An Interpretive Structural Modeling Approach for Biomedical Innovation Strategy Models with Sustainability

Abstract

In recent years, the biomedical startup industry has flourished, and yet, it still faces challenges in adapting to changing market demands. Meanwhile, the widespread use of single-use medical devices generates significant waste, posing threats to environmental sustainability. Addressing this issue has become a critical challenge for humanity today. The study aimed to delve into the specific difficulties faced by Taiwanese biomedical entrepreneurs during the innovation and development of medical devices from a sustainability perspective and to explore solutions. This study collected first-hand experiences and insights from Taiwanese biomedical entrepreneurs through a literature review and expert questionnaires. It employed Interpretive Structural Modeling to analyze the development stages and interrelationships of biomedical device startups for building sustainable biomedical innovation. The Clinical Needs Assessment is revealed as the most influential factor, shaping Regulatory Feasibility Evaluation, Clinical Trial Execution, and Market Access Compliance. Our findings provide a structured problem-solving framework to assist biomedical startups in overcoming challenges while incorporating energy-saving and carbon reduction processes to achieve environment sustainability goals. The results of this study show that biomedical innovation practitioners should prioritize integrating sustainability considerations directly into the earliest stage of a Clinical Needs Assessment.

1. Introduction

Climate change is a major global issue that is having a profound impact on human society and the environment. COP26 called for urgent climate action to halve global greenhouse gas emissions by 2030 and achieve net-zero emissions by 2050 to address the global climate crisis. Rachel Hennein et al. noted that the U.S. healthcare industry is one of the largest polluting industries with significant adverse effects on human health [1]. Taiwan is globally renowned for its development and manufacturing of hardware technologies, and the vast majority of countries worldwide engage with Taiwan as part of technology- or product-related interactions. Therefore, in the context of global efforts to reduce carbon emissions, the sustainable development of the healthcare industry is a key area of focus for Taiwan. Taiwan’s total greenhouse gas emissions increased from 137,881 thousand metric tons in 1990 to 297,007 thousand metric tons in 2021, representing a 115.41% increase. Of this total, carbon dioxide emissions accounted for 95.32% [2]. In recent years, the biomedical startup industry has flourished but still faces challenges in adapting to changing market demands. In addition, the widespread use of single-use medical devices generates significant waste, posing a threat to sustainability and the environment. Biomedical technology innovation, encompassing new drugs, devices, and diagnostics, is crucial for advancing healthcare quality and reducing costs [3]. While the integration of technology is essential, financial constraints challenge R&D [4]. The COVID-19 pandemic underscored the urgent need for enhanced healthcare capacity and green healthcare, further emphasizing the importance of biomedical innovation and prompting calls for increased government support [5]. Currently, Taiwan does not have mandatory regulations requiring sustainable medical devices, resulting in a lack of incentives for many startups and small- to medium-sized medical device companies to proactively engage in sustainable design. However, in 2021, the Taiwanese government declared carbon reduction as a policy goal and introduced a series of policy announcements and draft legislation. The biomedical sector is already governed by complex legal frameworks. Nevertheless, for biomedical professionals in Taiwan, how to integrate sustainability concepts into existing business operations remains at an exploratory stage.

This study analyzes the annual number of outpatient visits in Taiwan, the average travel distance per visit, the average fuel efficiency of gasoline-powered vehicles, and the gasoline carbon emission coefficient announced by the Ministry of Environment. Based on these parameters, it is estimated that medical travel in Taiwan generates approximately 1,280,824 metric tons of CO₂-equivalent emissions annually, as shown in

Table 1. Annual carbon emissions from healthcare visits by people in Taiwan.

Item	Value
Average medical distance	17.68 km
Carbon footprint coefficient of 1 L of automotive gasoline	2.92
Number of medical visits	412,587,914 times/year
Average fuel consumption	16.63 km/L
Annual carbon emissions	1,280,823,728 kg of CO2 equivalent (CO2e)
Total CO2e emissions	17.68/16.63 × 2.92 × 412,587,914 = 1,280,823,728 kg/annual CO2e emissions

[6,7,8,9], highlighting the urgent need to integrate biomedical innovation and green healthcare in order to address climate-related health risks [1,10,11,12,13,14,15].

Biomedical innovation enhances care quality and may reduce costs, but rising cost scrutiny and the COVID-19 pandemic have highlighted the urgent need for green healthcare, resilient infrastructure, and greater public investment in medical R&D and manufacturing [3,5,12,13,16].

Medical device development, a process that requires rigor and high risk, emphasizes the safety and efficacy of the end product [17,18]. Biomedical innovation research remains fragmented and overly specialized, lacking structured models to manage complex innovation steps and the assessment of emerging risks such as AI integration. This gap, along with insufficient strategic analysis in the medical industry, highlights the need for a comprehensive innovation framework [4,19,20,21,22,23]. The current business model in medical device manufacturing prioritizes single-use items over durable or reusable equipment, hindering environmentally conscious design [24]. The estimated carbon footprint per case is 2.40 kg CO₂ for single-use flexible cystoscopes and 0.53 kg CO₂ for reusable flexible cystoscopes [25].

Given the high environmental impact of the medical device industry, in continuous medical device innovation, while there have been studies investigating the carbon emissions associated with the medical device innovation process and the devices themselves, a research gap exists regarding the systematic integration of sustainable consideration to achieve environmental sustainability. To address the gap, this study aims to provide medical device stakeholders with an innovation strategy to enhance the sustainability of the medical device industry.

Interpretive Structural Modeling (ISM) is a modeling tool used to analyze the relationships between specific aspects of a defined problem, classify variables based on their importance, and provide managerial insights. It offers valuable information for deeper understanding and practical applications, particularly in sustainability research [26].

Biomedical innovation is an essential approach for enhancing the status of a nation’s biomedical industry, and the development of medical devices constitutes a significant aspect within the domain of biomedical innovation. The Biotechnology Industry White Paper published by Taiwan’s Industrial Development Bureau of the Ministry of Economic Affairs also includes medical devices within the scope of the biotechnology industry. Specifically, it leverages Taiwan’s outstanding capabilities in hardware development and manufacturing and revenue from the medical device sector, including that of healthcare devices, which accounts for 50% of the total biotechnology industry revenue in Taiwan [27], as these are representative within Taiwan’s biomedical industry.

This study identifies authoritative experts and scholars within Taiwan’s biomedical industry. Using expert opinions as a foundation, it integrates considerations in the areas of clinical needs, regulatory evaluations, product manufacturing, and market commercialization to establish a structured strategic model for sustainable biomedical innovation.

This study employs ISM and the Matrix of Cross-Impact Multiplications Applied to Classification (MICMAC) to identify the key drivers of sustainable medical device innovation and their interrelationships. It provides an effective innovation strategy to address this research gap, advancing biomedical innovation and environmental sustainability.

Specifically, although previous studies have addressed the development processes of medical devices and the carbon emissions related to these, most have concentrated on specific medical devices or healthcare fields. There is a lack of research integrating sustainability considerations throughout the biomedical innovation process, and systematic strategic models are often absent. This study fills this gap by employing ISM analysis to systematically perform a hierarchical analysis of influencing factors, identifying their interrelationships, and determining critical sustainability-related factors in biomedical innovation. Distinct from previous research focusing solely on carbon emissions or the specific development processes of medical devices, this study provides strategic recommendations for sustainable biomedical innovation aimed at entrepreneurs and industry practitioners. The biomedical innovation process and sustainability are still two distinct concepts. Therefore, how to integrate these two into a sustainable biomedical innovation process constitutes the research question that this study aims to address.

2. Literature Review

Successful biomedical innovation and the complex Medical Device Development Process (MDDP) necessitate rigorous processes and a systematic approach [21]. Growing competition demands comprehensive risk assessment throughout the MDDP, covering technical, manufacturing, regulatory, and market aspects. The uniquely complex and high-stakes MDDP, unlike general manufacturing, requires a proactive risk management approach across its technical design, manufacturing, compliance, and market adoption phases [22].

The Medical Device Innovation Process (MDIP) is more complex than the MDDP, integrating new concepts and technologies. Structural models assist in addressing innovation challenges and guiding development. MDIP factors influence the entire innovation process, from concept to market, ensuring devices are safe, effective, and meet the needs of patients and healthcare providers. However, the process of innovative development is multifaceted.

The core focus of this study is on medical device innovation within the biomedical industry. This choice is based on the fact that medical device development entails a high level of technical expertise and complex regulatory and clinical processes, representing a significant aspect of Taiwan’s biomedical industry structure and serving as a critical area for achieving green and sustainable transformation in Taiwan.

2.1. Biomedical Innovation

Producing medical devices that truly meet users’ needs requires the consideration of a variety of factors beyond simply meeting clinical needs, minimizing human error, and ensuring patient safety [28]. It is crucial to carefully analyze the clinical context to uncover unmet or inadequately addressed needs that can be improved through innovative proposals [4]. The clear identification and addressing of unmet clinical needs are crucial for successful medical device development. Thus, understanding and resolving potential mismatches are essential for successful device adoption [22]. Clinical studies with medicinal products or medical devices pose increasingly complex requirements for sponsors and participating centers. Over the past two decades, sponsors have been increasingly delegating regulatory and organizational study tasks to Clinical Research Organizations (CROs) [29]. The realization of a device from an idea, occurring during prototype construction, typically involves drafting dimensions, materials, and operation; creating a 3D model; developing a proof of concept to demonstrate functionality and feasibility; and building the prototype to identify the most efficient and marketable version [4,19].

Assessing a product’s technical concept, addressing technology gaps, and considering its clinical application help ensure comprehensive medical device design, facilitating progression to subsequent development stages [4,22,30]. This study demonstrates that a key component in the MDIP is focusing on the discovery and description of needs, which differs from the traditional “technology push” model. Physicians play a vital role in this process, especially in needs identification and description [19].

Choosing the right regulatory pathway, such as the FDA’s regulations in the US and the medical device regulations in the EU, is crucial for market entry and compliance [22,31]. Early in the MDDP, it is essential to assess market acceptance and stakeholder support; establish a viable business model, covering distribution, finance, and funding strategies; and formulate a clear marketing and value proposition. Articulating competitive advantages over existing solutions is critical for a successful product launch [19,32].

Identify strengths, weaknesses, and technological improvements, acquire key technologies, set product specifications, and align with clinical needs to enhance the acceptance of new technology, as this ensures the product aligns with specific clinical requirements [30]. To ensure efficient workflows, pilot production plans must adapt production processes to current technologies, integrating additive manufacturing and conventional techniques [33,34]. Medical device development demands precise clinical trial planning aligned with FDA and EU regulations, where the FDA oversees centralized processes, and EU regulations rely on notified bodies [19,31]. Obtaining market approval for a medical device requires meeting regulatory standards, including risk-based classification, compliance with FDA or EU regulations, and the provision of clinical data to demonstrate safety and efficacy [28,31]. Selecting appropriate supply chain partners and adapting to supply chain trends can ensure stable product supply and quality, preventing disruptions like those experienced during the COVID-19 pandemic [16].

A significant challenge for new medical device companies is the complexity of reimbursement, often requiring substantial evidence of clinical efficacy and cost-effectiveness, scrutinized by regulators and payers to ensure benefits over existing solutions [30,32]. Effective medical device development management requires a structured operating model that tracks progress through stages. As technology matures, developers consider commercialization strategies, with a robust business model central to decisions [4,30,32]. Medical device companies focus on variety-based and needs-based positioning, alongside market segmentation. Differentiation, alignment with operational efficiency, execution, and conducting market research to identify target customers and distribution channels are crucial [4,32,33].

2.2. Medical Device with Sustainability

The development of medical devices, as part of a traditionally risk-averse industry, has lagged behind other sectors in addressing environmental issues. Energy saving and carbon reduction remain critical environmental challenges.

For example, from conceptualization to prototype testing, the innovative controller prototype, which relies on model predictive control and aims to minimize operating costs and/or energy, is being developed to enhance operational efficiency [35]. In the same way, healthcare management must evaluate workflow, outcomes, economy, and environmental impacts when adopting new medical technologies [36].

Providing life cycle assessments that compare the environmental effects of single-use, reprocessable single-use, and reusable devices could facilitate health systems’ purchasing decisions [1]. Another approach to reducing carbon emissions in the medical device industry is Medical Device Reprocessing (MDR), which reduces waste, carbon emissions, and costs while enhancing supply resilience but requires regulatory and manufacturer approval [15]. In the United Kingdom, the life cycle assessment of carbon emissions from clinical trials followed the UK National Health Service’s sustainability guidance on care pathways [11]. As is well known, medical devices carry high levels of medical risk, and thus, product quality stability is necessary. Quality function deployment enables the analysis of the relationship between customer demands and product technologies and evaluates customer satisfaction levels, e.g., by demonstrating the hierarchical framework of the customer need for green smartphones [37].

MDR is a sustainability solution that can reduce hospital waste, decrease carbon emissions, cut costs, and enhance supply chain resilience; however, only a small proportion of FDA-approved devices have actually been reprocessed. The study findings emphasize the importance of educating hospital staff on MDR, ensuring the industry provides accurate and transparent life cycle assessments of medical devices, and conducting audits of MDR processes [1].

A study has suggested that the most important opportunity for implementing Design for the Environment (DfE) in the medical device industry lies mainly in the early stages of the design process. The authors also argue that the nature of the business model is critical, emphasizing systemic considerations rather than focusing solely on the product itself [24].

The medical device industry is highly regulated, and the cost and complexity of carrying out medical services to achieve market legitimacy reduce benefits to small-scale entrepreneurs, who lack the capital, knowledge, and technology to bring their products to market [38].

However, increasing the responsiveness level of the supply chain can enhance the sustainability of various areas, including job opportunities, safety, carbon emission reduction, and economic aspects [1,10,39].

On the other hand, upon considering the effects of health insurance reimbursement and consumers’ low-carbon preferences, it becomes evident that government subsidies play a significant role in promoting low-carbon investments by pharmaceutical companies and low-carbon marketing by medical institutions [40]. Moreover, it is beneficial for corporations to comply with environmental regulations and adopt sustainable business practices, which can enhance corporate financing capabilities and promote long-term sustainable development [1,15]. With the development of the low-carbon economy, it is necessary for manufacturers to produce low-carbon products and choose appropriate sales channels [41].

The increasing popularity of single-use equipment is mainly due to the reduced risk of cross-contamination and patient-to-patient infections. There are ongoing studies that are evaluating the health economics and environmental impacts of implementing these technologies on a large scale [36]. To reduce the carbon footprint generated by the use of single-use endoscopes in gastrointestinal endoscopy, promoting reusable equipment, measuring institutional emissions, setting benchmarks, and utilizing virtual platforms, such as telemedicine, are essential for achieving sustainability [25].

Recent research compared greenhouse gas emissions (in CO₂ equivalents) for reusable and single-use ureteroscopes [42] and the financial and environmental costs of single-use versus reusable anesthetic equipment [43]. MDR also enhances sustainability. Carrying out comparative life cycle assessments of the environmental effects of single-use, reprocessable single-use, and reusable devices could aid in health system purchasing decisions [1]. However, focusing on circular single-use products, where materials are reprocessed into new or the same type of products, is effective for reducing environmental impacts, provided the reprocessed material achieves comparable quality [36].

In summary, the relationship between a product’s life cycle and its environmental burdens is complex, as design decisions and external parameters, like use context or end-of-life, also shape its overall environmental impact [24,36].

Research on medical device innovation has predominantly focused on regulatory processes [31], technical feasibility [21,22], and market entry strategies [19,32]. Although some studies have addressed the development processes for specific medical devices [4,19,20,21,22,23], they have not considered integrating these processes with green innovation strategies. Sustainability research on medical devices has explored reusable versus single-use devices or MDR, with emphasis largely placed on cost or carbon emission comparisons of specific devices [42,43] or the design of reprocessing workflows [1]. Other research highlights the importance of considering the environmental impact during the early stages of medical device design [24]. However, a comprehensive strategic model for sustainable biomedical innovation has yet to be developed.

Previous studies have applied the Interpretive Structural Modeling (ISM) method to explore how to develop environmentally friendly sustainability within specific industries. For example, research on developing green supply chain management in the automotive industry has identified that maintaining environmental awareness is the primary barrier [44]. In a study focusing on sustainable supply chain management, further investigation into the eco-design of products revealed that eco-design on sequential supply chain activities at the deployment phase is a key influencing factor in achieving a sustainable supply chain [45]. In research on developing sustainable banking services based on Industry 4.0, the creation of higher economic value has been highlighted as the main factor influencing the sustainability of banking services [26].

In the healthcare sector, studies on sustainable green ergonomics have identified the design principle as a critical influencing factor [46]. In research on green operations strategies in healthcare organizations, vision and structure, as well as stakeholder pressure, have been recognized as the main factors affecting green operations strategies [47].

Fourteen process factors, classified through a literature review, are presented in

Table 2. Medical device innovation process literature review and factor content analysis summary.

No.	Factor	Description	Study
1	Clinical Needs Assessment	Confirm with physicians the impact and necessity of the product/technology on the target medical process to determine whether the clinical need is a must-have or nice-to-have.	A, B, G
2	Clinical Research Organization Cooperation	Collaborating with a CRO company to conduct clinical trials related to the product. Outsourcing to CROs offers benefits like increased flexibility, access to specialized expertise, rapid technology adoption, cost reduction, and faster time to market, which also support energy saving and carbon reduction.	K
3	Prototype Concept Assessment	Construct the product concept through prototype trials, confirm the quantitative and qualitative specifications and functional indicators, and validate the safety and efficacy methods and standards. From conceptualization to testing, an innovative controller prototype using model predictive control is developed to enhance efficiency by minimizing costs and energy use.	E, G, N
4	Technical Feasibility Evaluation	Assess whether the technical concept of the product can be realized and analyze current technical gaps. Studies show healthcare management must consider workflow, clinical outcomes, economy, and environmental impacts when implementing new technology.	A, G, H, M
5	Regulatory Feasibility Evaluation	Clarify the regulatory pathways that medical device regulations can follow to assess the appropriateness of the product design concept. MDR can reduce hospital waste, lower carbon emissions, cut costs, and improve supply chain resilience but requires clear guidelines from regulatory bodies and manufacturers.	A, J, O
6	Market Feasibility Evaluation	Evaluate market acceptance and stakeholder acceptance, as well as the reimbursement business model involving analysis of distribution channels, funding strategies, and value proposition to ensure a successful product launch.	C, E
7	Product Technology Implementation	Understand strengths and weaknesses and areas where technology needs improvement in order to acquire key technologies and set product specifications and operating mechanisms to realize product technical concepts or advantages. Quality function deployment analyzes the relationship between customer demands and product technologies and evaluates customer satisfaction levels, e.g., by showing the hierarchical framework of customer need for green smartphones.	H, P
8	Production Process Design	Develop pilot production and production planning for the product to adjust production processes accordingly. Providing life cycle assessments that compare the environmental effects of single-use, reprocessable single-use, and reusable devices could facilitate health systems’ purchasing decisions.	D, F, Q
9	Clinical Trial Execution	Establish clinical trial plans according to regulatory standards for product safety and functional efficacy. Life cycle assessments consider carbon emissions of all clinical trial phases, following sustainability guidelines by the UK National Health Service.	E, J, L
10	Market Access Compliance	Ensure the team follows regulatory and verification standards, manages certification items, and schedules planning to comply with market approval regulations. The complexity and costs associated with achieving market legitimacy can limit benefits for small-scale entrepreneurs in the context of carbon offset projects.	B, J, R
11	Supply Chain Management	Collaborate with supply chain partners to ensure smooth upstream and downstream supply development of the product. Increasing supply chain responsiveness can enhance sustainability aspects, including job opportunities, safety, carbon emission reduction, and economic factors.	I, Q, S, T
12	Healthcare Insurance Reimbursement	Assess healthcare reimbursement systems in various countries based on market layout and create strategies, such as seeking external support or consulting firms to assist in insurance code application planning. Government subsidies significantly promote low-carbon investments by pharmaceutical companies and low-carbon marketing by medical institutions.	C, H, U
13	Business Model Design	Develop an operational model for team milestones and product progress at various phases. Complying with environmental regulations and adopting sustainable practices can also enhance financing capabilities and promote long-term development.	C, G, H, O, Q
14	Sales Channels Design	Conduct market research to identify target customers and distribution channels for sales. With the development of a low-carbon economy, it is necessary for manufacturers to produce low-carbon products and choose appropriate sales channels.	C, F, G, V

A—[22]; B—[28]; C—[19,32]; D—[34]; E—[19]; F—[33]; G—[4]; H—[30]; I—[16]; J—[31]; K—[29]; L—[11]; M—[36]; N—[35]; O—[1,15]; P—[37]; Q—[1]; R—[38]; S—[10]; T—[39]; U—[40]; V—[41].

, which includes their sources and descriptions. The authors of this study used these 14 factors to develop a questionnaire for experts, through which their opinions were collected to conduct an ISM analysis, thereby building a structured model for sustainable biomedical innovation. ISM is used to structure the causal relationships of complex problems by collecting expert judgments on the interrelationships among the factors, leading to the development of an expert model that emphasizes the convergence of collective expert experience. It has been widely applied in emerging topics or areas that require multi-expert knowledge integration, such as sustainable supply chain management [44], sustainable growth for small- and medium-sized enterprises, and sustainable banking [26].

3. Methods

This study carried out the stratified sampling of biomedical industry participants in Taiwan, screening those with over 10 years of experience. Out of 60 questionnaires sent, 52 were returned (86.67% response rate), but 17 were removed due to irrelevant answers, leaving 35 valid responses.

Table 3. Background information of questionnaire participants—Interpretive Structural Modeling.

Background Information	Category	Sample Size	%
1. Function (N = 35)	Executives/leaders	22	62.86%
	Academics/researchers	7	20.00%
	Professionals/experts	6	17.14%
2. Current position (N = 35)	Innovation team member	27	77.14%
	Review committee	8	22.86%
3. Years of investment in the biomedical industry (N = 35)	11~20 years	25	71.43%
	More than 21 years	10	28.57%

presents the participants’ background information. Furthermore, the experts selected for this study consist of review committee members for government-sponsored biomedical projects, physician–entrepreneurs, and biomedical professionals with many years of practical experience in Taiwan. This study adopted a majority rule approach to validate expert opinions and determine whether an influence exists between the factors.

This study outlines the phases of the MDIP by comparing the literature review findings with actual processes and establishes a hierarchical model of the factors involved in the MDIP.

A first-phase questionnaire was designed and pre-tested to ensure its feasibility. Experts in medical device development were selected for the first-phase questionnaire of the survey in order to understand the importance and impact of the innovation process. After collecting responses from the first-phase questionnaires, numerical analysis was performed to confirm the opinions and feedback of the respondents. Based on first-phase responses, we designed a second-phase questionnaire. This analysis informed the design of a second-phase impact questionnaire, surveying 60 experienced biomedical practitioners in Taiwan, ultimately obtaining 35 valid responses to explore the relationships between various factors. The background information of questionnaire participants is shown in Table 3.

Approximately 30% of the experts who completed the valid questionnaires had more than 20 years of practical experience, demonstrating strong representativeness of and substantial influence in Taiwan’s biomedical industry. We questioned these experts on possible influences between various factors and used this as the basis to construct the ISM. This study uses the ISM to structure sustainable biomedical innovation strategies. Its hierarchical structure clarifies the causal relationships among various influencing factors, thereby enabling the development of sustainable biomedical innovation strategies.

The data collected from the second-phase questionnaire were analyzed using ISM to provide a basic model.

3.1. Development of Hierarchical Structure

ISM is a computational methodology designed to construct and understand the relationships between elements in complex situations [48]. Originally proposed by Warfield [49], ISM is particularly suitable for addressing complex engineering problems involving numerous directly or indirectly related factors [50,51].

ISM works as an interactive learning process that transforms unclear and poorly articulated mental models of systems into visible and well-defined models for different purposes [52,53]. The primary objective of ISM is to identify and establish relationships between the various elements present in a given problem [54].

ISM is widely adopted for its ability to simplify complex problems, showing versatility and effectiveness in various fields [51]. The steps are given below.

First:

The factors to be considered in the analysis need to be identified through a literature review by senior industry experts/committees. The present MDIP study identifies the process assessments.

Second:

To build the Structural Self-Interaction Matrix (SSIM), four variables to describe the relationships between the indicators need to be selected. Any of the four variables can be used for any entry in the matrix, as shown in

Table A1. Structural Self-Interaction Matrix (SSIM).

Variables	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1		O	O	O	O	O	V	O	V	O	O	O	O	O
2			O	O	O	O	O	O	A	O	O	O	O	O
3				X	O	O	X	O	O	O	O	O	O	O
4					O	O	X	O	O	O	O	O	O	O
5						V	O	O	V	X	O	O	O	O
6							O	O	O	O	O	O	O	V
7								V	O	O	O	O	O	O
8									O	O	O	O	O	O
9										X	O	O	O	O
10											O	O	O	O
11												O	O	O
12													O	O
13														X
14

(Appendix A).

V: Variable i influences variable j, but j does not influence i.

A: Variable j influences variable i, but i does not influence j.

X: Variables i and j influence each other.

O: No influence between variables i and j.

Third:

To build the reachability matrix, the SSIM is converted into a binary matrix. The symbols V, A, X, and O are replaced by 1 and 0.

Fourth:

When building the “Final Reachability Matrix (FRM)”, the FRM is checked for transitivity; this means that if element A affects element B and element B affects element C, then element A should affect element C.

In this step, the SSIM format is initially converted into an initial reachability matrix format by transforming the information in each cell of the SSIM into binary digits (i.e., 1 or 0) in the initial reachability matrix (RM), as shown in

Table A2. Reachability matrix (RM).

Variables	1	2	3	4	5	6	7	8	9	10	11	12	13	14	Driving Power
1	1	0	0	0	0	0	1	0	1	0	0	0	0	0	3
2	0	1	0	0	0	0	0	0	0	0	0	0	0	0	1
3	0	0	1	1	0	0	1	0	0	0	0	0	0	0	3
4	0	0	1	1	0	0	1	0	0	0	0	0	0	0	3
5	0	0	0	0	1	1	0	0	1	1	0	0	0	0	4
6	0	0	0	0	0	1	0	0	0	0	0	0	0	1	2
7	0	0	1	1	0	0	1	1	0	0	0	0	0	0	4
8	0	0	0	0	0	0	0	1	0	0	0	0	0	0	1
9	0	1	0	0	0	0	0	0	1	1	0	0	0	0	3
10	0	0	0	0	1	0	0	0	1	1	0	0	0	0	3
11	0	0	0	0	0	0	0	0	0	0	1	0	0	0	1
12	0	0	0	0	0	0	0	0	0	0	0	1	0	0	1
13	0	0	0	0	0	0	0	0	0	0	0	0	1	1	2
14	0	0	0	0	0	0	0	0	0	0	0	0	1	1	2
Dependence Power	1	2	3	3	2	2	4	2	4	3	1	1	2	3

(Appendix A). Thus, the FRM is obtained, as shown in

Table A3. Final reachability matrix (FRM).

Variables	1	2	3	4	5	6	7	8	9	10	11	12	13	14	Driving Power
1	1	1 *	1 *	1 *	1 *	1 *	1	1 *	1	1 *	0	0	1 *	1 *	12
2	0	1	0	0	0	0	0	0	0	0	0	0	0	0	1
3	0	0	1	1	0	0	1	1 *	0	0	0	0	0	0	4
4	0	0	1	1	0	0	1	1 *	0	0	0	0	0	0	4
5	0	1 *	0	0	1	1	0	0	1	1	0	0	1 *	1 *	7
6	0	0	0	0	0	1	0	0	0	0	0	0	1 *	1	3
7	0	0	1	1	0	0	1	1	0	0	0	0	0	0	4
8	0	0	0	0	0	0	0	1	0	0	0	0	0	0	1
9	0	1	0	0	1 *	1 *	0	0	1	1	0	0	1 *	1 *	7
10	0	1 *	0	0	1	1 *	0	0	1	1	0	0	1 *	1 *	7
11	0	0	0	0	0	0	0	0	0	0	1	0	0	0	1
12	0	0	0	0	0	0	0	0	0	0	0	1	0	0	1
13	0	0	0	0	0	0	0	0	0	0	0	0	1	1	2
14	0	0	0	0	0	0	0	0	0	0	0	0	1	1	2
Dependence Power	1	5	4	4	4	5	4	5	4	4	1	1	7	7

Note: The asterisk (*) in the table indicates an indirect influence.

(Appendix A).

Fifth:

To achieve Level Partitioning (LP), rank each row and column by comparing it to its respective type. Once the reachability matrix has been constructed, the reachability set and the prior set of an individual goal are determined. The reachability set consists of the item itself and other items it can reach, while the prior set consists of the item itself and other items it can reach. The intersection of these sets is then derived for all elements. These elements are considered top-level elements and have the same accessibility as the intersection set. To reach the next level, the top element is isolated from the other elements, and the same technique is repeated. The four levels are divided by the number of influences on the variable, with more influenced variables at lower levels. Arrows indicate that higher-level variables affect lower-level ones. The ISM of process interaction LP is shown in

Table A4. Level Partitioning (LP).

Variables	Reachability Set R(Mi)	Antecedent Set A(Ni)	Intersection Set R(Mi)∩A(Ni)	Level
1	1,	1,	1,	4
2	2,	1, 2, 5, 9, 10,	2,	1
3	3, 4, 7,	1, 3, 4, 7,	3, 4, 7,	2
4	3, 4, 7,	1, 3, 4, 7,	3, 4, 7,	2
5	5, 9, 10,	1, 5, 9, 10,	5, 9, 10,	3
6	6,	1, 5, 6, 9, 10,	6,	2
7	3, 4, 7,	1, 3, 4, 7,	3, 4, 7,	2
8	8,	1, 3, 4, 7, 8,	8,	1
9	5, 9, 10,	1, 5, 9, 10,	5, 9, 10,	3
10	5, 9, 10,	1, 5, 9, 10,	5, 9, 10,	3
11	11,	11,	11,	1
12	12,	12,	12,	1
13	13, 14,	1, 5, 6, 9, 10, 13, 14,	13, 14,	1
14	13, 14,	1, 5, 6, 9, 10, 13, 14,	13, 14,	1

(Appendix A).

Sixth:

The Conical Matrix (CM) is shown in

Table A5. Conical Matrix (CM).

Variables	2	8	11	12	13	14	3	4	6	7	5	9	10	1	Driving Power	Level
2	1	0	0	0	0	0	0	0	0	0	0	0	0	0	1	1
8	0	1	0	0	0	0	0	0	0	0	0	0	0	0	1	1
11	0	0	1	0	0	0	0	0	0	0	0	0	0	0	1	1
12	0	0	0	1	0	0	0	0	0	0	0	0	0	0	1	1
13	0	0	0	0	1	1	0	0	0	0	0	0	0	0	2	1
14	0	0	0	0	1	1	0	0	0	0	0	0	0	0	2	1
3	0	1 *	0	0	0	0	1	1	0	1	0	0	0	0	4	2
4	0	1 *	0	0	0	0	1	1	0	1	0	0	0	0	4	2
6	0	0	0	0	1 *	1	0	0	1	0	0	0	0	0	3	2
7	0	1	0	0	0	0	1	1	0	1	0	0	0	0	4	2
5	1 *	0	0	0	1 *	1 *	0	0	1	0	1	1	1	0	7	3
9	1	0	0	0	1 *	1 *	0	0	1 *	0	1 *	1	1	0	7	3
10	1 *	0	0	0	1 *	1 *	0	0	1 *	0	1	1	1	0	7	3
1	1 *	1 *	0	0	1 *	1 *	1 *	1 *	1 *	1	1 *	1	1 *	1	12	4
Dependence Power	5	5	1	1	7	7	4	4	5	4	4	4	4	1
Level	1	1	1	1	1	1	2	2	2	2	3	3	3	4

Note: The asterisk (*) in the table indicates an indirect influence.

(Appendix A). The reachability matrix is converted to a conical (lower triangular matrix) format by placing elements according to their hierarchy.

Seventh:

To build the directed graph and form the ISM, a preliminary directed graph/layer is obtained from the reachability matrix, including links. Then, the graph is transformed into ISM by replacing factor nodes with statements. The final stages of the ISM process involve transforming the Reduced Conical Matrix (RCM) into the final ISM, replacing node numbers with variable names and representing nodes in rectangular shapes. This transformation aims to minimize unnecessary edges while preserving the levels, structure, and reachability of variables, ensuring the clarity and readability of the final model.

In constructing ISM, the process begins with generating a preliminary directed graph or layer from the RCM shown in

Table A6. Reduced Conical Matrix (RCM).

Variables	2	8	11	12	13	14	3	4	6	7	5	9	10	1	Driving Power	Level
2	1	0	0	0	0	0	0	0	0	0	0	0	0	0	1	1
8	0	1	0	0	0	0	0	0	0	0	0	0	0	0	1	1
11	0	0	1	0	0	0	0	0	0	0	0	0	0	0	1	1
12	0	0	0	1	0	0	0	0	0	0	0	0	0	0	1	1
13	0	0	0	0	1	1	0	0	0	0	0	0	0	0	2	1
14	0	0	0	0	1	1	0	0	0	0	0	0	0	0	2	1
3	0	1 *	0	0	0	0	1	1	0	1	0	0	0	0	4	2
4	0	1 *	0	0	0	0	1	1	0	1	0	0	0	0	4	2
6	0	0	0	0	1 *	1	0	0	1	0	0	0	0	0	3	2
7	0	1	0	0	0	0	1	1	0	1	0	0	0	0	4	2
5	1 *	0	0	0	0	0	0	0	1	0	1	1	1	0	7	3
9	1	0	0	0	0	0	0	0	1 *	0	1 *	1	1	0	7	3
10	1 *	0	0	0	0	0	0	0	1 *	0	1	1	1	0	7	3
1	0	0	0	0	0	0	1 *	1 *	0	1	1 *	1	1 *	1	12	4
Dependence Power	5	5	1	1	7	7	4	4	5	4	4	4	4	1
Level	1	1	1	1	1	1	2	2	2	2	3	3	3	4

Note: The asterisk (*) in the table indicates an indirect influence.

(Appendix A), incorporating connectivity links between factors. Nodes and edges are derived from this graph to visually represent the relationships among factors. The directed graph is then converted into ISM by translating factor nodes into corresponding statements. This transformation provides a visual depiction of the intricate relationships among various elements, facilitating the understanding and systematic analysis of the system’s structure.

The concept of ISM is to identify the relationships among factors within complex systems and to build a structured model through the hierarchical structure and causal relationships between these factors [45]. This method is useful for obtaining valuable insights. Moreover, ISM is considered an effective method for representing the causal relationships among elements, as it provides a hierarchical structure to analytical models and plays an important role in the decision-making process [26]. It has been widely applied to complex systems across various fields. ISM uses expert opinions as development variable [50,51], and the advantage of the expert-driven approach is that the opinions of experts in the field themselves carry authority. Furthermore, experts are often few in number, and the selection criteria are not based on quantity but instead on quality [55], which is suitable for small sample sizes in an analysis.

3.2. MICMAC Analysis Applied to Classification

Matrice d’Impacts Croisés-Multiplication Appliquée á un Classification, commonly abbreviated as MICMAC, operates on the principle of matrix multiplication [39]. MICMAC analysis serves to evaluate both the driving power and dependence of various enablers within a system. The objective is to identify the key enablers that exert significant influence across different categories in the system and then clarify their roles and statuses within it [56]. In the current analysis, enablers are classified into four categories based on their levels of driving power and dependence as follows:

• Autonomous variables: These variables have weak driving power and weak dependence power. These variables are shown in Quadrant I.
• Dependent variables: This category includes enablers, which have weak drive power but strong dependence power. These enablers are shown in Quadrant II.
• Linkage variables: These have strong driving power, as well as strong dependence power. These enablers are shown in Quadrant III.
• Independent variables: These have strong driving power but weak dependence power. These enablers are shown in Quadrant IV.

Including MICMAC can compensate for the limitations of ISM. While ISM builds the structural model by typically applying a binary “yes/no” relationship judgment, it tends to overlook differences in the strength of influence between factors. MICMAC analysis performs a dynamic quantitative analysis by calculating the driving power and dependence power of each factor from the reachability matrix obtained through ISM [45].

Based on the aforementioned classification results, the factors that exert significant influence in the process of green biomedical innovation were identified, and a driving power-dependence quadrant diagram was constructed.

4. Result

4.1. Results of ISM Analysis

To understand the impact of various factors on sustainable biomedical innovation and identify those that are able to influence other factors, a sustainable biomedical innovation strategy model is proposed; this study used ISM analysis to analyze the questionnaire data and develop a structured model. The ISM result is shown in Figure 1.

In Figure 1, Clinical Needs Assessment (1) is classified as Level 4, which is the lowest level. Regulatory Feasibility Evaluation (5), Clinical Trial Execution (9), and Market Access Compliance (10) are classified as Level 3.

Prototype Concept Assessment (3), Technical Feasibility Evaluation (4), Product Technology Implementation (7), and Market Feasibility Evaluation (6) are classified as Level 2. Clinical Research Organization Cooperation (2), Production Process Design (8), Business Model Design (13), and Sales Channels Design (14) are classified as Level 1 (as shown in Figure 1).

The graph in Figure 1 shows that the Level 4 process factor is Clinical Needs Assessment (1), which affects all Level 3 process factors (Regulatory Feasibility Evaluation (5), Clinical Trial Execution (9), and Market Access Compliance (10)). The Level 3 process factors (Regulatory Feasibility Evaluation (5), Clinical Trial Execution (9), and Market Access Compliance (10)) affect each other and affect the Level 2 process factor (Market Feasibility Evaluation (6)) and the Level 1 process factor (Clinical Research Organization Cooperation (2)). The Level 2 process factors Prototype Concept Assessment (3), Technical Feasibility Evaluation (4), and Product Technology Implementation (7) affect each other and influence the Level 1 process factor (Production Process Design (8)). The Level 1 process factors (Business Model Design (13) and Sales Channels Design (14)) affect each other. The Level 2 process factor (Market Feasibility Evaluation (6) affects the Level 1 process factors (Business Model Design (13) and Sales Channels Design (14)). Supply Chain Management (11) and Healthcare Insurance Reimbursement (12), on the other hand, are not affected by the aforementioned factors. It can be observed that Clinical Needs Assessment (1) is a very significant process factor as it forms the base of the ISM hierarchy.

4.2. Results of MICMAC Analysis

The influencing factors of sustainable medical device innovation are classified based on their driving power and interdependence. The factors are divided into four quadrants: I—Autonomous Variables; II—Dependent Variables; III—Linkage Variables; IV—Independent Variables.

Subsequently, the diagram of driving power vs. dependence power for the process factors was constructed, as shown in Figure 2.

Clinical Needs Assessment (1) is classified into Quadrant IV, indicating that Clinical Needs Assessment (1) has a stable and high impact on medical device innovation and is not influenced by other factors.

Clinical Research Organization Cooperation (2), Prototype Concept Assessment (3), Technical Feasibility Evaluation (4), Market Feasibility Evaluation (6), Product Technology Implementation (7), Production Process Design (8), Supply Chain Management (11), and Healthcare Insurance Reimbursement (12) are classified into Quadrant I, indicating that they have low driving power and dependence power.

In Quadrant I, Market Feasibility Evaluation (6) has a driving power of 3 and a dependence power of 5. Therefore, in Figure 2, it is positioned in a location which corresponds to these values. Similarly, the remaining process factors are positioned corresponding to their driving and dependence powers [39,57,58].

There are no process factors in Quadrant II and Quadrant III. However, Regulatory Feasibility Evaluation (5), Clinical Trial Execution (9), and Market Access Compliance (10) are classified as being between Quadrant I and Quadrant IV on the x-axis, indicating that they have relatively high independence, are less likely to be influenced by other factors, and exhibit neutral driving power. Business Model Design (13) and Sales Channels Design (14) are located between Quadrant I and Quadrant II on the y-axis, indicating lower driving power and higher susceptibility to influence from other factors.

5. Discussion

5.1. The Critical Role and Cross-Level Influence of Clinical Needs Assessment

Clinical Needs Assessment (1) is positioned at Level 4 in the ISM model (as shown in Figure 1), characterized by high driving power that enables it to independently influence other factors. It is of great significance and is categorized as an independent variable, making it a priority consideration for achieving sustainable medical device innovation. Further observation reveals that Clinical Needs Assessment (1) not only indirectly impacts Level 2 through Level 3 but also surpasses Level 3 to directly influence Level 2 factors, such as Prototype Concept Assessment (3), Technical Feasibility Evaluation (4), and Product Technology Implementation (7). This further highlights the critical importance of Clinical Needs Assessment (1) in the process of sustainable medical device innovation.

According to the results of this study, Clinical Needs Assessment (1) influences multiple factors and should be prioritized in the MDIP. This aligns with the study of Fernández-Feito et al., who describe the device development process from conception to initial commercial exposure and emphasize Clinical Needs Assessment as a primary consideration directly impacting the outcome [4].

From the hierarchical diagram of ISM, it can be observed that Clinical Needs Assessment (1) not only directly impacts Level 3 factors, such as Regulatory Feasibility Evaluation (5), Clinical Trial Execution (9), and Market Access Compliance (10), but also cross-level influences Level 2 factors, including Prototype Concept Assessment (3), Technical Feasibility Evaluation (4), and Product Technology Implementation (7). In practice, the product design, technical feasibility, and implementation of medical devices must meet clinical needs. Without clinical needs, medical devices would lack access to or profitability in the market.

Regulatory Feasibility Evaluation (5), Clinical Trial Execution (9), and Market Access Compliance (10), positioned at Level 3, influence Market Feasibility Evaluation (6) at Level 2 and exhibit mutual interdependence. Furthermore, Regulatory Feasibility Evaluation (5), Clinical Trial Execution (9), and Market Access Compliance (10) cross-level influences to impact Clinical Research Organization Cooperation (2). The field of medical devices is highly regulated. Collaborations with CROs to jointly develop medical devices must comply with relevant regulations and clinical trial data requirements, which are essential prerequisites for these devices to enter the market.

Prototype Concept Assessment (3), Technical Feasibility Evaluation (4), and Product Technology Implementation (7), placed at Level 2, collectively influence Production Process Design (8), which is situated at Level 1. This effectively illustrates that Prototype Concept Assessment (3), Technical Feasibility Evaluation (4), and Product Technology Implementation (7) are essential considerations in the product production process, and they mutually influence one another.

Lastly, Market Feasibility Evaluation (6) is not influenced by other Level 2 factors but independently impacts Level 1 factors, such as Business Model Design (13) and Sales Channels Design (14). For business operators, the feasibility of market adoption of innovative medical devices significantly affects how business models and sales channels are shaped. Furthermore, Business Model Design (13) and Sales Channels Design (14) exhibit mutual interdependence.

The empirical results of this study show that among the factors affecting the sustainable biomedical innovation process, Clinical Needs Assessment serves as the starting point, and many other factors exhibit cross-level influence. This demonstrates that Clinical Needs Assessment is the most critical factor, and it is necessary to integrate factors with cross-level influence to collaboratively develop strategic planning. These findings are consistent with those of previous studies [55].

Medical device innovation companies are mostly small- and medium-sized enterprises (SMEs). In the highly regulated field of medical device innovation, any small impact can influence overall regulatory compliance and market entry outcomes. Therefore, formulating an effective strategy is crucial for medical device innovation. The results of this study suggest that, in addition to focusing on Clinical Needs Assessment (1), which has a broad impact on multiple factors, sustainable medical device innovation strategies should emphasize interdependence among various factors.

Surprisingly, the empirical results of this study reveal that Supply Chain Management (11) and Healthcare Insurance Reimbursement (12) are not influenced by other hierarchical factors (as shown in Figure 1). The reasons for this could be attributed to Taiwan’s unique healthcare environment and the characteristics of the questionnaire respondents. The survey was distributed to individuals such as operators of biomedical startups, key reviewers of Taiwan’s government-funded medical device development subsidy projects, and leaders of innovation projects. Their roles are primarily focused on decision-making and value evaluation in the area of medical device innovation. These two factors are not involved in the assessment of raw material procurement. The empirical study shows that in sustainable medical device innovation, supply chain management (11) is not considered during early innovation stages or government subsidy review phases. This finding aligns with the research conducted by James Moultrie et al., which suggests that medical device designers lack sufficient education in DfE [24]. Unlike other countries where medical insurance is provided by private insurance companies, Taiwan’s healthcare insurance is managed by a government agency, the Ministry of Health and Welfare, and is offered to all citizens through National Health Insurance (NHI), which is more aligned with social welfare compared to traditional insurance models. According to Taiwan’s National Health Insurance Act and the reimbursement standards for pharmaceutical items and medical devices, Taiwan’s healthcare insurance system operates under unified government guidelines, which define subsidized and non-subsidized medical services. Medical devices used in reimbursable medical procedures must pass review and approval by the National Health Insurance Administration (NHIA), a division of the Ministry of Health and Welfare. Given Taiwan’s unique healthcare insurance structure, the reimbursement process is not determined by individual insurance companies but instead by government authorities, which regulate the items and standards for reimbursement. For novel medical devices, additional approval from government agencies is required, and the review criteria vary for each device. This explains why, in the empirical study, Healthcare Insurance Reimbursement (12) remains independent of other factors, as it is exclusively governed by Taiwan’s Ministry of Health and Welfare.

In addition, in the MICMAC diagram (as shown in Figure 2), Clinical Needs Assessment (1) is placed in Quadrant IV of MICMAC, exhibiting high driving power. Supply Chain Management (11) and Healthcare Insurance Reimbursement (12) are located in the lower-left corner of Quadrant I in the MICMAC diagram, demonstrating the weakest driving power and dependency, consistent with the hierarchical structure of ISM. Prototype Concept Assessment (3), Technical Feasibility Evaluation (4), and Product Technology Implementation (7) are influenced by Clinical Needs Assessment (1). These three factors collectively impact Production Process Design (8), which, while having driving power and dependency, is not dominant and is located in Quadrant I of the MICMAC diagram.

In sustainability-oriented biomedical device innovation strategies, product design and innovation are key aspects, encompassing factors such as material selection, modular design for single-use or reusable products, green manufacturing processes, and cost-effectiveness [1]. These factors are often intertwined and mutually influential. At this time, there is no structured model for sustainable biomedical innovations that fully demonstrates the causal relationships among all influencing factors. As a result, biomedical innovation companies, working with limited resources, are unable to make the most effective investments, leading to resource waste and impacting business operations. This study found that Clinical Needs Assessment is the most influential factor in sustainable biomedical innovation, consistent with the conclusions of Jonathan G. Schwartz et al. [19]. This study shows that in sustainable biomedical innovation processes, Clinical Needs remains the most critical factor that innovation teams need to consider. There is also a need for the comprehensive consideration of cross-level influences among factors, aligning with the finding that DfE is most significant during the early stages of the design process [24]. We recommend that medical device innovators prioritize integrating sustainability considerations directly into this earliest stage of the needs assessment and evaluate environmental impacts in addition to clinical efficacy.

5.2. Interdependence and Cross-Level Linkages Among Different of Factor Levels

In addition to Clinical Needs Assessment (1), the second-level factors—Clinical Trial Execution (9), Regulatory Feasibility Evaluation (5), and Market Access Compliance (10)—not only influence each other but also exert cross-level influence. Regulatory Feasibility Evaluation (5) is an important threshold for medical devices to reach during their advancement onto the market. Although Taiwan does not yet have regulations specific to sustainable medical devices, sustainability has already been incorporated into national policy goals. Clinical Trial Execution (9) is the prerequisite for bringing innovative products to market, while Market Access Compliance (10) represents the final market entry barrier after product testing and certification. These influencing factors all involve compliance and are highly relevant to whether innovations can successfully enter the market. Therefore, biomedical innovation teams should consider these cross-level influencing factors at the early stage of innovation to understand the overall regulatory framework for biomedical innovation and sustainability.

Resources are extremely valuable during the innovation stage, and how to allocate them to achieve sustainable biomedical innovation is critical. The product design concept begins at the Clinical Needs Assessment stage, influencing other stages of biomedical innovation. P. Thamsatitdej et al. also reached a similar conclusion in their study on green supply chains [45]. Their study recommends that stakeholders in biomedical innovation, such as innovation teams or entrepreneurs, should prioritize Clinical Needs Assessment (1) while also comprehensively considering Clinical Trial Execution (9), Regulatory Feasibility Evaluation (5), and Market Access Compliance (10), in addition to all other influencing factors to achieve the goal of sustainable biomedical innovation. Furthermore, this study advises that the sustainable biomedical industry in Taiwan is still developing and that innovation teams should pay close attention to ongoing regulatory trends.

As shown in previous studies, the key influencing factors within specific industries are highly related to the characteristics of those industries [26,44,45]. This study reached a similar conclusion, indicating that Clinical Needs Assessment (1) has a guiding influence on medical device innovation. Moreover, the results of this study show that Clinical Needs Assessment (1) and regulatory restrictions are critical influencing factors that significantly impact the design of medical devices, which is consistent with the findings of Albi Thomas et al. [46].

5.3. Strategic Suggestion

The findings of this study indicate that multiple factors exhibit mutual influence, highlighting that in the highly regulated fields of medical device innovation under regulatory and clinical trials constraints, a change in one factor could lead to the risk of regulatory validation failure or the product being unable to enter the market. This study proposes a medical device innovation strategy, recommending that during the Clinical Needs Assessment stage, business operators or entrepreneurs in medical device innovation should comprehensively consider the mutual influence of all factors. A thorough understanding of the relationships between these factors can ensure the sustainability of medical device innovation and contribute to a successful and sustainable medical device innovation model.

This study proposes strategic recommendations for developing a sustainable biomedical innovation process. First, sustainability considerations should be incorporated at the early clinical assessment stage of product development. This is because Clinical Needs Assessment directly or indirectly influences other factors within the biomedical innovation process. Integrating sustainability at this stage helps ensure that subsequent stages of development embed sustainability elements throughout the process.

Furthermore, regulatory requirements are essential conditions for biomedical devices to gain market access. If a biomedical product fails to meet regulatory standards, it cannot be legally marketed or sold. The biomedical innovation process is a capital-intensive endeavor; failure to bring a product to market is an outcome that entrepreneurs and investors cannot afford to accept.

Taking Taiwan as an example, the regulatory framework for biomedical devices and the legal framework for sustainability fall under separate areas of law, with entirely different legislative purposes and logic. Therefore, this study recommends that, in developing sustainable biomedical devices, in addition to considering biomedical device regulations, teams must also take into account relevant sustainability regulations.

Based on clinical needs and regulatory constraints, biomedical innovation teams should develop and evaluate product concepts and technical feasibility while simultaneously assessing market feasibility. Such an integrated approach can substantially enhance the sustainability of biomedical innovation and increase the likelihood of success for sustainable biomedical device development.

Finally, biomedical devices are significantly influenced by regulatory restrictions and the local socio-environmental context, resulting in region-specific applicability. The experts selected for this study, as well as the relevant regulations referenced, are primarily based in Taiwan; therefore, the findings of this study are applicable only to Taiwan and may not be applicable to other countries directly. However, under conditions that are commonly shared across different regions, certain strategic recommendations may still be relevant to biomedical innovation teams or entrepreneurs in other countries. For instance, the findings of this study indicate that Clinical Needs Assessment is the most critical first step, as it affects all subsequent factors, and market demand arises only from unmet clinical needs. Moreover, while each country imposes different regulatory requirements on biomedical devices, regulatory compliance is a universal prerequisite for market entry. Therefore, based on the findings of this study, biomedical innovation teams or entrepreneurs should incorporate sustainability considerations when assessing clinical needs and conducting regulatory research in order to enhance the overall sustainability of the innovation process.

6. Conclusions, Implications, Limitations, and Future Research Agenda

6.1. Conclusions

This study utilized questionnaires completed by various stakeholders, including innovation team members and review committee members, to explore challenges in the MDIP. Problem-solving structure modeling for the MDIP, using ISM, revealed the relationships among 14 factors, presented in levels where higher numbers indicate greater influence. As shown in Figure 1, Clinical Needs Assessment influences factors at lower levels. It is significant as it aids in determining the real needs of patients during the design of successful medical devices. CROs often specialize in different medical fields, and finding the right CRO partnership is vital during the MDIP. This factor highlights the need for managers to engage in broader stakeholder ecosystems and foster interdisciplinary collaboration to enhance innovation efficiency and effectiveness.

The Clinical Needs Assessment was identified in this study as the most influential factor. This study recommends that biomedical innovation practitioners prioritize integrating sustainability considerations directly into the earliest stage of needs assessment and further consider other factors that exhibit cross-level influence, thereby developing a sustainable biomedical innovation process. By addressing the cross-level influencing factors, sustainable biomedical innovation can be promoted.

In conclusion, the ISM approach to the MDIP provides a comprehensive framework for understanding and managing the complex relationships between various factors in medical device innovation. This model serves as a valuable tool for both researchers and practitioners, offering a more nuanced view of the innovation process and guiding strategic decision-making in medical device development.

In addition, regulatory-related factors can directly or indirectly influence other factors at the subsequent level. Therefore, following the clinical needs assessment, compliance with sustainability regulations should be further considered to enhance the likelihood of market access. Finally, the product concept and technical feasibility of the biomedical device should be evaluated.

6.2. Implications

Existing studies have explored the reduction in carbon emissions during the recycling or reprocessing stages of medical devices. However, there has been no discussion on incorporating the concept of sustainability into the overall process of medical device innovation or proposing strategic solutions for this issue. This study integrates environmental sustainability into the MDIP. It also provides a strategy for medical device innovators, business operators, and entrepreneurs to enhance the sustainability of the medical device industry.

The outcomes of this study outline the key influencing factors that practitioners should consider during the medical device innovation stage through a review of the literature. Using ISM in combination with MICMAC analysis, this study achieves the following:

• Identifies the factors of medical device innovation;
• Explores the direct and indirect relationships among influencing factors;
• Evaluates the directions of influence among these factors and further proposes green medical device innovation strategies that contribute to environmental sustainability.

Medical device innovation strategies are crucial for successful innovation processes. The results of this study reveal that, in addition to focusing on the Clinical Needs Assessment (1), which has a broad impact on multiple factors, it is also essential to emphasize the interdependence among various factors. This approach can contribute to the success of MDIPs, serving as a reference for business operators and entrepreneurs.

This study fills a gap in MDIP research by identifying key factors that impact medical device development and providing a clear pathway to support the medical industry and teams or companies seeking to develop innovative medical devices. Beginning with an examination of medical device development, this research progressively explores the state of biomedical device innovation processes in the field. Through a questionnaire analysis provided to stakeholders, this study has gained further insight into the state of the medical device industry. In addition, this study helps identify potential pathways and strategies for teams or companies developing innovative medical devices.

This study integrates sustainability concepts into the MDIP and establishes a structured strategic model, supplementing the previous literature, which generally focused on either MDDPs or on reducing carbon emissions from medical devices. The strategic model proposed in this study helps innovation teams, entrepreneurs, or practitioners in the medical device field identify which influencing factors should be prioritized in the innovation process to achieve sustainable biomedical innovation processes and enable the effective utilization of resources. The results of this study provide practical suggestions for startups or entrepreneurs developing sustainable medical devices.

The implications of this study suggest that, in a sustainable biomedical innovation process, clinical needs, regulatory requirements, and product concept and technical feasibility are important aspects that need to be considered. Biomedical innovation teams or entrepreneurs should focus on these areas in order to achieve a sustainable biomedical innovation process.

6.3. Limitations and Future Research Agenda

The results of this study are based on expert opinions from Taiwan, and the model was developed accordingly; therefore, it is applicable only to Taiwan’s biomedical industry. This study is limited to Taiwan’s biomedical industry, and the proposed final model for the MDIP may not be entirely applicable to biomedical innovation processes in other countries. Different regions have varying regulations, market approval requirements, patent laws, and intellectual property rights, which necessitate adhering to local guidelines and result in differences in innovation processes. This aspect requires further discussion and improvement. Additionally, this study focuses solely on the pre-market phase of the MDIP, aiming to help teams or companies effectively plan their biomedical innovations. This focus is intended to prevent excessive time, manpower, and resource waste due to over-investment in the early stages. Biomedical innovation encompasses the development of drugs, devices, diagnostics, and other medical technologies. Although drug development is not directly related to the MDIP, the packaging and container design of drugs could benefit from the considerations outlined in the final model proposed by this study.