connection between big data analytics and various aspects of supply chain management within the context of hospitals or healthcare institutes^18,19. Additionally, previous research has generally examined the impact of big data analytics on organizational risk-taking behaviours^20–22 and decision-making quality^23–25, primarily in broader organizational contexts, such as companies or firms. However, studies specifically investigating these relationships within hospital settings remain scarce, highlighting a notable gap and underscoring the need for further targeted research in this area. Moreover, an absence of studies on the mediating effects of supply chain innovation, decision making quality, risk taking in the relationship between big data analytics capability and environmental performance in either hospital/ healthcare organization studies or other in general. The current study aims to fill this important gap in the literature.

Objectives of the study

The main objective of this study is to fill knowledge gaps revealed in the literature through developing a new framework to use big data analytics for improving hospital environmental performance and stimulating supply chain innovation. This paradigm has the capability to incorporate sophisticated analytical tools into strategic management of hospitals and environmental policy planning, and to unlock the substantial though not fully utilised power of big data to the healthcare practice. In particular, the model is designed to create an integrated approach to the improvement of decision-making quality and risk management in the hospital supply chain in order to improve environmental performance quite significantly. Furthermore, this paper makes a theoretical contribution of sourcing chain innovation, quality of decision-making, and risk-taking compared to the traditional. The framework is developed to systematically examine large datasets to identify inefficiencies, predict future trends, and recommend sustainable practices. Additionally, the study seeks to evaluate the necessity for conducting assessments of healthcare outcomes resulting from environmental and supply chain activities. By bridging these critical gaps, the proposed framework offers innovative theoretical insights and practical guidance to healthcare organizations committed to achieving operational excellence while adhering to sustainability principles.

Novelty of the study

The novelty of this study lies in its comprehensive exploration of the relationship between big data analytics capability and environmental performance within the context of hospitals, particularly by examining previously underexplored mediating factors: supply chain innovation, decision-making quality, and risk-taking behaviours. Unlike prior studies that typically focused on big data analytics in broader organizational contexts, this research uniquely integrates these specific mediators within the hospital sector, thereby extending existing theoretical frameworks and offering practical insights tailored to healthcare sustainability. Moreover, by explicitly utilizing the Stimulus-Organism-Response (S-O-R) theory, the study provides a fresh, structured perspective on how big data analytics capabilities trigger internal organizational processes, ultimately leading to improved environmental outcomes in healthcare settings.

Research questions

1. How does big data analytics capability influence the environmental performance of hospitals?

2. What roles do supply chain innovation, decision-making quality, and risk-taking play in mediating the relationship between big data analytics capability and hospital environmental performance?

Research hypotheses development

This section presents the theoretical basis and empirical justification for the proposed hypotheses of the study. For clarity and coherence, in our study we classify the hypotheses into clear blocks defined by the nature of the relationships among them: the first group addresses the direct effect of big data analytics on environmental performance; the second of big data analytics on the improvement of internal organizational capabilities; the third on the impact of these capabilities in terms of an improvement in environmental outcomes; and the fourth group on the mediating roles that these capabilities play. This structured approach serves as a guideline for a systematic inquiry into the extent to which big data analytics helps sustain hospital operations.

Group 1: big data analytics capability and hospital environmental performance

Applying big data to hospital operations It is a major step in advanced environmental performance in healthcare. Big data analytics poses to be a highly promising tool to hospital managements, as it enables them to analyse massive and complex data sets from various sources (such as electronic health records, energy consumption monitoring systems, and waste management records), and gain actionable insights that exceed the potential of conventional data analysis methods²⁶. In a similar vein, predictive analytics can help to detect high points in energy usage, allowing hospitals to adapt and minimise the level of energy that is utilised²⁷. Likewise, analysing waste management data can reveal specific areas where recycling or composting efforts can be intensified, leading to meaningful reductions in landfill waste. These data-driven insights empower hospitals to adopt targeted strategies for conserving energy, minimizing waste, and optimizing resource utilization, collectively contributing to a smaller ecological footprint. Furthermore, the ability to process and interpret extensive datasets in real time supports continuous improvement and the advancement of environmental sustainability practices²⁸. Taken together, these capabilities underscore the central role of big data analytics in strengthening hospitals’ commitment to sustainable operations. Based on these insights from the literature, the present study proposes the following hypothesis.

H1

Big data analytics capability is positively related to hospital environmental performance.

Group 2: effect of big data analytics capability on supply chain innovation, decision making quality, and taking risk

Big data analytics capability exerts a substantial influence on supply chain innovation, decision-making quality, and risk-taking within the healthcare industry.

BDAC and SCI

A substantial body of literature has highlighted the transformative role of BDA in driving innovation in supply chains across various industries. In concept, BDA has been acknowledged as offering the capability for organisations to shift towards proactive, integrated, and data-driven operations, as opposed to reactive and fragmented ones in the supply chain^29,30. Using predictive and prescriptive analytics, firms can forecast market requirements and inventory levels for optimal performance, as well as receive real-time alerts to bottlenecks in the supply chain, enabling them to develop new and creative solutions to age-old supply chain problems. These conceptual arguments are supported by empirical research. Rashid, Baloch⁷ also found that firms that invest in significant data capabilities are far more innovative in areas such as supply chain redesign, including just-in-time delivery, green logistics, and digital supply chain integration. Similarly, Kumar and Raj³¹ found evidence that developed BDA capabilities drove firms to adopt greener sourcing and circular economy initiatives, thus directly associating analytics capability with innovative supply chain outcomes. This can also have particularly big potential in hospitals, where this sort of capability can mean that healthcare organizations can get more effective with procurement and reducing wastage of medical products (and move much faster when it comes to responding to changing patient care loads). Based on this evidence, we propose our hypothesis: H2a with a high level of big data analytics capability are associated with a higher level of SCI, leading to gains in both operational and environmental performance.

H2a

Big data analytics capability is connected in a positive way to supply chain innovation.

BDAC and DMQ

Organisations traditionally employ big data analytics capabilities as a key influencer in quality decision-making²⁴. BDAC is acknowledged as a key component of quality decision-making, with both conceptual perspectives and empirical evidence. Theoretical frameworks suggest that BDA enables managers to access factual and timely information, thereby minimizing uncertainty and informing decisions based on evidence^32,33. Through the application of pattern recognition and predictive modelling, BDA tools enhance leaders’ comprehension of complex systems and prospective trends, thereby solidly underpinning their strategic decisions. Empirical evidence supports these theoretical observations. Kampoowale³⁴ observed that organizations using BDA on average registered statistically significant increases in decision accuracy, responsiveness, and agility in supply chain operations. Sumrit³⁵ also demonstrate that big data capabilities enhance managers’ confidence and decision-making speed, thereby indirectly leading to a competitive advantage. Accuracy of decisions in a hospital context is crucial not only for effectiveness and patient safety, but also for environmental impact. Hospitals that have BDA capabilities can leverage various sources of data to inform resource allocation, improve treatment planning, and optimize sustainability strategies³⁶. Accordingly, we posit that big data analytics capability has a positive impact on decision-making quality, enabling hospitals to make suitable decisions for both long-term and short-term success.

H2b

Big data analytics capability is connected in a positive way to decision making-quality.

BDAC and risk taking

A level of risk-taking is also influenced by an organization’s capacity to analyse big data, a defining aspect for fuelling innovation and strategic disruption²⁰. In parts, this is also because the availability of broad and fresh data volumes helps reduce perceived uncertainty, encouraging some measure of higher risk aversion that organizations would otherwise shy away from^21,37. Beer believes that scenario analysis, forecasting, and simulating potential outcomes contribute to an organisation’s ability to plan and respond to risk more effectively. Concrete support for this point of view comes from an empirical study: Basile, Carbonara³⁸ also observed that healthcare services with mature BDA competencies were encouraged to engage in trial projects and adopt recent technologies and procedures in the presence of high levels of uncertainty. Similarly, Alrfai, Maabreh³⁹ demonstrated that companies with a stronger BDA were more likely to undertake sustainability-based innovations, although such radical innovations may involve financial and operational hazards. In the hospital environment, where higher levels of risk aversion are based on patient safety issues and regulatory requirements, BDA may provide decision-makers with the confidence to pilot new sustainability practices or invest in eco-friendly equipment. Such calculated risks can have significant, long-term payoffs in operational and environmental performance. Therefore, we propose:

H2c

Big data analytics capability is connected in a positive way to risk-taking.

Group 3: linking supply chain innovation, decision making quality, and taking risk to hospital environmental performance

SCI and environmental performance

SCI is a powerful lever for enhancing environmental performance in healthcare organizations. By integrating sustainable practices and advanced technologies into supply chain operations, hospitals can effectively minimize waste, reduce emissions, and optimize resource use. These practices, which include green logistics, sustainable sourcing and circular economy models lead to improved operational performance and a lower environmental footprint of healthcare supply chains⁴⁰. This includes strategies for transportation and energy efficiency, which could even decarb the region. In addition, using digital technologies and big data analysis, hospitals can have valuable insight into the ecological effects of their supply chain, that help them to decide informed actions toward sustainability⁴¹. These technologies serve a double advantage: on one hand they help preserving the environment, and on the other, they help in response to the increasing social and regulatory pressure for sustainable health services, improving the institutions’ image while being competitive.

H3a

Supply chain innovation is connected in a positive way to hospital environmental performance.

DMQ and environmental performance

To enhance decision-making regarding environmental performance, sustainability issues must be integrated into decision-making in a structured manner. It means considering the environmental impact of how we operate, while also doing so with an eye to the bottom line and the schedule. Now, making informed decisions based on data and analytics for assessing and predicting the ecological impacts of alternative courses of action enables healthcare leaders to choose strategies that align with sustainable goals. In addition, a strict and transparent decision-making process may enhance accountability and foster an organization-oriented culture^42,43.

H3b

Decision making-quality is connected in a positive way to hospital environmental performance.

Risk-taking and environmental performance

Additionally, accepting managed risks in the pursuit of environmental innovation can yield significant gains in sustainability. Hospitals that are willing to innovate by investing in new technologies and pursuing non-traditional operational methods can find more efficient ways to reduce the size of their ecological footprint⁴⁴. Through their use of data-driven analysis and risk assessment processes, organisations such as these can justify bold steps towards innovative solutions that provide significant environmental benefits^45,46. Taken together, supply chain innovations, best-practice decision-making, and informed risk-taking present a comprehensive and strategic roadmap for hospitals to reduce their ecological footprint and make a real contribution to the world’s environmental well-being.

H3c

Risk-taking is connected in a positive way to hospital environmental performance.

Group 4: mediating roles of supply chain innovation, decision making quality, and taking risk in the relationship between big data analytics capability and environmental performance

Hospitals can gather detailed supply chain information, from sourcing to consumption, by optimally utilizing big data analytics. These lessons also enable hospitals to identify and address inefficiencies, waste sources, and opportunities for reducing emissions, thereby supporting a more sustainable adoption of practices and technologies⁴⁷. For example, with the help of advanced data analytics, demand for new supplies can be predicted more accurately, resulting in less overproduction and waste of medical resources⁴⁸. By utilizing these data-driven insights, hospitals have the potential to significantly enhance their supply chain operations significantly, creating more efficient processes, achieving cost savings, and reducing their overall environmental impact⁴⁹. This indicates that supply chain innovation addressed through big data analytics is not only instrumental in driving operational excellence and sustainability but also in improving hospitals’ competitiveness in the marketplace and meeting environmental performance targets.

Moreover, big data analytics also enables hospitals to gain various priceless insights that help enhance operational efficiency and achieve greener outcomes. In addition to improving decision-making regarding the sustainability implications of different options, prioritize projects that aim to minimize environmental impacts, and contrast organizational strategies with environmental targets, etc⁵⁰. Analytics can uncover the granular aspects of each hospital process’s carbon footprint, enabling managers to make data-informed decisions that reduce emissions. Environmental considerations should be integrated into decision-making processes to ensure sustainability becomes a key consideration in hospital planning, leading to more responsible resource usage, reduced waste, and improved environmental care. The incorporation of big data analytics into these processes improves the quality of decision-making, highlighting the importance of critical thinking and strategic planning in reapplying data towards actionable sustainability outcomes⁵¹.

In addition, big data analytics can help hospitals focus on targeted areas for sustainability enhancement, including innovative techniques and technologies that could be utilized⁵². Yet, introducing such solutions is rarely straightforward, and there can be uncertainty, upfront investment, or operational disturbances. This is where hospitals will need to take risks, turning data-driven insights into action for sustainable practices in times of uncertainty. Taking well-measured strides, hospitals can implement new technologies, try out sustainable materials, and implement high-impact operating process change initiatives, all of which can make a large-scale impact on the environment. The appetite to take smart risks, informed by data analysis, can significantly accelerate progress in reducing emissions, minimizing waste, and optimizing resources.

Accordingly, big data analytics is essential not just for identifying opportunities but also for converting potential environmental performance improvements into innovative strategies and tangible results, which rely on the willingness to take calculated risks.

As the preceding discussion highlights, supply chain innovation, high-quality decision-making, and thoughtful risk-taking play essential roles in enhancing environmental performance in hospitals, and these elements operate synergistically with the power of big data analytics. Based on these insights, the following hypotheses are proposed:

H4a

Supply chain innovation positively mediates the relationship between big data analytics capability and hospital environmental performance.

H4b

Decision making quality positively mediates the relationship between big data analytics capability and hospital environmental performance.

H4c

Risk Taking positively mediates the relationship between big data analytics capability and hospital environmental performance.

Figure 1 illustrates the proposed model of this study.

Fig. 1.

Research proposed model.

Methods

Stimulus-organism-response theory

The Stimulus-Organism-Response Theory⁵³, also known as S-O-R Theory, is a conceptual framework that originates from the field of psychology. It describes how individuals react to external stimuli through internal processes, which ultimately result in predetermined responses. In this study, the S-O-R model is innovatively adapted to assess hospital environmental performance through a multidisciplinary perspective linking psychological and organizational theories to healthcare sustainability. Specifically, we clarify that external stimuli include regulatory pressures, societal expectations, and technological innovations influencing hospitals to develop internal capabilities such as big data analytics. Thus, big data analytics capability is conceptualized as an internal organizational capability (Organism), developed in response to external stimuli, and subsequently influencing internal processes such as supply chain innovation, decision-making quality, and risk-taking. Ultimately, these internal processes shape observable outcomes (Response), specifically the environmental performance of hospitals. By clearly positioning external pressures as stimuli and big data analytics capability as an internal organizational capacity developed in response to these pressures, we provide a more coherent and theoretically consistent application of the S-O-R framework.

Measures

Within the scope of the present investigation, measuring scales were utilized that have previously been subjected to thorough testing and validation by other researchers. To evaluate the participants’ responses, a five-point Likert scale was used. The scale ranged from 1, indicating severe disagreement, to 5, indicating strong agreement. The theoretical literature and research topics related to latent variables are presented in Table 1 and Supplementary Appendix A, which provides a comprehensive overview of their contents.

Table 1.

Theoretical support for measurement variables.

Latent variable	Quantity of inquiries	Theoretical support
Big data analytics capability	10 items	Wamba, Dubey54
Decision making quality	3 items	Shamim, Zeng55
Supply chain innovation	14 items	Paulraj56
Risk taking	8 items	Covin and Slevin57
Environmental performance	7 items	Kim, Kim58

Pilot study

The researchers delivered one hundred questionnaires to individuals who were affiliated with hospitals in the Zhejiang province in China. This was done as a preliminary effort to investigate a particular feature of hospital operations across Zhejiang province. As a result of this distribution, they obtained 86 valid responses, which represents an 86% response rate. This represents an exceptionally high level of involvement for studies within this category. It was discovered through the preliminary analysis of the pilot survey that the constructs that were being measured exhibited satisfactory reliability and validity. This finding indicates that the design of the questionnaire was sound and that the contents of the questionnaire could potentially be relevant. An analysis of power was carried out with the help of the G*Power software to further strengthen the foundation of the study. The findings of this research showed that to obtain results that are statistically significant, the study would need to have a minimum sample size of 615 persons. This is because the expected effect size is 0.2, the alpha value is set at 0.05, and the required power is 0.9. The result of this inquiry highlights the importance of conducting a more comprehensive sampling to guarantee the robustness of the investigation and the trustworthiness of its future conclusions.

Data collection

Data collection was systematic and comprehensive, supporting methodological rigor. Zhejiang, a typical province in China, was chosen for this study due to its well-developed medical facilities, broad application of big data in hospitals, and firm policy promoting the practice of environmental sustainability. Forty-three hospitals were initially randomly chosen and mobilized to receive an email that described the research’s purposes. Of the 26 hospitals that were prompted were willing to participate in the research.

Within these working hospitals, a random sampling method was employed, targeting a wide range of healthcare professionals, including hospital managers, directors, supply chain and information technology (IT) specialists, and environmental sustainability officers. Respondents were purposely selected for their experience and participation in big data analytics and environmental sustainability practices within their hospitals, ensuring they had the necessary information to answer the survey questions truthfully and accurately. Internship students working in these institutions were tasked with distributing and collecting measurements to aid in data collection and increase the response rate.

The questionnaire was created in English and then translated and back translated by a certified, professional translator to ensure cultural and linguistic equivalence. Two bilingual experts ensured that the translated version accurately reflected the original content and maintained the same meaning as the survey items. The survey was approved by the Ethics Review Board of the Yongjia County People’s Hospital (Ref. No.: 2024-YL01-LXQ-01), and all methods were carried out in accordance with the relevant guidelines and regulations. Participants were informed of the study’s aims and objectives, as well as their right to participate or decline participation, or to discontinue at any time. It was made known to the respondents that their responses would be kept confidential and used only for academic purposes. All participants provided written informed consent.

In total, 700 paper-and-pencil questionnaires were distributed, resulting in an impressive response rate of 96.1% (n = 673). Following the exclusion of 10 incomplete questionnaires, a final sample of 653 was retained for analysis. To reduce, to the extent possible, the contribution of standard method variance resulting from the use of self-report measures, explicit instructions, assurances of anonymity, and positional isolation of predictor measures from criterion measures were included in the questionnaire design. The analysis of the response rate, comparing “early responders vs. late responders”, showed no significant differences, suggesting a low non-response bias. Possible reverse causality was controlled for by grounding the conceptual model and the items measuring the constructs in well-established theoretical approaches and in the literature, thereby ensuring unambiguous directionality in the hypothesized relationships. In addition, the study employed SEM, which allows for more rigorous testing of causal assumptions. These methodological protections correspond to those suggested by Guide Guide Jr and Ketokivi⁵⁹ and Lu, Ding⁶⁰, which, as stressed, enhance the strength, trustworthiness, and transferability of the study results.

Expanded methodological details

Study region and rationale

The research in this paper takes place in Zhejiang Province, China, which was chosen due to its well-developed healthcare structure, active policy on digital transformation, and regional focus on green and sustainable development. Hospitals in this region invested significant resources in big data analytics and environmental efforts, thus providing a context ripe for investigating the relationships proposed in our research model.

Respondents’ roles and expertise

The recipients of this study were professionals directly involved in data analytics, supply chain management, environmental sustainability, and strategic decision-making at the hospital site. The respondents consisted of hospital managers, high officials, IT specialists, supply chain managers, and environmental managers. All contributors were experienced and well-versed in big data applications and scalable solutions, particularly when it came to sustainability planning and provisioning within their organizations. To enhance the quality of responses, we employed a filtering question to verify whether the participants were actively involved or in a position to make decisions.

Sample representativeness and justification

Of the 43 hospitals approached in urban and rural Zhejiang Province, 26 agreed to participate in the trial, providing a varied range of organizational and geographical settings. From these, a final sample of 653 usable responses was obtained. It has an adequate sample size for Structural Equation Modeling (SEM), and the range of roles of the respondents and type of hospitals contributes to a satisfactory degree of internal representativeness of hospitals in the area. Nevertheless, as emphasized in the revised Limitations section, we are aware that generalizability to other industries or countries should be considered carefully.

Addressing methodological risks

We employed procedural remedies to diminish common method bias, such as anonymity guarantee, physical separation of measures in the instrument, and the use of reverse-coded items. We checked for non-response bias by comparing early with late responders on selected demographic and construct-related variables and found no differences. From a reverse causality perspective, we theoretically based the model on the S–O–R model and utilized the recommendations of Guide Jr and Ketokivi⁵⁹ and Lu, Papagiannidis⁶¹, which were cross-sectional but theoretically directional in study design. Although we are unable to make a claim of causality, the model design, participants’ experience, and the strength of the SEM analysis support some.

Results

Descriptive statistics

There was a significant gender gap within the sample, as evidenced by the fact that the majority of the participants in the study were male (68%), while only 32% were female. The demographic composition of the study’s participants provides a broad mix. There is a significant age gap between the younger and middle-aged groups, as indicated by the distribution: 23% of the participants are under the age of 25, which indicates that they are relatively early in their careers; the age group that is the largest is between 31 and 40 years old, which accounts for 38% of the total, followed closely by those who are between 26 and 30 years old, which accounts for 33% of the total, demonstrating a substantial representation from the core age groups of the workforce; participants who are over 40 years old are the least represented, accounting for 6% of the total. The sample is diverse from a professional standpoint: Hospital Administrators and Executives make up 25%, highlighting a significant involvement from top-level management; Supply Chain Managers and Professionals account for 16%, indicating a focused interest from those directly involved in logistics and operations; IT Specialists are the largest group at 37%, reflecting the increasing importance of technology in healthcare settings; Environmental Sustainability Officers constitute 22%, underscoring the growing emphasis on sustainable practices within hospitals.

Validity, reliability, and multicollinearity analysis

Fornell and Larcker⁶² have established criteria that must be met to certify the validity and reliability of a Structural Equation Modeling (SEM) analysis. To ensure the accuracy of each underlying variable in a study, it is essential to attain a Cronbach’s alpha coefficient of 0.7 or above. This threshold guarantees that the items used to measure a hidden variable are adequately correlated, indicating that they can be considered reliable indicators of that variable. The data reported in Table 2 of the study demonstrates that all latent variables satisfy this requirement, as indicated by Cronbach’s alpha values exceeding the threshold of 0.7. This accomplishment provides substantial evidence for the credibility of the study, verifying that the variables are assessed with precision and consistency, therefore accurately representing the intended notions.

Table 2.

Validity, reliability, and multicollinearity analysis.

Variables	Cronbach alpha	AVE	VIF
Big data analytics capability	0.823	0.638	[3.87, 4.27]
Decision making quality	0.761	0.702	[3.76, 4.68]
Supply chain innovation	0.773	0.568	[3.09, 4.16]
Risk taking	0.723	0.602	[3.11, 4.13]
Environmental performance	0.768	0.619	[2.96, 3.72]

In addition, the Average Variance Extracted (AVE) is an important metric for assessing the dependability of a construct. It quantifies the amount of variance in the indicators that can be attributed to the underlying variable. Segars⁶³ suggests that an AVE value of 0.5 indicates good reliability, indicating that the construct explains over half of the variance of its indicators. Furthermore, the study considers the Variance Inflation Factor (VIF) to assess the possibility of multicollinearity among the variables, in addition to the aforementioned measures of validity and reliability. Based on the research conducted by Hair, Black⁶⁴, VIF values that are less than 5 are considered acceptable, suggesting a manageable level of linear dependence between the variables. By meeting the requirements for Cronbach’s alpha, AVE, and VIF, the study not only confirms the reliability and validity of its measurement model but also guarantees the accuracy of its structural model analysis. This offers a strong basis for evaluating the correlations between variables.

Model fitting

Fit indices are crucial in the field of the SEM since they assess the degree of correspondence between a proposed model and the observed data. Several indices are available to evaluate various elements of model fit, and values above 0.9 often indicate a strong correspondence between the model and the data. The study being discussed shows that the Goodness of Fit Index (GFI), Relative Fit Index (RFI), Incremental Fit Index (IFI), Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), and Normed Fit Index (NFI) all indicate acceptable scores, indicating a strong agreement between the model and the data. Furthermore, the confirmation of the model’s suitability is reinforced by the fact that the Root Mean Square Error of Approximation (RMSEA) value is less than 0.05. Lower RMSEA values indicate a stronger alignment with the observed data. The comprehensive information shown in Table 3 of the study strongly confirms the suitability of the research model’s alignment, hence establishing confidence in the model’s structural integrity.

Table 3.

Model fitting analysis.

Index	Value
GFI	0.918
RFI	0.931
IFI	0.922
CFI	0.923
TLI	0.939
NFI	0.913
RMSEA	0.031

The Tucker-Lewis Index (TLI), which stands out for having the highest value among the fitting indices in this study, warrants further consideration. The TLI is unique because it evaluates both the adequacy of the model and takes into account the level of complexity of the model by applying a penalty. Contrary to other indices, a high TLI signifies that any increase in complexity in the model, such as adding more parameters or relationships, is justified by a substantial improvement in fit. Essentially, if the TLI is the greatest among the examined fit indices, it indicates that the suggested model provides a more accurate representation of the data compared to a null model, which implies complete independence across variables. Hence, a high TLI score not only confirms a strong match but also indicates that the model’s intricacy is warranted, resulting in a more detailed and precise representation of the underlying data pattern. This feature emphasizes the effectiveness of the model in accurately representing the connections between variables without additional complexity, emphasizing the trade-off between the complexity of the model and its ability to explain phenomena.

Structural model

The presented data in Table 4 is a set of connections between big data analytics Capability and many important organizational characteristics, with each aspect being measured using correlation or effect sizes. The influence of Big Data Analytics Capability on Supply Chain Innovation is very substantial, with a notable effect size of 0.654. This indicates that being skilled in managing large volumes of data is significantly associated with enhancements and progress in supply chain operations. Data analytics competence has a positive impact on Decision Making Quality and Environmental Performance. The effect sizes for these impacts are 0.521 and 0.523 respectively. This means that having a strong skill in data analytics contributes to improved decision-making and environmental strategies within a business. The impact of Big Data Analytics Capability on Risk Taking is somewhat favourable, with a coefficient of 0.432. This suggests that while significant, it is just one of several factors that can influence an organization’s inclination to undertake more risky endeavours.

Table 4.

Direct estimations.

H	Relationships			Standardized	P-value
				Estimates
H1	Big data analytics capability	→	Supply chain innovation	0.654**	< 0.001
H2a	Big data analytics capability	→	Decision making quality	0.521**	< 0.001
H2b	Big data analytics capability	→	Risk taking	0.432**	< 0.001
H2c	Big data analytics capability	→	Environmental performance	0.523**	< 0.001
H3a	Supply chain innovation	→	Environmental performance	0.605**	< 0.001
H3b	Decision making quality	→	Environmental performance	0.367**	< 0.001
H3c	Risk taking	→	Environmental performance	0.302**	< 0.001

Significant *p < 0.05, **p < 0.001.

The impact of Supply Chain Innovation on Environmental Performance is notably significant (0.605), indicating the direct advantages of innovative supply chain methods on sustainable results. The association between Decision Making Quality and Risk Taking and Environmental Performance is favourable, but not very strong. The impact size for Decision Making Quality is 0.367, while the effect size for Risk Taking is 0.302. both data indicate that although these factors have an influence on environmental outcomes, their effect is not as significant as Supply Chain Innovation. This suggests that the impact of both factors is more indirect or complementary in driving environmental performance. In summary, the data emphasizes the significant impact of Big Data Analytics Capability on creating a favourable environment for creativity and efficient performance, with varied levels of influence on different organizational elements.

Mediation analysis

Bootstrapping is a highly rigorous statistical technique, and the present study used the bootstrap technique to test the mediating effects of supply chain innovation, decision-making quality, and risk-taking in the relationship between big data analytics capability and environmental performance. A bootstrapping analysis using 5,000 bootstrap samples for the direct and indirect effects of the mediator variables was conducted using the AMOS program, and bias-corrected and percentile confidence intervals were employed. Results showed that all (both direct and indirect) paths were significantly different from zero (p < 0.001), as indicated by 95% confidence intervals not crossing zero and p-values < 0.05.

In this paper, we have employed partial mediation to address what we consider a contradiction with only simultaneous mediation, highlighting the meaningfulness of both the direct (See Table 4) and indirect paths (See Table 5) between big data analytics capability and environmental performance. More precisely, the direct effect of big data analytics capability on environmental performance (β = 0.523, p < 0.001) was still exposed among the indirect ones through supply chain innovation (β = 0.396, p < 0.001), decision-making quality (β = 0.191, p < 0.001) and risk taking (β = 0.130, p = 0.021). The standardized coefficient for the direct effect is larger than those for the indirect effects, suggesting that the mediations are partial and relatively weak; however, the significance supports the important and complementary mediating role of supply chain innovation, decision-making quality, and risk-taking in each. Our interpretation is consistent with that of Zhao, Lynch Jr⁶⁵, who argued that partial mediation occurs when both the direct and indirect paths are significantly different from zero. We specified the type of mediation and our interpretation of these effects in the revised manuscript.

Table 5.

Indirect analysis.

Path	Effect	S.E.	P-value	Bias-Corrected 95% CI		Percentile 95% CI
				Lower	Upper	Lower	Upper
BDAC → SCI → EP	0.396	0.067	< 0.001	0.267	0.419	0.303	0.434
BDAC → DMQ → EP	0.191	0.083	< 0.001	0.136	0.231	0.133	0.209
BDAC → RT → EP	0.130	0.029	0.021	0.088	0.162	0.079	0.169

BDAC Big Data Analytics Capability; SCI Supply Chain Innovation; DMQ Decision Making Quality; RT Risk Taking; EP Environmental Performance

Discussion

This study aims to fill a notable research gap by investigating the potential advantages of investing in big data analytics capability, particularly in the context of enhancing supply chain innovation and improving the environmental performance of hospitals. The underlying concept is that hospitals may utilize big data analytics to improve the efficiency and resilience of their supply chains, as well as increase their sustainability practices and decrease their environmental impact. The study utilizes the SOR paradigm⁵³ to investigate the correlation between big data analytics capability, supply chain innovation, decision making quality, risk taking, and hospital environmental performance. This framework is employed to comprehend the influence of external stimuli (Stimulus: big data analytics capability) on the internal state of an organization (Organism: supply chain innovation, decision making quality, and risk taking), which ultimately shapes its responses (Response: environmental performance). The proposed model in this study was rigorously examined using SEM, an advanced statistical method meant to clarify the intricate network of connections, intermediary effects, and causal directions among variables.

The Cronbach’s α value ranges from 0.631 to 0.872, while the factor loading in the final model for each measurement variable ranges from 0.711 to 0.832. The AVE for each latent variable ranges from 0.501 to 0.615. Therefore, the reliability and validity of the study model have been confirmed. The results of the SEM analysis indicate that all the hypotheses proposed in the study were confirmed. This is consistent with the theoretical predictions and expectations.

The results of this study support this group of hypotheses by verifying a significant positive association between big data analytics capability and hospital environmental performance. It illustrates the pivotal contribution of data-based methodologies not only to enhanced operations but also to the overall enhancement of sustainable development in healthcare. This result is consistent with previous findings reported by Benzidia, Bentahar¹⁷, Benzidia, Makaoui¹², and Batko and Ślęzak²⁷. Hospitals that strategically invest in and effectively implement big data analytics tools and techniques generally experience notable improvements in their sustainability initiatives. The observed positive correlation clearly indicates that an increase in big data analytics capabilities directly translates into enhanced environmental outcomes. This improvement can be attributed to various factors, such as more efficient resource management, reduced waste achieved through optimized supply chain operations, and the integration of eco-friendly practices driven by insights gained from data analytics.

The second set of hypotheses (H2a–H2c) emphasizes the strategic role of big data analytics capability in enhancing internal processes and its impact on shaping supply chain innovation, decision-making quality, and risk-taking within hospitals. These are based on the observation that big data analytics enables healthcare organizations to handle large-scale and complex datasets, thereby providing them with the opportunity to achieve more creative, informed, and flexible responses to operational and strategic challenges. Through increased analytical capabilities, hospitals can more effectively transform their supply chains, utilize data to inform decisions, and take targeted risks that advance long-term sustainability objectives.

Findings that support the second group of research hypotheses suggest that big data analytics capability has a direct positive effect on innovative supply chain practices. This finding aligns with previous studies^7,29,31, which also emphasize the importance of big data analytics in enhancing supply chain innovation. Novelty: The paper provides empirical evidence supporting and extending previous insights regarding the positive impact that organizations dedicated to enhancing their big data analytics capabilities (i.e., investments in technologies, tools, and skilled professionals to collect, manage, and interpret large datasets) have on their supply chain innovation. In addition, our findings further generalize the findings by presenting that the hospital sector can use big data to improve logistics and inventory management, realize more sustainable sourcing, and respond more effectively to dynamic market conditions³⁰. Unlike previous studies, mainly in the context of manufacturing and retail, our research contributes by illustrating the impact of big data analytics capabilities on supply chain innovation in the healthcare sector, where efficiency and sustainability are becoming increasingly important. This hospital-based extension constitutes an original contribution to the literature on big data and supply chain management.

Hypothesis H2b of the study posits that the capability for big data analytics is positively associated with the quality of decision-making within hospitals. This investigation provides support for certain previous studies in general and healthcare/ or hospital studies^32,34,36 which suggests that as companies enhance their ability to gather, process, and analyse large datasets, the quality of their decision-making processes improves. One possible reason is that big data analytics provides a wealth of information and insights, which can help decision-makers understand complex situations better, forecast future trends with more accuracy, and assess the potential outcomes of different decisions more effectively. This improvement can be attributed to the ability of big data analytics to reduce uncertainty and provide a solid empirical basis for decisions. By leveraging data-driven insights, organizations can minimize the risks associated with decision-making under uncertainty, leading to choices that are more aligned with strategic objectives and expected outcomes^34,35.

Hypothesis H2c examines the effect of big data analytics capability on hospitals’ risk-taking tendencies, suggesting that this capability enhances their willingness of hospitals to experiment and take calculated risks. This hypothesized association is in line with previous studies such as those by Tipu and Fantazy²¹, Jalali, Palalić²⁰, and Basile, Carbonara³⁸, who stress that envelope-pushing environments may foster strategic risk behaviour. Yet, much of the literature has focused on firms in the manufacturing or commercial sector, and thus, little is known about this relationship in healthcare, a significant gap that the current study addresses. Hospitals that have embraced big data to a greater degree are not just freer with their risk- taking; they are also more capable of assessing, controlling and selectively acting on such risks based on predictive models and proof- based planning³⁹. This enables them to innovate in treatment protocols, operational models, or technology implementation with reduced risk and uncertainty. For example, such analytics could help identify excess capacity or predict the impact of using new medical technologies³⁷, and thus reduce the high levels of uncertainty that often impair innovation in the healthcare system. In doing so, we offer a critical empirical extension and analysis of this relationship within hospital settings, thereby providing unique and timely insights that contribute to the literature on digital transformation and organizational risk behaviour in complex, regulated contexts, where the role of big data remains largely underexplored.

The third group of hypotheses (H3a–H3c) examines how internal organizational capabilities; specifically, supply chain innovation, decision-making quality, and risk-taking; directly impact hospital environmental performance. These constructs represent critical levers through which hospitals can translate strategic intent into sustainable outcomes. Supply chain innovation enables greener logistics and resource efficiency; high-quality decision-making ensures that sustainability is factored into operational choices; and a willingness to take calculated risks facilitates the adoption of novel environmental practices and technologies. Together, these capabilities reflect the internal mechanisms that support an institution’s shift toward environmental sustainability.

The current study uniquely incorporates supply chain innovation (H3a), decision-making quality (H3b), and risk-taking (H3c) as a complementary set of analytic lenses, providing a more comprehensive understanding of how supply chain innovation, decision-making quality, and risk-taking collectively contribute to hospital environmental performance. Although such antecedents have previously been studied in isolation (e.g^40,42,46, relatively little attention has been paid to their combined effects in a hospital context, particularly in the context of big data analytics capabilities. This study addresses a significant theoretical gap in the literature regarding the collaborative impact of connected managerial capabilities on achieving environmental outcomes in healthcare organizations. H3a examines the relationship between supply chain innovation and hospital environmental performance. Although studies like that of Fu, Yang⁴⁰ and Alkhatib⁴¹ recognize that sustainability supply chain practices are essential, their applicability in hospitals is less understood. Our results indicate that hospitals can achieve significant environmental savings by integrating sustainable logistics, applying predictive analytics to inventory management, and aligning their suppliers with ecological criteria. In contrast to extant literature that primarily overlooks the healthcare sector, we develop a theoretical perspective where healthcare organizations are considered as active players in sustainable evolution and development, driven to impact upstream and downstream supply chain management practices. This is not only beneficial for the environment but also aligns with the broader social (public health) responsibility of the healthcare industry. By framing our results within this cross-sectoral conversation, we challenge the current siloed approach and call for an integrated model of environmental practice—one that is compatible with the complex and hybrid nature of hospitals and their capacity for environmental leadership.

This study demonstrates that the quality of decision-making has a significant influence on the environmental performance of hospitals, as supported by earlier research^42,43. One of the limitations we found in the literature is the lack of attention to how the quality of decision-making and risk-taking, characterized by big data analytics, enhances sustainability performance in hospitals. Our results help to fill such a gulf, indicating that sound environmental decision-making in healthcare settings will hinge on making timely, evidence-based decisions that balance operational constraints against longer-term ecological consequences. This reinforces and builds upon prior research^45,66 in this area, which demonstrates how hospitals that adopt sustainability assessments (e.g., environmentally preferable purchasing, sustainable supplier evaluations, investments in energy-saving technologies) achieve greater consistency between operational choices and environmental targets. Moreover, while previous research (e.g⁴⁶, has established a link between organizational risk-taking and environmental innovation, few studies have explored this dynamic within hospitals. Our enabling factor, H3c, raises awareness that taking calculated risks is a prerequisite for environmental leadership in healthcare, particularly in embracing disruptive green technologies and novel waste management practices. These costly and uncertain efforts, however, can reap significant environmental rewards. With this conclusion situated within the broader literature on organisational sustainability, our work not only ratifies but also advances the debate by demonstrating how hospitals can be agents of environmental transformation, driven by open, data-enriched decision-making and courageous risk leadership. This fusion of empirical and theoretical dialogue directly addresses the previously identified voids and expands the academic conversation regarding sustainability within the healthcare industry.

The study’s findings indicate that supply chain innovation plays a crucial role as a partial mediator in the connection between several organizational aspects (such as big data analytics capability, decision-making quality, and risk-taking) and the environmental performance of hospitals. Consequently, the influence of these organizational characteristics on environmental performance is partially mediated by supply chain innovation. Put simply, these characteristics have a direct impact on environmental performance, but they also indirectly promote innovations in the supply chain.

This outcome aligns with the assertions of the study, confirming the crucial significance of supply chain innovation in improving environmental sustainability in hospitals. Supply chain innovation include strategies such as developing environmentally friendly procurement policies, optimizing logistics to decrease carbon emissions, and embracing circular economy concepts to minimize the generation of waste⁶⁷. Supply chain innovation acts as a mediator, transforming the organization’s capability and strategic behaviours into concrete environmental gains.

The study’s findings emphasize a notable pathway by which the capability of big data analytics might improve the environmental performance of hospitals. The quality of decision-making and the willingness to take risks play crucial roles as mediators in this relationship. This result highlights the complex relationship between an hospital’s analytical abilities and its strategic actions, demonstrating how the former can lead to enhancements in sustainability through a two-step mediated process. At first, the capability of big data analytics enhances the quality of decision-making by offering thorough, precise, and timely information. The abundance of data aids hospitals in making well-informed decisions, enabling them to thoroughly assess the environmental consequences of their actions and select methods that are in line with their sustainability objectives¹². Improved decision-making quality acts as the initial mediator, converting data into practical insights that can result in superior environmental results. According to the study, hospitals are more likely to take strategic risks in order to achieve their environmental performance targets as a result of using this enhanced decision-making process⁶⁸. The second mediating stage is represented by the willingness to take risks, which is fostered by the confidence in their decision-making processes. Hospitals may choose to include novel, eco-friendly technologies or embrace inventive operating methods that, although potentially hazardous, hold the potential to substantially diminish their ecological impact⁶⁹. Hence, engaging in risk-taking serves as a link between making well-informed choices and carrying out actions that have the potential to result in environmental enhancements.

Theoretic implications and applications

The theoretical implications of these findings are significant, as they enhance our comprehension of the complex interactions among big data analytics capability, decision-making quality, risk-taking, and environmental performance, specifically in the realm of hospital operations. This study contributes to the existing body of knowledge by presenting a sequential mediation model that illustrates how the abilities in big data analytics increase the quality of decision-making. This, in turn, promotes prudent risk-taking, resulting in enhanced environmental performance. This model emphasizes the significance of taking into account both the immediate and indirect impacts of big data analytics on sustainability results, indicating that the route to environmental enhancements is complex and requires a sequence of internal organizational changes. The results enhance the SOR theoretical framework by presenting empirical proof of how particular organizational capability (stimuli) impact internal processes (organism responses) that result in specific outcomes (responses). This offers a detailed comprehension of how technology-driven capability can influence organizational behaviour and sustainability efforts.

From an application perspective, these results have practical ramifications for hospital administrators and legislators who are striving to enhance environmental sustainability. Hospitals may improve their data analytics capability to not only enhance operational efficiency but also use it as a strategic tool for sustainability by acknowledging the importance of decision-making quality and risk-taking as mediators. This entails allocating resources towards the development of data analytics infrastructure and providing training to employees. It also involves cultivating a work environment that encourages effective decision-making and strategic risk-taking. Additionally, it requires incorporating sustainability factors into the decision-making procedures. These findings emphasize the need of legislators implementing policies that promote the adoption of big data analytics in healthcare facilities, with a specific focus on sustainability. This may be achieved by providing incentives for green technologies and investing in data-driven decision-making frameworks. In summary, the report offers a clear plan for using technology to improve environmental performance, directing both professionals and politicians towards more sustainable practices in the healthcare industry.

Policy implications

Applying the results of our study, we provide with five customized policy implications that can then be articulated for hospital administrators, health policy makers, and regulators:

Advocate for data-driven hospital environmental policies

Hospitals should be incentivized to incorporate the institutional capacity for big data analytics into their environmental strategy. Policies allow investment in digital infrastructure, which allows for real time monitoring of energy usage, waste management, and resource use. Hospitals can further harmonize their operations with their sustainability objectives by incorporating data analytics into environmental compliance and reporting requirements.

Enhance data literacy and long-term support for building capacity

Considering the importance of quality decision-making, decision-makers should commit themselves to training programs that improve hospitals’ employees’ skills in interpreting data related to sustainable activities. This encompasses tailored workshops, certifications, and continuing education with a focus on data analytics, green healthcare operations and evidence-based environmental planning.

Rewarding supply chain innovation in healthcare

To enhance the mediating role of supply chain innovation in environmental performance, the healthcare policies should develop procurement reforms, supplier assessment criteria and eco-innovative logistics trends as well. Government-friendly facilities, tax breaks, or funding grants could be offered to hospitals that implement sustainable supply chain changes brought on by big data.

Institutionalize risk-informed innovation policies

Our results validate risk-taking as an intermediate factor between big data capability and environmental performance. Lawmakers should therefore encourage systems that incentivize responsible innovation in new eco-technologies or business models applied within hospitals. This could mean regulatory sandboxes or innovation testbeds in the public health space.

Big data strategies, regional vision

In view of the results of our study, which focuses on Zhejiang Province, it is essential for policymakers to recognize that opportunities to drive big data roll-out must be context-specific. Local government entities can develop region-specific guidance for hospitals that considers the technological preparedness, regulatory readiness, and environmental needs of a region, making the policies more relevant and increasing the likelihood of successful implementation.

Limitations and future studies

A key limitation of our study is the focus on the hospital sector of only one province in China. Although the focus on the Zhejiang healthcare system represents a strength of our study, it also introduces limitations, as the findings may not be generalizable to different regions, nations, and industries. Hospitals operate within unique regulatory, cultural, and institutional environments that may not be applicable to other industries or geographies. Therefore, caution should be exercised when generalizing the results to other medical settings. This research is to be seen as exploratory, with a specific interpretative horizon and not as making any general claims for universal validity.

Furthermore, the self-reported nature of the data raises questions regarding response bias, particularly in areas such as environmental performance and organizational risk-taking, where the social desirability response set may influence answers. While actions were taken to maintain the respondent’s anonymity and minimize bias, this remains a methodological limitation.

Another limitation is the dynamic nature of big data technologies and sustainability standards. Given the recent development in these domains, the associations found in this study may, in turn, be time attenuated. Future work should explore more longitudinal designs to determine how these relationships evolve as technology advances and regulations change.

In addition, the quantitative design of the study, although appropriate for testing the model developed, may fail to address the more interactive and qualitative aspects of human decision-making and risk-taking in a hospital setting. Prospective research may use mixed approaches such as combining quantitative data with qualitative interviews or case study data to gain a richer understanding of the mechanisms underlying the associations. Cross-sector and multi-country comparisons are also proposed to confirm and generalize the findings.

Conclusions

Considering these constraints, future research endeavors could broaden their investigation to encompass a wider array of sectors, enabling a comprehensive analysis of how the connections between big data analytics capability, decision-making quality, risk-taking, and environmental performance materialize in diverse settings. Qualitative research methods, such as case studies or interviews, can enhance quantitative findings by providing more profound insights into the mechanisms and processes by which big data analytics impact decision-making and risk-taking. Furthermore, longitudinal research would yield useful insights into the progression of these linkages over time, specifically as technology and sustainable practices continue to develop. Further exploration of these areas would not only overcome the limits of the current study but also enhance our comprehension of the strategic significance of big data analytics in advancing environmental sustainability in many sectors.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

SOR

Stimulus organism response

SEM

Structural equation modeling

AVE

Average variance extracted

VIF

Variance inflation factor

GFI

Goodness of fit index

RFI

Relative fit index

IFI

Incremental fit index

CFI

Comparative fit index

TLI

Tucker-Lewis index

NFI

Normed fit index

RMSEA

Root mean square error of approximation

BDAC

Big data analytics capability

SCI

Supply chain innovation

DMQ

Decision making quality

RT

Risk taking

EP

Environmental performance

Author contributions

L.X, Y.X, Y.S collected the study data. H.S.J & N.S wrote the article and performed statistical analyses. L.X & Y.X read the article and made the necessary checks for its correction. Then all of them approved the article.

Data availability

The data are not publicly available due to the Yongjia County People’s Hospital Ethics Review Committee rules and regulations. The data that support the findings of this research are available upon reasonable request from the corresponding author and with permission of the Yongjia County People’s Hospital Ethics Review Committee.

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The survey was conducted with the Yongjia County People’s Hospital Ethics Review Committee (Ref No: 2024-YL01-LXQ-01) approval. Participants of the study were informed about the purpose, objectives, and their right to participate, decline participation, or withdraw their participation in the research activities by verbal. Respondents have been notified that the information given was private and confidential which only going to use for academic purposes only. Written informed consent was obtained from all respondents.

Informed consent

Informed consent was obtained from all subjects involved in the study. Written informed consent has been obtained from the participants to publish this paper.