Strengthening healthcare supply chains: a social–ecological perspective on resilience Open Access

The healthcare supply chain (HSC) is a complex network of organisations providing healthcare products and services required for treating and managing acute or chronic diseases (Senna et al., 2023). In this context, where well-being, health, or even lives are at stake (Aldrighetti et al., 2019; Scala and Lindsay, 2021), effective HSC management is paramount. Although HSCs share similarities with other supply chains, they require tailored approaches to guarantee effective operations and continuity of care (de Vries and Huijsman, 2011). Additionally, over the next decades, HSCs will need to adapt and evolve to effectively provide healthcare services to a growing elderly population in increasingly volatile contexts (Samani et al., 2020; Senna et al., 2023), reaching new quality and robustness levels while reducing costs. Therefore, although the efficiency of the HSC is a primary goal for managers and policymakers, it is even more important to ensure the continuity of operations, so that healthcare services and products can be guaranteed for all patients in need under every condition (Almeida, 2024).

To maintain continuity, the HSC must be able to withstand disturbances, ranging from high-frequency-low-impact disturbances – such as suppliers' delays, quality issues and general services interruptions – to low-frequency-high-impact disturbances – such as pandemics, earthquakes, floods, famine and wars (Chopra et al., 2007; Ivanov, 2021; Oke and Gopalakrishnan, 2009; Tang, 2006). To ensure the continuity of healthcare operations and services, HSC resilience, interpreted as “the capability of the supply chain to provide uninterrupted treatments and healthcare products to patients in the event of a disturbance, consequently maintaining continuity of operations” (Mandal, 2017), is paramount. Yet, this interpretation of the resilience concept is limited, as it fails to recognise disturbances as an opportunity for transforming to a more desirable state than the original one (Davoudi et al., 2013; Hohenstein et al., 2015; Scala and Lindsay, 2021). Including such a transformative dimension aligns the definition of resilience to the social–ecological (S–E) perspective of resilience as discussed in Wieland and Durach (2021), where supply chain resilience is defined as the capacity of a supply chain to persist, adapt and transform in the face of change.

Resilience in the S–E sense can be achieved through initiatives, efforts and interventions – collectively referred to in this paper as resilience actions – designed to enhance the HSC's capabilities to persist, adapt and transform. These actions can be implemented either before or after a disturbance occurs (Ali et al., 2017): in the former case, the actions focus on preparing for possible future disturbances (pre-disturbances actions), whereas in the latter, they adjust processes and structures after a disturbance (post-disturbances actions) (Ivanov et al., 2017). The effectiveness of these actions is influenced by potential barriers, that is, context-specific elements that may hinder the achievement of HSC resilience. Furthermore, the concept of supply chain resilience, as viewed from the standpoint of S–E systems, can be bolstered by seven fundamental principles (Biggs et al., 2012). These were identified in the context of management and governance within intertwined systems of people and nature and were contextualised to the supply chain literature by Wieland et al. (2023).

HSC resilience is an evolving and still-developing area of research, as reflected in the literature's inconsistent use of terminology (Scala and Lindsay, 2021). Studies have primarily focused on specific healthcare segments, such as pharmaceuticals (Golan et al., 2021) and personal protective equipment (Cuvero et al., 2021), resulting in fragmentation and a lack of consensus and unanimity among scholars. Additionally, in the HSC resilience literature, barriers have received less attention compared to actions (Ganguly and Farr, 2023).

Therefore, given the relevance of HSCs, the lack of a homogeneous perspective on the topic and the interpretative power of the S–E perspective as a theoretical lens, this study aims to contextualise S–E resilience within the HSC context. Particular attention is given to S–E resilience principles, introduced in detail in Section 2.3, that can be discerned intuitively in the HSC resilience literature. The RQs guiding the research are:

The paper conducts a contextualised literature review (CLR) beginning with an initial theoretical framework (Section 2). Section 3 details the CLR protocol, followed by a descriptive analysis (Section 4) illustrating the essential research elements required to answer the research questions. Sections 5 and 6 present the findings and insights that answer the research questions. Then, the discussion and future research avenues are presented (Section 7). Finally, theoretical and managerial implications are reported, and the research conclusions are summarised (Section 8).

The HSC is a network of organisations involved in manufacturing, distribution and provision of healthcare products (i.e. drugs, medical equipment) and services (Dobrzykowski, 2019; Pinna et al., 2015; Rakovska and Stratieva, 2018; Senna et al., 2023). While sharing core principles with supply chains in manufacturing and service industries, it stands out for its unique characteristics, which often require tailored approaches (de Vries and Huijsman, 2011). The HCS is characterised by objectives beyond profit, such as treating diseases and improving health (Senna et al., 2023). Consequently, HSC has significant social impact and ethical responsibility, as its reliability and stability directly affect public health in regulated settings (Hussain et al., 2018; Furstenau et al., 2022).

HSC healthcare products and services are highly personalised and tailored to individual needs (Minvielle et al., 2014). Often, patients do not actively choose them but need them for medical necessity, making demand unpredictable and difficult to control (Dobrzykowski, 2019). Additionally, many healthcare products are costly, complex and require special handling (Rakovska and Stratieva, 2018).

Because of its critical flows, the HSC must have a very high service level, close to 100% (Aldrighetti et al., 2019), making operational continuity essential. Enhancing HSC resilience improves disturbance management, boosts performance (Senna et al., 2023) and benefits organisations by ensuring ongoing operations (Ali et al., 2017). Ignoring these factors can pose major risks to the entire HSC (Aigbogun, 2023).

Given these characteristics, the HSC can be viewed as a social–ecological system, that is, a complex adaptive system (CAS) where social and technical aspects are closely linked (Redman et al., 2004; Wieland, 2021). This interpretation challenges the traditional, engineerable perspective often applied to HSC resilience and instead advocates for a socio-ecological approach that accepts transformation.

Resilience, first studied in social sciences, describes how communities, institutions and economies respond to disturbances (Ponomarov and Holcomb, 2009). While its meaning varies across fields, resilience is often defined as an element's ability to withstand disturbances and return to stability (Bhamra et al., 2011; Ivanov et al., 2017). This aligns with engineering resilience, which aims to minimise disturbance impacts and quickly return to the previous state (Holling, 1996). The S–E view expands on this, focusing on transitions to a better post-disturbance state and viewing resilience as the ability to reorganise and seize opportunities from disturbances (Holling, 1996; Wieland and Durach, 2021).

The S–E resilience perspective is not yet common in supply chain resilience studies and is largely unexplored in the HSC resilience literature, yet it offers benefits. An engineering resilience view may oversimplify HSCs, focusing only on quick recovery to the pre-disturbance state. In contrast, the S–E perspective better suits HSCs as CASs (Redman et al., 2004; Wieland, 2021), emphasising transformation, adaptability and persistence for long-term resilience.

S–E resilience is linked to three capabilities (Wieland et al., 2023): persistence (the ability to withstand adverse events without operational interruptions (Araujo et al., 2022; Chowdhury et al., 2024)), adaptation (the ability to adjust the system in response to actual or expected changes, ensuring operational continuity during disturbances (Shweta et al., 2022; Vann Yaroson et al., 2023a)), and transformation (the ability to transform the system's structure and process more radically in response to changing conditions or disturbances (Wieland et al., 2023)). These three capabilities are not mutually exclusive and may be used sequentially or simultaneously.

In the literature, the adaptive cycle – a model describing complex systems with four phases: (1) conservation with resource buildup; (2) release following disturbances; (3) reorganisation as the supply chain adapts; and (4) exploiting the new state (Holling, 1987; Wieland, 2021) – has been used to explain the concept of S–E resilience and, consequently, the three resilience capabilities (Holling, 1986). The S–E resilience capabilities are related to different phases of the adaptive cycle. In the conservation phase, typically characterised by persistence as the system strives to maintain its core functions and structure despite emerging pressures, efficiency, stability and optimisation are prioritised through pre-disturbance actions. A disturbance may precipitate the transition from the conservation to the release phase (Wieland, 2021), in which post-disturbance action can be implemented to respond to the disturbance. Post-disturbance action can favour the transition from release to the reorganisation phase. If this transition is achieved thanks to small, gradual changes, it favours adaptation capability. More profound and radical changes, pushing the transition from release to reorganisation, are associated with transformation capabilities involving significant changes in structure, processes, or governance. Adaptation and transformation can enable the system to reach a new, more resilient state during the subsequent exploitation phase, thus initiating a new cycle of growth and development (Wieland et al., 2023) (Figure 1).

The concept of supply chain resilience, as viewed from the standpoint of S–E systems, can be bolstered by seven fundamental principles (Biggs et al., 2012; Wieland et al., 2023). The following overview highlights the main principles that inform the concept of supply chain resilience, serving as reference points for understanding its application in HSCs. The first of these highlights the importance of diversity and redundancy, which enable SCs to adapt to change. The second principle pertains to the management of connectivity, underscoring that while a well-connected SC can expedite recovery from disturbances, an excess of connectivity can amplify the propagation of crises. The third principle emphasises the importance of monitoring slow variables and providing feedback, as gradual changes can lead to irreversible transformations if not addressed promptly. Principles four and five advocate complex adaptive systems thinking and encourage learning, both of which are essential for dealing with uncertainty and improving adaptability. Finally, principles six and seven emphasise the importance of broadening participation and adopting polycentric governance, with multiple autonomous and flexible decision-making entities, to ensure effective and resilient management (see Wieland et al. (2023) for details).

The adaptive cycle has been instrumental to developing and refining the research questions addressed in this study, and informing the subsequent CLR (Durach et al., 2021), as the three capabilities, persistence, adaptation and transformation, provided a conceptual lens to distinguish how HSCs respond to disturbances and evolve towards a new, better post-disturbance state.

We opted for a CLR that helps create or improve our knowledge of “for whom,” “in what circumstances,” and “when” certain phenomena can be observed (Durach et al., 2021). The phenomenon under observation is thus the development of S–E resilience in HSCs, as influenced by specific resilience actions and hindered by contextual barriers. Furthermore, the CLR examines to what extent the principles of S–E resilience have already been applied within the HSC context to address its unique challenges. The “for whom” dimension is focused on the point of care delivery to patients. The “in what circumstances” element considers the type of disturbance. The “when” dimension distinguishes between pre-disruption and post-disruption, which maps directly onto the timing of resilience actions and into the resilience capabilities of persistence, adaptation and transformation. The CLR followed a systematic approach (Tranfield et al., 2003) and was conducted in accordance with guidelines for systematic literature reviews in supply chain management (Durach et al., 2017; Sauer and Seuring, 2023). A detailed description is provided in Figure 2.

While our review encompasses both healthcare product and service supply chains within the broader HSC context, our unit of analysis is defined more narrowly: we focus on supply chain disturbances, whether originating in the product or service domain, that directly affect the continuity of care at the point of delivery to patients. This framing reflects the strong interdependence between product and service flows in healthcare delivery and allows us to analyse resilience at the level of care provision rather than at the level of isolated supply chain segments.

Based on a discussion among the authors, the inclusion and exclusion criteria have been defined considering the research questions (content-related criteria) and the expected quality of the primary studies (quality-related criteria). The relevant literature was identified by defining the keyword used to build a search string, finding key articles that provided background information on supply chain resilience and reviewing the references used to determine a list of articles as complete as possible, keeping the number of irrelevant hits as low as possible, following general guidelines (Tranfield et al., 2003). The selected keywords included multiple synonyms to obtain a broad baseline sample and were combined using Boolean operators to construct the search string (reported in Table 1).

The inclusion/exclusion criteria, presented in Figure 2, were applied to reduce the number of papers in line with guidelines for systematic literature reviews on supply chains (Sauer and Seuring, 2023). Each author independently reviewed each title and abstract, selecting those that met the inclusion criteria. Disagreements over paper inclusion were resolved through discussion till a consensus among the authors was reached. Finally, a citation analysis was used to complement the article search and identify additional relevant works not found with the initial database search (Wohlin, 2014). Figure 2 provides a detailed illustration of the selection process and the rationale behind excluding certain papers. The complete list of references is available as supplementary material (S1).

Following the recommendations set out in the literature (Durach et al., 2017), to ensure the alignment of the analysis with the CLR's objective (Schorsch et al., 2017), the research questions were used to determine which data to extract and to define an a priori coding structure, presented in supplementary material S2. As illustrated in Figure 2, open coding, representing the initial stage of abstraction, enabled the extraction of concepts in each category in their original form. Given the diversity of the intervention mechanisms considered, integrating and synthesising the collected concepts was necessary, grouping them into subcategories (see  Appendix Tables A1, A2, and A3). Concepts that could not be aggregated with others were treated as singular actions, principles, or barriers. Finally, codes were related to one another, revealing connections and relationships among concepts. The synthesis of the codified information thus enabled greater abstraction of the framework (Durach et al., 2015), facilitating an understanding of the relevant factors influencing HSC resilience. In the selective coding phase, resilience capabilities served as an analytical framework for coding the resilience actions identified in the literature. Actions were interpreted in terms of which capability they supported. This approach allowed to link resilience actions to S–E resilience capabilities and, ultimately, to the post-disturbance state they enabled (whether previous status quo or a new, better post-disturbance state). These capabilities were not explicitly coded during the earlier rounds, as they were not directly reported in the reviewed papers, but their application in the selective coding helped contextualise the S–E resilience within the HSC context.

The descriptive analysis in this Section aims to offer an overview of the essential elements required to address the two research questions.

As reported in Table 2, the majority of the corpus aims to summarise and describe the features of the phenomenon (descriptive analysis) (47.2%). This observation suggests that HSC resilience research is currently in a transition period between the nascent and intermediate states.

Six theoretical lenses emerged from the analysis. These were used by only 8.6% of the corpus, indicating limited adoption of explicit theoretical lenses. In particular, the complex adaptive system (CAS) theory has been increasingly employed to explain supply chain resilience in volatile, uncertain and complex environments (Aigbogun, 2023). Furthermore, it supports the selection of S–E perspective for the analysis of the resilience of the HSC, as a S–E system can be regarded as a particular instance of CAS (Wieland and Durach, 2021).

The analysis of disturbances in the corpus reveals key research priorities and interests in the scientific community. Most studies focus on low-frequency-high-impact disturbances (68.5%), whereas 18.0% examine high-frequency-low-impact events – primarily equipment breakdown – and 13.5% address both. However, resilience is still primarily framed from an engineering rather than a S–E perspective.

As evidenced by the descriptive analysis, only one paper in the corpus explicitly employs the theoretical framework of S–E resilience. Consequently, there is an absence of explicit reference to the principles of S–E resilience (Biggs et al., 2012; Wieland et al., 2023). However, some elements of these principles can be intuitively discerned in the HSC resilience literature (answering RQ1). See Appendix A1 for definitions.

Recognising that there is a limit to what any single organisation can achieve without support from other stakeholders (Vanvactor, 2011), one of the most important elements referenced in the corpus is collaboration, defined in the HSC as the ability of hospitals, suppliers, manufacturers and government entities to work together effectively to anticipate, withstand and recover from disturbances (Aigbogun, 2023; Ganguly and Farr, 2023). This element (mentioned in about 34.8% of the corpus) is a component of the broaden participation and manage connectivity principles. Manage connectivity is also supported by two other element: agility – defined within the HSC as the ability to react quickly and competently to disturbances and patient requirements (Shweta et al., 2022) and mentioned in about 21.3% of the corpus – and visibility – defined as the extent to which actors within a supply chain have access to or share information which they consider key or useful to their operations (Ganguly and Farr, 2023) and mentioned in about 21.3% of the corpus. Another important resilience element is flexibility (mentioned in about 28.1% of the corpus), defined as the ability to adapt to both positive and negative impacts within the HSC by reconfiguring operations under variable conditions (Bø et al., 2023; Shweta et al., 2022). This element is a component of the foster CAS thinking principle. The manage slow variables and feedbacks principle is at the root of three other elements: agility, visibility and flexibility. Two relevant resilience elements of maintain diversity and redundancy principle can be found in HSC resilience literature. These are redundancy and robustness. The former (mentioned in about 9.0% of the corpus) is defined as the strategic incorporation of surplus capacity or resources to mitigate disturbance impacts (Zamiela et al., 2022) whereas the latter refers to the capability of the supply chain network to accommodate and deal with unexpected events endure disturbance risks (Hasani, 2021) (mentioned in about 10.1% of the corpus). Redundancy is also an element of the broaden participation principle.

These elements, and therefore the S–E resilience principles, are also linked to the resilience capabilities. Persistence capability is strongly linked with maintaining diversity and redundancy. This principle fosters persistence by acting proactively in the face of potential disturbances (Shweta et al., 2022). In the aftermath of a disturbance, foster CAS thinking and manage slow variables and feedback assume paramount importance in facilitating adaptation. These principles enable HSC to sense threats, react and adapt to changing requirements with minimal time, effort, cost and performance drop (Lima et al., 2018). Similarly, broaden participation is essential for ensuring resilience as a transformation capability, generating competitive advantages, reducing uncertainty, and improving the level of integration between SC entities (Araujo et al., 2022). Some examples taken from the paper corpus are reported in Table 3. S–E resilience principles can be used to help HSC progress on the adaptive cycle and can be used as a lever to reduce the factors (such as barriers) hindering the progression toward a more desirable HSC state.

We did not identify in the corpus any element components of the encourage learning and promote polycentric governance principles. Regarding learning, Dagenais et al. (2023) present the lessons learned from the changes observed during the pandemic in different countries. In contrast, Sawyerr and Harrison (2022) stated that there was not much evidence for lessons learnt after the pandemic. This suggests that learning in the HSC is uneven and context-dependent. In particular, the lack of elements linked to encouraging learning may be due to the fact that some of the lessons learned cannot be applied directly by HSC actors, as they fall under the responsibility of other levels (e.g. government policy) (Dagenais et al., 2023). This misalignment may prevent learning from being activated or formalised at the supply chain level, limiting its traceability in empirical contributions. Furthermore, the governance structure of the sector, which is largely centralised, constrains the autonomy of individual HSC actors and limits their ability to apply lessons learned independently, reinforcing the challenge of translating learning into practice. Regarding polycentric governance, the centralised nature of decision-making within the HSC contrasts with the decentralised governance structures, which can enable more adaptive and localised responses. For instance, Dagenais et al. (2023) report that the integration of health and social services facilities into decisional centres reinforces their lack of autonomy to plan their stocks to anticipate resource scarcity. Within the healthcare system context, the realisation of decentralised governance is inherently challenging. All actors, products and services must meet rigorous quality and safety requirements. A centralised approach ensures compliance with these requirements (Aigbogun, 2023; Mandal, 2017).

In summary, the analysis presented in Section 5, which answers RQ1, highlights how the elements intuitively discerned in the HSC resilience literature can be traced back to the principles of socio-ecological resilience and are consistently linked to the related resilience capabilities. This demonstrates that socio-ecological resilience is an applicable and useful approach for the HSC context, as it provides a solid theoretical framework for interpreting the sector's ability to persist, adapt and transform in the face of disruptions. Consequently, the following section will explain the context-specific insights that enable the achievement of HSC socio-ecological resilience.

In this section, the actions that influence HSC resilience are analysed using the S–E perspective, enriching the theoretical framework presented in the Background section, thereby answering RQ2. The resilience actions referenced in the corpus are most often implemented before a disturbance (48.1%), whereas 32.1% are implemented both pre- and post-disturbance. The remaining 19.8% focused on resilience in the post-disturbance phase (Figure 3) (see Appendix A2 for definitions). The following sections will describe how these actions affect the three resilience capabilities.

Pre-disturbance actions focus on the time frame preceding a disturbance to the HSC (Götz et al., 2024) and aim to increase its persistence. Pre-disturbance actions imply that the supply chain implements ex-ante measures to cope with disturbances, with no adaptation needed during times of change (Vann Yaroson et al., 2021). In other words, actions taken in preparation for future disturbances are expected to enable “operations as normal”, i.e. operational continuity (Ganguly and Kumar, 2019).

Indeed, most of the supply chain actions highlighted in the analysed corpus are implemented before a disturbance (Ganguly and Kumar, 2019); some examples are stockpiling medical devices, equipment and products (Ash et al., 2023; Vanvactor, 2011), implementing multiple sourcing (Rehman and Ali, 2021), developing a warning system based on technologies (Furstenau et al., 2022) and reinforcing the facilities (Hasani and Khosrojerdi, 2016). Some of these actions require a long implementation period, such as facility reinforcement or process standardisation (Bruckler et al., 2024). If these pre-disturbance actions successfully contrast a disturbance upon its occurrence, the HSC maintains the status quo and operational continuity through persistence.

Nevertheless, actions implemented in anticipation of a future disturbance cannot guarantee effectiveness, as these actions may prove inadequate in averting the adverse consequences of actual disturbances. The reasons for this are manifold within the HSC. First, managers are reluctant to justify enough investments in actions to mitigate disruptive events that have a low probability of occurrence (Ganguly and Kumar, 2019). High pre-disturbance investments are hardly compatible with healthcare organisations' historical pursuit of the lowest purchase price (Beaulieu et al., 2024). This results in implementations of actions that are not structured or pervasive enough to handle high-impact disturbances effectively. Moreover, given the scale, scope and complexity of today's HSCs, it is infeasible for HSC managers to anticipate and prepare for every potential risk. As supply chains inevitably become more global, their vulnerability also increases (Aigbogun, 2023). Even in cases where contingency plans are established, they typically remain dormant for extended periods, potentially becoming irrelevant when disturbances of low predictability and high impact occur (Bø et al., 2023). The case of pandemic preparedness plans exemplifies this: these plans were present but not updated in several countries, and they proved to be inadequate during the COVID-19 pandemic, thereby requiring the implementation of additional post-disturbance actions to maintain operational continuity (Butler, 2024; Giuffrida and Boseley, 2020; Mnyanda, 2023).

After a disturbance hits, if pre-disturbance actions are insufficient to maintain operational continuity, the HSC is compelled to implement additional post-disturbance actions (Ward and Hargaden, 2019; Vann Yaroson et al., 2021; Zaza et al., 2022). Some examples of post-disturbance resilience actions are lateral transhipment, i.e. hospitals that have major availability of drugs try to compensate for the shortcomings in other hospitals (Aldrighetti et al., 2019), backup suppliers and/or flexible distribution systems (Vann Yaroson et al., 2023b) and establishing a response team (Vanany et al., 2022). The quicker the HSC reacts to the disturbance, the faster it will return to its normal state; conversely, the HSC may encounter the risk of being unable to guarantee operational continuity (Vanany et al., 2022).

Post-disturbance actions aim to respond as quickly as possible by implementing a predefined set of responses (Tortorella et al., 2022). The term “predefined” here refers to the fact that these actions are already known to the decision-makers, in contrast to “emergent” actions, which are context-specific to the disturbance and unknown a priori, as discussed later in Section 6.2. Implementing effective predefined post-disturbance actions facilitates a return to the pre-disturbance status quo. It can thus be stipulated that such predefined post-disturbance actions increase the resilience of the HSC in terms of persistence, in a manner similar to pre-disturbance actions. Furthermore, pre-disturbance actions can enhance post-disturbance ones (Vann Yaroson et al., 2023b). For instance, strategic alliances built in preparation for future disturbances can support joint decision-making during disturbances (Vann Yaroson et al., 2021).

Predefined post-disturbance actions are sometimes ineffective due to an underestimation of the disturbance impact (Aigbogun, 2023), the selection of inappropriate actions (Doroudi et al., 2018), or the absence of effective and feasible predefined actions to counteract the specific effects of the ongoing disturbance. For these reasons, after a disturbance, HSCs explore, design and implement additional measures to enhance resilience and mitigate the disturbance's ongoing effects. If the measures are small and incremental, the system develops adaptation capabilities (Wieland et al., 2023), while more radical changes are associated with transformation (as described in Section 6.3). Adaptive actions differ from the aforementioned persistence actions, which are designed to overcome disturbances by remaining rigid and not adapting. This may be particularly true in the case of low-frequency-high-impact disturbances, where predefined actions often fail to account for the unique characteristics of the disturbance and the adverse effects it generates. In other words, to develop adaptation capability, HSCs must implement emergent post-disturbance resilience actions. These actions should not be viewed as part of a predefined set, but rather as context-specific, relatively small adjustments to the disturbance and the changes it causes (Aigbogun, 2023). Indeed, during low-frequency-high-impact disturbances, decisions must be made quickly and based on incomplete and contextual information (Furstenau et al., 2022). In this case, the HSC must adjust and modify its normal functioning in ways that cannot be predetermined.

The implementation of emergent post-disturbance actions leads to a new post-disturbance state, through incremental changes that allow the HSC to better serve the new needs that emerge after the disturbance. Therefore, this transition is determined by the HSC's ability to reorganise itself and modify its functioning to better respond to the ongoing disturbance, thus building adaptation capability. This adaptive capability is evidenced by the creation of new services in response to changing community needs (Jamal et al., 2020). For example, pre-crisis contingency plans, such as those ensuring medicine supply during shortages, require revisions due to the unexpected scale of the emergency (Jamal et al., 2020). Another example is the expansion of HSC capacity to supply the social care sector with personal protective equipment at the height of the pandemic (Scala and Lindsay, 2021).

However, supply chains, and HSC in particular, have long neglected the dynamic nature of supply chains and the need to constantly adapt and evolve in response to disturbances (Ward and Hargaden, 2019). In fact, even after the reorganisation phase has been reached, it is common for the HSC to revert to its previous status quo of conservation, abandoning the changes implemented in response to disturbance. For example, to address the shortage of health personnel in the first COVID-19 wave, hospitals recruited short-term contractors and organised rapid training for candidates. However, at the end of the first wave, managers dismissed many of these contractors. The sharp increase in the number of COVID-19 patients during the second wave necessitated the recall of those workers, but most refused to return (Dagenais et al., 2023). During shortages, HSC actors modified their standard procedures by adopting alternative forms of treatment, adjusting dosage, volume, or formulation, to ensure the continuity of care (Vann Yaroson et al., 2021). This operational shift reflects an adaptive response, whereby the HSC altered its functioning to meet emergent patient needs, rather than maintaining pre-disturbance norms. Moreover, during the Syrian war, shelters were established in hard-to-reach areas, and emergency health teams were deployed to areas of active conflict (Jamal et al., 2020). However, once the disturbance subsided, these practices were discontinued. When such actions are withdrawn and the HSC reverts to pre-disturbance practice, this reflects persistence rather than adaptation, indicating that the actions were context-specific and not integrated into long-term operations (Vann Yaroson et al., 2021). This phenomenon can be interpreted as a “bounce back” to the prevailing status quo. In other words, the achievement of a more desirable post-disturbance state is not sustained. Consequently, the concept of adaptive capability entails the HSC's capacity to modify its functioning in response to disturbances and, in some cases, consolidate incremental yet enduring changes that enhance S–E resilience. When changes are abandoned instead, this is more reflective of persistence rather than true adaptation.

In certain cases, transformative resilience is achieved, involving a radical departure from the existing system in response to a disturbance. Transformative actions permanently modify the system or processes to better cope with the post-disturbance environment. There are documented cases of success. For example, at Rady Children's Hospital in San Diego, designers created an intensive care unit with a 60-bed floor plan that can be converted into 20 fully isolated rooms for infectious disease patients (Kamin, 2022). New healthcare services were permanently introduced in Syria to mitigate the war's negative consequences to patients and the staff. These actions led to radical changes, in which disturbance serves as an opportunity for improvement and innovation, enabling the HSC to implement new procedures or services and thus evolve by adapting to the new post-disturbance environment. As HSCs respond to disturbances, they gain valuable experience that can inform their pre-disturbance actions (Harland et al., 2021). Therefore, post-disturbance actions (emergent or predefined) could become part of the pre-disturbance actions implemented by HSC to prepare for future disturbance in a new adaptive cycle. Examples of transformative resilience are limited, primarily because most papers adopt an engineering resilience perspective. Moreover, transformative actions are characterised by increased complexity in implementation, greater effort requirements and barriers that impede execution.

Barriers, often specific to certain industries, can hinder S–E resilience in the HSC, inhibiting persistence, adaptation and transformation. Only 19.0% of the corpus addresses these barriers (with each paper discussing one or more). Table 4 outlines the key barriers that affect HSC resilience (see also Appendix A3). In this section, barriers are presented in relation to adaptation and transformation capabilities, as these capabilities are connected to achieving the new post-disturbance state, enabling the exploitation of disturbances as an opportunity for improvement.

For instance, in the pharmaceutical industry, the structural and regulatory complexity of the supply chain, characterised by stringent regulations, long lead times and a limited number of suppliers, can reduce the effectiveness of certain resilience actions (Vann Yaroson et al., 2023b). Similarly, in public HSCs, a growing focus on efficiency, cost minimisation, inventory reduction and maximisation of utilisation across the entire supply chain (due to limited resources) can undermine the long-term benefits of resilience actions (Götz et al., 2024). For example, maintaining a low inventory to pursue cost-efficiency in the public healthcare system prevents it from responding to sudden increases in demand (Götz et al., 2024). As another example, in publicly funded healthcare systems, budget constraints (a lack of resources) may limit the implementation of resilience actions, such as investing in advanced forecasting tools, which are deemed too costly despite their potential to mitigate severe disturbances (Thomson et al., 2022). Due to public health budgetary austerity, several countries have closed or merged public health bodies and reduced spending on public health, including preparedness for pandemics (Thomson et al., 2022). Additionally, the presence of numerous stakeholders (fragmentation) with contrasting objectives may determine healthcare-specific barriers (Friday et al., 2021). During the COVID-19 pandemic, the fragmented application of existing risk management capabilities hindered efforts to enhance resilience against medical stockouts (Friday et al., 2021). These examples illustrate that, while HSC may implement emergent, adaptive and transformative post-disturbance actions to cope with disturbances, structural, regulatory and resource constraints can hinder the permanent adoption of these actions and the progression toward a new, better post-disturbance state, thereby reversing the HSC to its previous status quo. These make it impossible to exploit disturbances as opportunities for improvement.

Our study demonstrates that the theoretical lens of S–E resilience (Wieland and Durach, 2021) is not only applicable in the context of HSC, characterised by high dynamism and variability, but is also advantageous because it offers a theoretical framework for interpreting the relationship between the actions, barriers and principles influencing resilience and resilience capabilities. This lens has broadened the discourse on HSC by focusing on the evolution towards a better status quo, more suited to survival in the post-disturbance environment, which was previously little considered in the literature. Using this theoretical lens allows us to assess the evolution of HSC over time, evaluating how to manage and facilitate transitions from one phase to the next of the adaptive cycle by enabling resilience capabilities, thus overcoming the previous rigid classification of actions into simple pre- and post-disturbance categories, and directly relating actions to the respective enabled capabilities. In addressing RQ1, the principles of S–E resilience have been associated with the capabilities they enable (see Table 3), thereby broadening the interpretation of S–E resilience provided in the literature to date. This analysis highlights promising areas for future research, particularly those that investigate the previously unexplored relationship between principles and capabilities. Applying the S–E resilience lens to HSC has enabled a better characterisation of the barriers that limit transformation and adaptation (RQ2), thereby hindering progress towards a better status quo and continuous improvement. The previous engineering perspective focused solely on responding to the acute phase of disturbances, without considering how to leverage disturbances as opportunities for improvement. S–E resilience allowed us to highlight this gap and propose an initial identification of these barriers. This extends the theoretical lens of socio-ecological resilience, highlighting that if barriers are not addressed, progression along the adaptive cycle can be blocked. Vice versa, our study showed that contextualising S–E resilience is instrumental to operationalising these barriers. Theorising HSC as a socio-ecological system enables us to explain the discrepancies and conflicting opinions identified in the literature, thereby clarifying its resilience. For example, this approach can help overcome the fragmented definitions of HSC resilience found in the literature (Araujo et al., 2022), which range from an emphasis on rapid response to disturbances (Bastani et al., 2021) to a focus on long-term growth (Scala and Lindsay, 2021). Furthermore, discrepancies are evident in how HSC resilience is measured, as there is currently no consistent, universally accepted method (Goodarzian et al., 2022; Scala and Lindsay, 2021).

The literature review results were formalised in the theoretical framework presented in Figure 4.

The framework describes the evolution of an HSC facing disturbances. Before a disturbance, the HSC can apply pre-disturbance actions as a mitigation lever promoting persistence. When a disturbance occurs, predefined post-disturbance actions are implemented to restore the previous status quo. In contrast, unexpected impacts require emergent post-disturbance actions, potentially leading to adaptation and transformation towards a new post-disturbance state. Adaptation and transformation-hindering factors can impede the transition to exploitation, returning the HSC to the previous status quo. Overcoming these hindering factors enables entry into the exploitation phase, thereby completing the transformation to a new, resilient state. Developing persistence, adaptation and transformation capabilities requires applying resilience principles progressively.

Furthermore, this perspective opens several avenues for future research, as detailed below. The first research avenue entails considering the HSC as a socio-ecological system, as we proposed here with our reframing of the existing literature. Following our example, researchers could apply the S–E system theory lens (Wieland and Durach, 2021) to study the HSC not as a static system, but as a continually evolving one, thus requiring it to be observed at different points in time to grasp its true nature. The same approach can also be used in other sectors. Taking the S–E system theory as a theoretical lens, the second research avenue proposes studying how different resilience actions can enable S–E resilience. Our study proposes that in the context of HSC, resilience research should distinguish between three types of actions: pre-disturbance, predefined post-disturbance and emergent post-disturbance. Each type can foster distinct resilience capabilities, and future research should investigate which specific actions activate this effect and how this mechanism unfolds. Overcoming the previous categorisation into proactive and reactive strategies (Ganguly and Kumar, 2019; Vann Yaroson et al., 2023b), these future research avenues can leverage SC resilience theory and the resource-based view to identify and characterise the actions that build persistence capability. Instead, dynamic capabilities would be a suitable theoretical framework for actions that support adaptation capability.

Our literature review highlighted that certain barriers to HSC resilience have been identified in previous studies. Nevertheless, we introduced a novel conceptualisation of these barriers as hindrances to the adaptation and transformation of the supply chain. In addition, barriers might be only one of the factors hindering the HSC transition towards a more resilient status. Therefore, the third research avenue focuses on identifying the factors hindering HSC adaptation and transformation, particularly the lack of formalisation of lessons learned and the explanation of underlying mechanisms. The SC resilience theory can effectively serve this exploration.

Finally, the last future research avenue builds on our novel conceptualisation of the HSC resilience capabilities as cumulative; thus, future research should embrace this cumulative nature to understand how and under which conditions it is possible to reach full resilience. The cumulative capabilities concept, dynamic capabilities and contingency theories can help frame this future research avenue. Furthermore, future research may investigate the still-unexplored relationships between resilience principles and capabilities.

This research has validated the applicability of the S–E resilience framework to the HSC context. S–E resilience principles were contextualised for the HSC context (RQ1), demonstrating that the S–E resilience theory is applicable and useful for studying resilience in HSC. Contextualising the theory helped identify additional elements that had not been theorised before and operationalise them, thanks to a narrower approach. From this perspective, this study identified actions that promote and barriers to the development of persistence, adaptation and transformation capabilities (RQ2). Reading the literature on HSC through this lens enables us to understand how to help HSC progress towards a better post-disturbance state, broadening the discourse on HSC resilience. Therefore, this study offers a conceptual foundation for analysing HSC resilience through an S–E lens, and lays the groundwork for future studies that aim to empirically validate these categories and refine the conceptual model.

HSC managers should shift their focus from maintaining short-term operational continuity to actively integrating adaptation and transformation into resilience actions. This approach ensures that the HSC not only survives disruptions but also evolves into a better state. To achieve this, managers need to identify and overcome sector-specific barriers that lead to a predominant focus on engineering resilience, thereby overlooking opportunities to turn disturbances into opportunities for improvement. Employing a S–E perspective, both before and after disruptions, can help managers determine which resilience capabilities are underdeveloped and need targeted actions.

The study's findings demonstrate that fostering S–E resilience in HSCs directly benefits patients, healthcare professionals and the broader community. Investing in adaptive and transformative resilience in HSCs is essential because it promotes an evolution towards better and more equitable conditions. HSCs with this capacity not only ensure continuity of care and rapid responses to the population's needs during disturbances but also reduce inequalities and strengthen patient confidence through structural and innovative changes. Communities benefit from more equitable access to care, while healthcare professionals can rely on improved resources and procedures to manage and work effectively during disruptions. The lack of this form of resilience can lead to delayed responses, the maintenance of existing inequalities and a loss of confidence, thus compromising the quality and equity of healthcare.

This paper has some limitations. The search query excludes all studies that discuss components of the resilience concept (e.g. response, recovery, preparedness). Moreover, an additional limitation pertains to the HSC under consideration. The present study draws on insights into the HSC in its broader sense, with particular emphasis on care delivery. This may have led to the exclusion of papers that referenced specific sub-components, such as the blood supply chain, pharmaceuticals and hospitals. Ultimately, the query contains terms related to actions, barriers and principles. It is essential to note that alternative keywords for these terms may not be included in the search string. All these search limitations need to be addressed in future research. Furthermore, differences in governance and reimbursement models across countries have not been considered at this stage, as the analysis was primarily concerned with physical flows. Including government agencies, policymakers and insurance companies in the analyses could enhance the understanding of the whole healthcare ecosystem and facilitate the identification of additional actions, principles and barriers.

The supplementary material for this article can be found online.