Corps de l’article

Over the last decades, the health sector has faced significant changes in terms of costs and resources and technological advances (Adams, 2016; Chahal et al., 2019; Haschar-Noé and Basson, 2019). In France, this sector is based on multiple structures and characterized by the involvement of various medical, paramedical, technical, administrative, and social actors—and it is considered a landscape for entrepreneurship and innovation (Gaudron, 2020; Nobre, 2013). The French health sector is under intense pressure to change its management. Hospitals, for example, have been subject to new reforms to improve their operation. The HPST[1] law of 2009 clearly defines a new health and medico-social organization. This law invites the regional health agencies to support local actors’ mobilization capacities, recognize their competencies, and sign local health contracts (LHC), ensuring the articulation between regional projects and local action (Honta and Basson, 2017). This sector is also characterized by a hybridity of organizations and professions (Gallouj et al., 2015). Similarly, the boundaries between the private and public sectors are gradually disappearing (Emery and Giauque, 2014), and public-private partnerships and the pooling of resources are developing to promote technological and social innovations and prevent individual risks (Haschar-Noé and Basson, 2019).

Furthermore, the COVID-19 pandemic has brought about a significant transformation, as changes emerge with the rise of innovations for contactless services and operational processes to improve organizational agility (Corond et al., 2020).

With the increase in innovations, healthcare organizations such as hospitals, clinics, and nursing homes have undergone transformations and radical changes at both the medical and organizational levels. The medical shift concerns innovation in products and technologies and medical and therapeutic innovation that translates into treatments for better care or to deal with novel diseases (Nobre, 2013). Furthermore, entrepreneurial orientation (EO), as reflected in innovativeness, proactiveness, and risk-taking, seems particularly important in these organizations. It enables them to deal with the rapid development of technologies and users’ needs for a higher quality of care and service. EO is also a way to improve patients’ healthcare, raising innovation and transforming these organizations. From an organizational perspective, it provides managers with tools to strengthen the use of their resources and capabilities (Lages et al., 2017).

Despite developing an entrepreneurial culture in healthcare organizations at the managerial and organizational levels (Simonet, 2013), few studies explain the factors that promote EO (Paula Monteiro et al., 2019). To bridge this gap, this study investigates the role of dynamic capabilities (DCs), expressed through information acquisition (IA), strategic flexibility (SF), and structural flexibility (SX), in promoting EO in the French healthcare sector.

DCs are defined as “the ability of an organization to integrate, build and reconfigure internal and external competencies to cope rapidly with changes in the environment” (David et al., 1997, p. 516). Scholars have demonstrated that DCs are associated with entrepreneurial behavior by facilitating a sense of opportunities and threats, seizing opportunities, and transforming them to maintain competitiveness (Teece, 2007; 2014). This dynamic is strongly present in healthcare organizations undergoing evolution on all levels, prompting managers to rethink their operations by restructuring different services to address internal and external pressures.

Therefore, we aim to analyze the impact of these DCs on the development of EO based on a quantitative study carried out with 133 managers of healthcare organizations in France. We test our model using structural equation modeling (SEM) and fuzzy set qualitative comparative analysis (fsQCA). The results provide interesting insights and offer three main contributions to the literature. Our first contribution is enriching the DC and EO literature by empirically confirming the significant role of these capabilities in developing entrepreneurial activities.

Second, using DC theory enables us to better understand how entrepreneurial capabilities such as innovativeness, proactiveness, and risk-taking manifest in organizations by configuring their internal and external competencies. This study contributes to the debate on EO antecedents by linking the three dynamic capabilities (IA, SF, and SX), which have been gaining interest in the management and entrepreneurship literature.

Third, the findings contribute to the emerging field of EO research related to healthcare organizations (Chahal et al., 2019, Lages et al., 2017, Miller and French, 2016). Prior work addressing the antecedents of EO in the service sector has been scant in general but more specifically in the healthcare sector (e.g., Martens et al., 2016; Vecchiarini and Mussolino, 2013). This study contributes to this stream of research by proposing DCs as critical determinants of EO. Through its findings, the study offers managerial implications for healthcare managers, who are fundamentally concerned about the reinvention of their organizations in a dynamic environment.

The rest of this article is organized as follows. Section 2 outlines the relevant arguments on the proposed relationships among IA, SF, SX, and EO to develop the hypotheses. Section 3 presents the methodology and research design, while Section 4 discusses the proposed model’s results. Finally, Section 5 highlights the contributions, limitations, and directions for future research.

Theory and hypotheses

The focal point of the study is to investigate the role of IA, SF, and SX in developing EO in healthcare organizations. This section discusses the theoretical framework and hypotheses.

Figure 1

Conceptual Model

Conceptual Model

-> Voir la liste des figures

Entrepreneurial orientation (EO)

EO refers to the organization’s strategic position toward entrepreneurial behavior at the individual and organizational levels (Wales et al., 2020). The EO literature has significantly advanced during the past decade regarding theoretical development and outcomes (Covin and Wales, 2019; Ferreira et al., 2019). Notably, its conceptualization has been the subject of several studies (Anderson et al., 2015; Covin and Wales, 2018). Since the work of Covin and Slevin (1989), several conceptualizations have emerged. Lumpkin and Dess (1996) defined EO using five dimensions: innovativeness, proactiveness, risk-taking, autonomy, and competitive aggressiveness. The past decade was also characterized by the development of this conceptualization work (Anderson et al., 2015;). Anderson et al. (2015) depicted it in terms of two dimensions: (1) entrepreneurial behaviors, which are mainly based on innovativeness and proactiveness, and (2) managerial attitude toward risk, which focuses on risk-taking.

Despite these diverse conceptualizations, there is a near consensus on the level of the three dimensions forming EO: (1) innovativeness is defined as the efforts made by an organization in terms of pursuing new products, processes, or business models; (2) proactiveness entails how an organization reacts to different trends in its environment by introducing new methods and technologies, etc. (3), and risk-taking is the willingness to take advantage of opportunities with a reasonably high chance of failure (Covin and Slovin, 1989, Miller, 1983)

Furthermore, some recent studies have focused on better understanding the factors that promote EO (e.g., Eshima and Anderson, 2016; Paula Monteiro et al., 2019). Scholars argue that internal and external factors may influence EO. It has been argued that EO is internally facilitated by the firm’s structure (Green et al., 2008) and the firm’s efforts to generate new alternatives and frameworks that promote strategic adjustments (Gao et al., 2018). EO may be externally influenced by the environment’s complexity and dynamism (Wales et al., 2013), political and business ties, and social capital (García-Villaverde et al., 2018).

In this part, we present our hypotheses. First, we discuss the influence of information acquisition on entrepreneurial orientation.

Information acquisition (IA) and entrepreneurial orientation (EO)

IA is the organization’s capability to collect and assimilate valuable information from internal and external sources. Internally, organizations collect and incorporate helpful information from reports, databases, internal analyses, and managers’ previous experiences (Santos-Vijande et al., 2012). Externally, organizations search for information to anticipate stakeholders’ needs, new tendencies, and potential opportunities (Parra-Requena et al., 2015). IA is part of scanning the environment and monitoring its activities (Keh et al., 2007).

Studies have shown that information, more specifically, the acquisition of new information, is critical in identifying entrepreneurial opportunities. The acquisition of information can take many forms, such as the discovery of new technology to better meet the user’s needs, new work reforms, or new sources of funding (e.g., Rezazadeh and Nobari, 2018). Nadkarni and Barr (2008) reported that managers interpret their dynamic environment to make sense of events and changes and are open to new information—enabling them to identify and pursue opportunities and new ways of performing activities. IA is vital to developing the entrepreneurial mindset’s first steps and is especially important in determining opportunities. Accordingly, studies assert that organizations with effective mechanisms to assimilate external resources and integrate them into the decision-making process are more favorable in innovative activities (Crescenzi and Gagliardi, 2008).

Similarly, in the entrepreneurial literature, information is associated with entrepreneurial alertness. Tang et al. define alertness as “consisting of three distinct elements: scanning and searching for information, connecting previously disparate information, and making evaluations on the existence of profitable business opportunities” (Tang et al., 2012, p. 77).

From a dynamic capability perspective, IA facilitates the early detection of fundamental changes, thus providing managers with additional time to react to external pressures (Teece and Leh, 2016).

In the health sector, IA facilitates the understanding of changes that affect the governance of health institutions, their organizational level, and their professions (Tissioui et al., 2016). The objective is to develop user-centered innovation models where patients and health professionals are at the center of the experimentation. This requires new ways of accessing data (Kletz and Marcellin, 2019).

IA facilitates this process by providing signals for managers to adapt their decision to respond to patients’ and users’ needs. This adaptation process is ensured not only through the firm’s internal resources but also through its ability to acquire information from different sources of critical stakeholders.

More importantly, Teece (2014) saw the DC framework as an entrepreneurial approach that includes (1) the internal facet related to the strategy and the critical resources and (2) the external facet that organizes the firm–external partners’ relationship. It focuses on information and knowledge acquisition to facilitate the discovery, evaluation, and exploitation of opportunities. Therefore, organizations with a high level of IA have a solid potential to attract and exploit the external resources inherent in their networks to capitalize on emerging opportunities and develop their EO.

In light of these arguments, the following is posited:

H1: Information acquisition has a positive impact on entrepreneurial orientation.

In the following, we analyze the mediating effect of strategic flexibility on the IA‑EO relationship. This part will be divided into three subhypotheses: (H2a), which addresses the impact of IA on SF; (H2b), which concerns the effect of SF on EO; and (H2c), which focuses on the mediating role of SF.

Mediating effects of strategic flexibility (SF)

SF refers to the organization’s ability to manage environmental risks by adopting a reactive or proactive stance regarding environmental threats and opportunities (Grewal and Tansuhaj, 2001). Scholars assess that SF is vital for organizational adaptation because it helps managers deal with internal and external changes (Brosovic, 2018). It facilitates the transformation of strategic choices by helping the organization respond to change (Sanchez, 1995). At this level, Volberda (1996) distinguishes two types of SF. Internal SF is the management’s capability to adapt the organization to new environmental conditions. External SF is defined as the organization’s ability to influence the environment through proposing new technologies or new products and services but also through anticipation.

Regarding its antecedents, previous studies arising from different disciplines have identified a set of factors affecting SF. These can be linked to two types of factors: the strategic orientation of an organization, which can be associated with learning orientation (Santos-Vijande et al., 2012), absorptive capacity (Volberda et al. 2010), knowledge acquisition (Miroshnychenko et al., 2020), and organizational design (Herhausen et al., 2021). It is important to note that information promotes this dynamic capacity by allowing managers to have a more accurate picture of the functioning of the organization and by adapting their management to the situation.

H2a: Information acquisition has a positive impact on strategic flexibility.

Thus, in a very dynamic context, a high level of flexibility favors entrepreneurial behaviors that make anticipating changes and meeting the expectations of various internal and external stakeholders possible. Li et al. (2008) argue that entrepreneurial organizations, through SF, are more oriented toward innovative practices than traditional ones. Therefore, this capability facilitates the implementation of innovative strategic initiatives and innovation in terms of products and services (Fan et al., 2013)

Developing this capability within an organization aims to identify changes and propose suitable answers by deploying and configuring resources and developing knowledge and skills. Patel et al. (2012) state that flexible organizations are more oriented toward understanding uncertainty factors to integrate them into decision-making. Similarly, SF enables organizations to enhance the repertoire of managerial capabilities and grants them the speed to mobilize them.

H2b: Strategic flexibility has a positive impact on entrepreneurial orientation.

SF is also facilitated through other dynamic capabilities necessary for its development within the organization and for its survival. Its interdependence with other variables has been explained well in the work of Miroshnychenko et al. (2020), who highlight the impact of absorptive capability on its development. They assert that flexibility is impacted primarily by the ability to acquire and use computerized news. Flexible organizations can proactively identify environmental changes by improving their IA activities. Analyzing their environment, technological changes, and relationships between stakeholders are always helpful because they provide an essential knowledge source (Roy and Thérin, 2008). The information and knowledge-gathering process allows organizations to anticipate market tendencies and renew strategic responses in light of new opportunities (Santos-Vijande et al., 2012).

Thus, entrepreneurial practices depend on the organization’s emphasis on the innovative strategy-making process (Anderson et al., 2015). EO can partly explain the strategic actions that permit adjustments to environmental changes. Furthermore, this strategic orientation can be used to respond to restructuring internal resources to ensure a better match with external changes. Resource-based view (RBV) proponents suggest that knowledge-based resources help organizations achieve sustainable competitive advantage (Dai and Si, 2019). These resources may facilitate organizations’ internal entrepreneurial activities through the acquisition and sharing processes. Additionally, Wiklund and Shepherd (2005) argue that EO depends on combinations of internal resources with environmental changes. SF is recommended to ensure the fit between two constraints: resources and change.

Furthermore, a growing body of literature considers entrepreneurial behavior a product of management practices’ capabilities. The methods that promote this view include the firm’s strategic management (Barringer and Bluedorn, 1999), and learning (Sirén et al., 2016). According to Lumpkin and Dess (1996), EO is analogous to Stevenson and Jarillo’s (1990) concept of entrepreneurial management. Similarly, Anderson et al. (2015) state that EO relates to entrepreneurial behaviors and managerial attitudes.

In sum, entrepreneurial organizations are defined through decision-making practices oriented toward pursuing opportunities with uncertain outcomes, creating a culture of innovation, and implementing an offensive strategy.

In light of these arguments, the following is posited:

H2c: Strategic flexibility positively mediates the information acquisition–entrepreneurial orientation relationship.

Finally, we analyze the moderating effect of structural flexibility (SX) on the SF-EO relationship.

The moderating effect of structural flexibility (SX)

SX is the ability of an organization to adapt its organizational structure and its decision and communication processes to new conditions and to reconfigure different resources to perform better further actions (Anser et al., 2020). It is considered a dynamic capability that enables organizations to reconfigure their structure and their communication processes (Volberda, 1998) to adapt decisions to environmental change.

SX also facilitates resilience to change, especially in sectors characterized by strong dynamism, such as healthcare. Healthcare organizations require a certain degree of flexibility to adapt to patients’ and customers’ expectations and respond promptly to their needs (Brozovic et al., 2016). Accordingly, SX facilitates introducing organizational changes in response to new circumstances, such as crises or competitive pressure (Huber, 1990). Since 2010, the health sector in France has seen many collaborative spaces aimed at encouraging innovation, whether in hospitals or other healthcare structures. The living lab, which aims to strengthen the partnership between private and public actors (companies, universities, users, etc.), is the most impactful space that encourages open innovation. It is a space that allows products and services to be tested in real conditions (Le Chaffotec, 2016).

The COVID-19 crisis has also demonstrated the importance of SX within these organizations through more dynamic coordination between different healthcare providers (Albert-Cromarias and Dos Santos, 2020, Defrancq et al., 2020). Several initiatives have been taken to improve interorganizational communication and to encourage collaborations between various partners, such as hospitals and other healthcare organizations. Similarly, the development of flexibility also requires the training of multifunctional teams and the adaptation of control systems to meet different needs. This phenomenon was widely observed in healthcare facilities during the crisis, where managers were forced to rearrange activity flows to ensure resilience.

Previous academic works explain that the organizational structure impacts information flows and the context and nature of interorganizational relationships. The organizational structure specifies the coordination mode, divides powers and responsibilities, and prescribes levels of formality and complexity. Miller and Friesen (1982) note that the structure and decision-making process ultimately have the same goals: to enact control, achieve predictability, and extend collective cognitive abilities. Therefore, these factors enable managers to handle uncertainty and understand environmental complexity.

Entrepreneurial organizations ensure internal coherence between their structure and decision-making processes. On the one hand, the implementation of entrepreneurial actions depends on the strategic behavior of an organization, its strategic choices, and the manager’s EO (Anderson et al., 2015). On the other hand, it depends on the organizational structure (e.g., Green et al., 2008; Wales et al., 2013), which is considered a key determinant of entrepreneurial behavior, such as innovation and proactiveness (Green et al., 2008). Nahm et al. (2003) argue that the organization’s structure can facilitate or prevent the implementation of innovations. An organic design enhances product innovation development, while a mechanistic structure enables the execution of product innovation.

Green et al. (2008) also discussed the effects of alignment in structure and decision-making processes on the ability to handle changes. They confirmed that this internal coherence enhances the organization’s response capability by recognizing relevant signals and information. Subsequently, the organizational structure contributes to the systemic discovery of innovative opportunities through facilitation and motivation (Pittino et al., 2016). Thus, SX, as a valuable organizational capability, may facilitate responding to new environmental circumstances, as organizations can evolve their structures by proposing, for example, new structural configurations. Consequently, the potential impact of SF on EO can be powerful in organizations with a high degree of SX.

H3. Structural flexibility moderates the strategic flexibility–entrepreneurial orientation relationship.

Methods

Sample

To test the hypotheses, the study examines empirical data from France’s healthcare sector, particularly healthcare organizations, such as hospitals, specialized clinics, and nursing homes. This sector has undergone profound changes for several years, including regulatory, social (aging population, chronic diseases, etc.), financial difficulties, and privatization (Noguera and Lartigau, 2009). To address these dynamic conditions, public and private actors must adapt and anticipate users’ needs and innovate in terms of the quality of service and care. At this level, collaborations between hospitals and private organizations have emerged to facilitate the exchange of knowledge and experiences.

This sector provides an appropriate setting to test the relationships between dynamic capabilities (IA, SF, and SX) and EO. In France, the legislative and regulatory context of health policy has experienced changes aiming to modernize and promote innovation by diversifying services and funding sources. It is also characterized by various actors, such as the state and private nonprofit and for-profit commercial actors and associations. Subsequently, this presents conflicting perspectives, along with promising opportunities, with various constraints relating to financial resources, governance, and the size of operators. Similarly, the sector is denoted by the diversity of the proposed activities, such as professions of care, medical-technical, research and innovation, and hygiene quality. Thus, dynamic capabilities are needed to address the complexity and high degree of change (Blanken, 2008; Rechel et al., 2009).

A questionnaire was prepared following an exhaustive literature review and was pretested with ten managers to eliminate statistical and semantic issues (Malhotra et al., 1996). Potential respondents were located through databases and professional and personal networks. Consequently, we compiled a database of 981 potential respondents and contacted them by email.

The initial email included a brief presentation explaining the purpose of the research and a detailed questionnaire incorporating all the variables in the study. The second email was a reminder, urging respondents to complete and return the questionnaire. Of the 146 respondents who returned the questionnaire, only 133 could be used for analysis due to missing data. The following characteristics were observed: 60% were women, 40% were men, and the age ranged between 45 and 55 years; the average experience in the field was ten years; and they occupied positions such as directors of the health structure (53%), administrative managers (25%), and heads of medical service (22%). Their operations were principally related to managing i) nursing homes, ii) specialized clinics, and iii) hospitals in France.

Common method bias

Since the data were collected from a single sample, we adopted specific practices to address common method bias. The recommendations of Podsakoff et al. (2003) were used to ensure confidentiality and protection of personal data. The questionnaire was divided into several sections based on the variables used in our model (e.g., IA, SF, SX, EO), separated into small paragraphs. The questionnaire was pretested with managers and researchers to ensure its clarity. We used Harman’s one-factor test to detect common method bias (Podsakoff et al., 2003). We conducted a factorial analysis of all the items used in this analysis. Six factors emerged from this analysis, with an eigenvalue more significant than one and a cumulative variance of 70.7%.

We employed the approach of Cote and Buckley (1987), which consists of comparing several models. In Model 1, all IA, SF, SX, and EO items were loaded on a single model. In Model 2, all items were assigned to their respective construct. Model 3 employed a common latent variable linking all the dimensions used in Model 2.

The findings indicate that Model 1 (χ2/df= 2.389, NFI= 0.69, CFI= 0.79, RMSEA=0.1) and Model 2 (χ2/df=1.436, NFI=0.81, CFI=0.93, RMSEA=0.0057) have a better quality of fit than Model 3 (χ2/df=4.66, NFI=0.37, CFI=0.418, RMSEA=0.16). The result confirms that common method bias is not a concern in this study.

Construct measurement and validation

All variables in the proposed model were measured using reflective indicators (Appendix I provides a complete list of these items). The SF includes five items adapted from scales by Tsai and Shih (2004), and Vorhies and Morgan (2003). We asked respondents to rate their organizations’ levels of expertise in rapidly responding to changes and detecting new opportunities and threats. All items in this construct were measured using a seven-point scale. The reliability was examined using Cronbach’s alpha value (α=0.82).

The SX construct was measured using items related to communication between departments and different services, decreasing bureaucracy, and structure flexibility through decentralized decision-making, flexibility in work, communication, and human resources management. The three items were adapted based on Rudd et al. (2008). This scale indicated a Cronbach’s alpha coefficient of 0.74.

Additionally, five items were employed for the IA construct to measure firms’ efforts regarding internal and external IA as adapted from the scale developed by López-Sánchez et al. (2011). Respondents rated their level of agreement on a seven-point Likert-type scale (1= strongly disagree; 7= strongly agree). The Cronbach’s alpha value (α=0.86) indicated good reliability.

Finally, we measured the EO construct through nine items adapted from previous studies (Covin and Wales, 2012; Hughes and Morgan, 2007) on a seven-point scale (1=strongly disagree; 7=strongly agree). Next, we performed exploratory factor analysis (EFA) and examined the psychometric proprieties of each dimension using Cronbach’s alpha (Innov= 0.82; Proac= 0.80 and Risk= 0.83). Table 1 presents the descriptive statistics of the variables.

Table 1

Descriptive statistics

Descriptive statistics

-> Voir la liste des tableaux

Statistical techniques

Hypotheses are tested using structural equation modeling (SEM) and fuzzy set qualitative comparative analysis (fsQCA). SEM is a variable-oriented technique that focuses on the net effect of the independent variable on the dependent variable (testing the significance of relationships between variables). It treats the independent variables as competing to explain variation in the dependent variables and relies on the principles of additive effects, linearity, and unifinality (Woodside, 2013). It was conducted to test the causal path’s potential, such as the direct, mediating, and moderating effects of the variables under investigation. However, fsQCA was utilized to provide an in-depth understanding of antecedents’ complex, nonlinear, and synergistic effects (e.g., IA, SF, and SX on EO). fsQCA is a case-based technique focusing on configurational products (Ragin, 2008). It considers multiple configurations and supposes that an outcome is rarely the result of a single cause, that these causes are rarely separate and that a specific cause can have effects of opposite sign (e.g., negative or positive), depending on the context. It assumes asymmetry between independent and dependent variables and equifinality, in which multiple paths and solutions lead to the same outcome. fsQCA also allows for multifinality, in which identical conditions can lead to different results (e.g., a configuration between SF, SX, and IA can generate a high level of EO or a low level of EO).

Results and analysis

This section discusses the measurement-testing results, followed by the model-testing results.

Measurement testing

The statistical software AMOS 20 was employed, and the maximum likelihood (ML) estimation method was used to avoid normality issues. The model fit was assessed using chi-square (χ 2= 294.938), comparative fit index (CFI=0.938), incremental fix index (IFI=0.939), Tucker-Lewis’s index (TLI=0.928), and root mean square error of approximation (RMSEA=0.05) tests, which suggest that the model fits the data. Furthermore, we used confirmatory factor analysis (CFA) with an ML estimation procedure to examine reliability and validity. The first set of the model included IA, SF, and SX. The second set comprised the three dimensions of EO: innovativeness, proactiveness, and risk-taking.

Item IA5, “The members of the organization use informal means to determine about recent events regarding the market, or the environment,” was dropped following a preliminary analysis because its loading on the IA construct was very low (0.12), negatively impacting the quality of the model. After its elimination, χ 2 declined from 350.3 to 258.5.

Subsequently, all variables were assessed for reliability and convergent and discriminant validity (Anderson and Gerbing, 1988). Reliability was evaluated by Cronbach’s alpha, which all satisfied the recommended threshold of 0.7 (Cronbach, 1951). Regarding composite reliability (Fornell and Larcker, 1981), all variables’ coefficients range from 0.78 to 0.89, suggesting satisfactory internal consistency. Validity was assessed using average variance extracted (AVE), in which a value greater than 0.5 confirms validity (Table 2). Additionally, Fornell and Larcker’s (1981) approach mandates that the square root of the AVE for every pair of variables must exceed the correlations between the latent variables.

Table 2

Construct measurement summary—Confirmatory factor analysis and scale reliability

Construct measurement summary—Confirmatory factor analysis and scale reliability

-> Voir la liste des tableaux

Structural model and result

The SEM results indicate that the structural model (Figure 2) is a good representation of the data collected (χ 2 = 274.605, IFI = 0.925; TLI = 0.912, CFI = 0.923). Table 3 synthesizes the structural model based on AMOS by focusing on the standardized path coefficients (β) and their significance (t values).

Figure 2

Structural model

Structural model

-> Voir la liste des figures

The results (t=2.351, p<0.05) support H1, which supposed that IA positively impacts EO. The results also support H2a (t=2.211, p< 0.05), indicating that a greater degree of IA is associated with a high level of SF. H2b suggested that SF has a positive effect on EO. This relationship is confirmed based on the critical ratio (t=1.988, p<0.05).

Table 3

Structural model—Standardized path coefficients and robust t values

Structural model—Standardized path coefficients and robust t values

-> Voir la liste des tableaux

H2c posited the mediating effect of SF on the IA–EO relationship. Following Baron and Kenny’s (1986) approach, we checked the three conditions of mediation. Concerning the first condition, the impact of the independent variable, IA, on the dependent variable, EO, must be significant. This condition is fulfilled. The second condition addresses the independent variable’s impact on the mediating variable. We previously concluded that IA significantly positively affects SF; therefore, the second condition is also fulfilled. The third condition requires a significant relationship between the mediating and dependent variables. This condition is corroborated as a significant t value (2.231) between SF and EO is observed. Therefore, the mediating role of SF meets the specified requirements. To verify whether it is a partial or complete mediation, we controlled the mediator (SF) and observed that the independent variable (IA) no longer influences the dependent variable (EO) (t= 0.193 and p=0.84). Thus, we conclude that SF completely mediates the IA–EO relationship. Hypothesis H2c is supported.

Moreover, when only the significance of the regression coefficients is analyzed, the results may lead to the confirmation of a statistically erroneous mediating effect. Therefore, examining the significance of this effect and its absolute value is recommended. According to Hayes (2009), it is necessary to mobilize a nonparametric approach through bootstrapping resampling to analyze indirect effects. This method allows for better control of the type I error. Thus, following the recommendations of Preacher and Hayes (2008), we analyzed the importance of indirect effects using a macro for SPSS 10 and extracting confidence intervals. The results of the indirect impact of IA on EO through SF confirmed its validity (Tables 4 and 5).

Table 4

Direct effect of IA on EO

Direct effect of IA on EO

* Lower limit confidence interval (95%)

** Upper limit confidence interval (95%)

-> Voir la liste des tableaux

Table 5

Indirect effect(s) of IA on EO

Indirect effect(s) of IA on EO

*** Bootstrap lower limit confidence interval (95%) **** Bootstrap upper limit confidence interval (95%)

-> Voir la liste des tableaux

Regarding the effect of SX, the results (Figure 3) show that it has a negative and significant moderating effect on the SF-EO relationship (t=-2.091 and p <0.05).

Figure 3

Structural model (moderation effect)

Structural model (moderation effect)

-> Voir la liste des figures

fsQCA results

In fsQCA, the calibration of different values is required. In line with previous research, 0.95, 0.5, and 0.05 quantiles were selected to represent full set membership, the crossover point, and no set membership, respectively (Ragin, 2008). Following Ordanini et al. (2014), while using the direct calibration method, the following threshold values were used: 5 and 6 for full membership, 3 for the crossover point, and 2 for full nonmembership. The fsQCA software version 3 was used for fuzzy-set calibration.

The next step involves an analysis of the necessity or configuration elements. A condition is considered necessary when its consistency score is above 0.8. Consistency indicates the degree of coherence of a subset relationship. It is analogous to the notion of statistical significance (Schneider and Wagemann, 2010), while coverage measures are analogous to R-square in regression analysis and must be above 0.75 (Ragin, 2000).

Table 6

Analysis of necessary conditions for predicting entrepreneurial orientation (EO)

Analysis of necessary conditions for predicting entrepreneurial orientation (EO)

-> Voir la liste des tableaux

To compute the degree of sufficiency, the fsQCA algorithm was used to produce a truth table. To avoid including less significant configurations, a frequency threshold of ten observations was adopted (Rihoux and Ragin, 2009), which did not lead to the exclusion of any case in the sample. Subsequently, to identify the configurations of sufficient conditions for organizations to achieve high levels of EO, this study applied the thresholds proposed by Skarmeas et al. (2014) for determining sufficiency and coverage: 0.74 and 0.27, respectively. The fsQCA results on sufficiency conditions are presented in Table 6. The analysis produces three configurations leading to a high level of EO.

Table 7 shows that the results of Solution 2a—high levels of IA and SF—provide a high degree of consistency (0.97) and explain an increased number of cases (0.84), indicating that the combination of high levels of IA and SF mainly contributes to high levels of EO.

Solution 3a has the highest degree of consistency (0.98) and the highest number of cases (0.92), which confirms the effect of the combination (SF*SX) on EO.

Table 7

Configurations for achieving high EO scores

Configurations for achieving high EO scores

-> Voir la liste des tableaux

Discussion

Despite the growing literature in recent years on the determinants of EO, research articulating the impact of different types of dynamic capabilities remains limited. To address this research gap, this study investigated the role of three dynamic capabilities in developing EO in the French healthcare sector—IA, SF, and SX. In light of the results, IA constitutes a critical dynamic capability that favors EO. It feeds decision makers with signals facilitating the understanding of the external environment and adjusting their decisions according to new circumstances. Such an approach enables the generation of strategic choices and innovations at different levels (Herhausen et al., 2021). In our context, as a part of their environmental scanning, healthcare organizations seek innovative information to help them become more entrepreneurial.

Additionally, the SEM results confirm the mediating effect of SF on the IA‑EO relationship. Such findings address recent calls in the extant entrepreneurship literature, including Wales et al.’s (2021) and Chahal et al.’s (2019) appeals for more studies on EO antecedents. Likewise, the findings shed more light on the importance of DCs in developing entrepreneurial activities in healthcare organizations to adjust to demographic, societal, and technological changes.

The results also indicate a significant but negative moderating effect of SX on the SF–EO relationship. This negative effect can be explained by the lack of fit between the organizational structure and strategic decision-making within these organizations. Indeed, the structure can be a handicap to adapting quickly to new conditions and acting in an entrepreneurial way.

Furthermore, using the configuration approach fsQCA, this paper addresses the complexity underlying these relationships, focusing on the healthcare sector. EO is more likely to be associated with various combinations of dynamic capabilities, thus confirming equifinality and complexity.

First, three patterns emerge from our analysis, confirming that entrepreneurial orientation cannot result from a single factor. Instead, it combines a set of factors (e.g., IA+SF and SF+SX). Thus, EO in this sector requires effort at several levels. Individuals, organizations, and institutions must work together to manage constraints (financial, administrative, hierarchical, etc.) to adapt to change and the imponderability of future developments. The development of EO is not a singular fact resulting from managerial behavior but also from an organizational context favoring it. The degree of this entrepreneurial orientation differs between the private and public sectors, as the institutions in these sectors need to be managed similarly.

Second, the results of the fsQCA highlight the importance of strategic flexibility, which is present in the two configurations. This could be explained by the complexity of this sector, which brings together several actors, several professions, and a multitude of innovations to be managed. Managers face a significant challenge in balancing structural and financial constraints with the demands and needs of users. Being flexible facilitates the introduction of new working methods and, above all, a certain agility in decision-making.

Research contributions

The findings provide several critical contributions to the literature on the determinants of EO (Anderson and Eshima, 2016; Gao et al., 2016; Rodrigo-Alarcon et al., 2017), a field of research that remains understudied (Wales et al., 2020). In their bibliometric study on this concept, Wales et al. (2020) observed that very little work has been done on the determinants of this strategic posture. They called for mediator–moderator models to identify the interactions between organizational, human, and process factors facilitating EO and translating it into performance.

First, the present research clarifies the interaction between several factors and their contribution to the development of EO within healthcare organizations. The study empirically investigates the direct, mediating, and moderating effects of IA, SF and SX on EO by analyzing simultaneous relationships in a structural equation model, which is novel research in this domain. Furthermore, fsQCA helps provide a more comprehensive understanding of EO antecedents. The aim is to adopt a more holistic method (i.e., configuration analysis) that captures complex interactions leading to EO. We therefore add further evidence to the nascent wider body of works calling for a complex approach when studying entrepreneurial behavior in general (Douglas et al., 2021; Sahin et al., 2019).

Second, our study contributes to the field of management of healthcare organizations. Studying entrepreneurial behavior in these structures sheds light on the factors that can facilitate innovation and change in their management methods. Only a few studies in English have explored the role of specific dynamic capacities in the health field by examining the literature in this field. This study answers theorists’ call for a greater understanding of EO in the context of the hospital and healthcare sector (Chahal et al., 2019;), which is characterized by strong entrepreneurial dynamism and profound change (Lages et al., 2017; Guo, 2006; Jansen and Moors, 2013; Brandt and Znodtka, 2019, Nobre, 2013, Haschar-Noé and Basson, 2019).

Managerial and policy implications

Our findings provide practical managerial and policy implications. In today’s competitive and challenging environment characterized by technological, health, social, and economic changes, there is a consensus on the need for healthcare organizations to establish an organizational culture encouraging entrepreneurial behaviors and healthy managerial attitudes toward risky activities to survive.

However, the ambiguity lies in the tools and means that can be mobilized to develop this posture. This study proposes three dynamic capabilities that enable these organizations to develop agility by sensing and monitoring the environment to respond rapidly to new threats or opportunities. Accordingly, flexibility in decision-making and organizational design is needed to enhance these organizations’ ability to adapt and anticipate unique needs. Flexibility in policies is widely recommended to cope with the seismic changes faced in uncertain environments. As a dynamic capability, it facilitates adaptation to internal and external changes and empowers managers to swiftly react to disruptions while benefiting from unexpected opportunities and challenges. Our findings may encourage healthcare managers to introduce innovative actions to support their resilience in coping with environmental changes and constraints.

Conclusion, limitations, and directions for future research

Empirical studies focus primarily on EO’s direct effect on performance (Covin and Wales, 2012; Shan et al., 2016; Lisboa et al., 2016), while less attention has been devoted to how firms can develop this strategic orientation. This study investigated the factors influencing EO in firms and demonstrated that an organization’s ability to promote its EO is determined mainly by the fit of its organizational structure and management practices with its entrepreneurial activities.

This study offers a perspective to understand EO antecedents, as these have rarely been examined in the entrepreneurship and strategy literature; nevertheless, many limitations must be addressed. One principal limitation arises from the sample, which includes data from a particular area (healthcare) in one country (France). Sector- and country-specific factors—as well as the industry’s rate of change—may significantly impact organizations’ strategic actions and structure. In addition, the distinction between the private and public sectors has not been clearly put forward in this study. The degree of flexibility is not the same in both sectors. Future studies could address this point and extend their study to other contexts to see the specificities of each country. Furthermore, new insights could also be derived by including other potential EO determinants, such as the role of legislation. The legal framework is a predominant external influencing factor in the healthcare sector. Extensive legal constraints are typical in the public sector, but private organizations are also tightly controlled and must meet specific requirements. The legislative branch encourages health actors to act entrepreneurially, but legal standards are inappropriate for implementing entrepreneurial culture in many cases.