Why serendipitous informal knowledge sharing interactions are key to boundary spanning and creativity

1Introduction

Innovation projects are characterized by a high and dynamic knowledge demand. Informal knowledge sharing interactions (IKSI), in which colleagues voluntarily provide their knowledge to solve the tasks of others, have proven to be the most important source for this [1–3]. They enable the efficient and timely identification and acquisition of relevant knowledge. Thereby, IKSI enhance the job satisfaction and well-being of employees in knowledge-intensive industries [4, 5]. Furthermore, IKSI have proven to be an important driver for the success of innovation projects. While early research focused on the number of IKSI, more recent work points more strongly to the importance of partner constellations. Beneficial are especially the IKSI, which involve colleagues who hold knowledge and competencies that are comparatively distinct from their own. It is argued that IKSI between heterogeneous partners are more likely to convey non-redundant knowledge and thus open up new perspectives on the own project. They thus make a particular contribution towards discovering further development paths for innovation projects and thus increase their chances of success as well as the job satisfaction and well-being of the responsible employees [4–7].

Against the background of their considerable impact, innovation research has intensively investigated the formation of IKSI. Thereby, robust formation mechanisms with high explanatory power have been identified [8, 9]. By applying these, the social, formal, and material embeddedness of individuals can be used to explain which colleagues are likely to involve each other through IKSI. Further, these insights are used to stimulate the formation of IKSI in practice. While recent research on the effect of IKSI emphasizes the heterogeneity of interaction partners, this has not yet been taken into account in explaining their formation. On the basis of existing theory, it is therefore not possible to explain the particularly valuable IKSI among heterogeneous colleagues. In this paper, I will argue that IKSI between heterogeneous interaction partners emerge in a systematically different way than those between similar ones. Given these differences, IKSI and the advice networks they constitute can be explained and stimulated in a more nuanced way.

The basic argument of the paper is derived from the literature on serendipity. In innovation research, serendipity describes the phenomenon “that researchers make unexpected and beneficial discoveries” [10]. The serendipity literature emphasizes that researchers who focused on a specific problem found its solution not solely through systematic research activities, but remarkably often as a result of unplanned incidents [11, 12]. Besides accidentally triggered chemical reactions, this literature also anecdotally refers to unplanned interactions as incidental impulses for innovation. In this paper, I take up this idea and make the case that serendipitous informal knowledge sharing interactions (SIKSI) are significantly more likely than planned informal knowledge sharing interactions (PIKSI) to connect heterogeneous interaction partners and, accordingly, are more likely to provide non redundant knowledge and thus open up new perspectives. To unfold the argument, I will shed light on the mechanisms identified to explain the formation of IKSI, asking in each case (1) to what extent they produce SIKSI or PIKSI and (2) whether similar or heterogeneous interaction partner constellations are to be expected. The analysis of the concepts refines and corroborates the argument. It shows that a bias towards local search leads PIKSI to bring together colleagues with similar knowledge and skills, while SIKSI break this bias and thereby produce more heterogeneous partner constellations. The main aim of the paper is to develop the theoretical argument by systematically integrating two unrelated strands of research. Besides, it finally provides a first empirical investigation of the argument. Based on the analysis of 132 IKSI related to product development projects at large companies, it is confirmed that SIKSI are significantly more likely than PIKSI to open up new perspectives.

2The interactive development of innovation

Already the founders of innovation research pointed out that innovations usually result from the successful recombination of existing knowledge [13, 14]. In this way, they emphasized that successful innovations cannot be attributed so much to the unrestricted course of a free creativity, but rather require detailed knowledge and competencies. This assumption is also shared in current creativity and innovation research [15, 16]. On the other hand, a significant difference exists with regard to who conducts the knowledge combinations. Whereas Schumpeter in particular focused on creative individuals, current innovation research shows that innovative knowledge combinations are primarily produced interactively. Because the expertise of individuals is usually limited to one or a few specific areas, the possibilities for novel combinations in the area of their knowledge are limited. In contrast, comparatively many new combination possibilities arise when individuals with different expertise combine their knowledge interactively. Interactions between individuals are therefore key to the development of innovations [17–19].

While the sharing of knowledge in basic research is conducted to a greater extent in written language, the development of novel products, services and processes in enterprises is characterized in particular by verbal knowledge sharing interactions, which are less precise but take significantly less time [20]. The interactive acquisition of knowledge and competencies from colleagues can significantly improve the (low) chances of success of innovation projects. Developers in companies are concerned with developing and prototyping ideas with innovation potential. At the beginning, the ideas are usually abstract and it is very uncertain whether they can actually be realized and successfully marketed [21, 22]. The task of developers is therefore to make ideas tangible. This requires creativity. To find the best possible path for development, developers must discover a wide variety of approaches and pursue alternative paths to test their suitability. On the one hand, development is an ongoing creative process that benefits from identifying as many potential paths as possible. On the other hand, because resources are scarce, selection is omnipresent. Individual development paths as well as entire innovation projects are continuously reviewed, retained or discarded [20, 23]. Both the identification of development paths and the selection of the most promising ones benefit significantly from additional knowledge and perspectives.

Such knowledge sharing interactions take place in a formal context between the members of an innovation project team and their superiors [24–26]. However, it has been shown that developers also engage in informal knowledge sharing interactions (IKSI), in which colleagues voluntarily provide their knowledge to solve the tasks of others together. The informal nature of these interactions means that there is no formal reason for the committed colleagues to contribute to the projects of others and that no working time is available for this purpose. The fact that they nevertheless take place and are so important has three main reasons.

First, the efficient handling of projects requires the smallest possible number of project staff. This keeps coordination costs low and avoids overburdening individual employees through parallel involvement in a large number of projects [27]. The perspectives represented in a project are therefore limited from the outset by the number of people involved. Second, innovation projects are characterized by a particularly high degree of fuzziness [28]. Precise activities are defined in the project on the basis of the current state of knowledge and, because this changes, are adapted again and again. As a result, at the start of the project it is very difficult to foresee which specific knowledge will be relevant later on. The specific knowledge required for the projects only becomes apparent during the projects. Staffing the projects with the most suitable developers is therefore only possible to a very limited extent, because it is unclear what knowledge will be required. Through IKSI the relevant knowledge for the project can be obtained. Third, the capacities of developers (and most other employees in a company) are allocated to projects in the medium term [23]. If it is recognized in the course of a project which specific knowledge is relevant to it, it is likely that the corresponding knowledge carriers will be unavailable at the moment. Since the success of projects also depends on whether developments can be completed faster than competitors, waiting for a specific expert to become formally available is usually not an alternative. The efficient organization of development processes in companies is therefore only compatible to a very limited extent with formally integrating the broadest possible spectrum of relevant knowledge carriers into projects. IKSI make it possible to satisfy this need and thus crucially contribute to the success of innovation projects [20, 29–31].

3Which interactions are particularly beneficial for innovation?

In line with what has been said so far, research further shows that not all IKSI are equally beneficial for innovation projects. Of particular importance here is the extent to which the knowledge stocks of the interaction partners are complementary. In the research on the relationship between the heterogeneity of knowledge stocks and innovation successes, two findings in particular have proven instructive.

First, differences between the knowledge bases of individuals and organizations are described as cognitive distances [17, 32, 33]. The larger this distance turns out to be, the smaller the overlaps between the knowledge stocks are. Applying this concept, it is stated that a low cognitive distance is not very fruitful because the potential for learning from each other and recombining knowledge is low. On the other hand, a very large cognitive distance causes communication difficulties that can create a great effort and bring the exchange of knowledge to a breakdown. A medium cognitive distance is therefore considered particularly promising. It characterizes a state in which the knowledge stocks of the interaction partners are on the one hand connectable, so that interaction partners can communicate without major difficulties, and in which they are on the other hand still so different that there is potential for learning and recombination.

Second, differences and similarities are placed in a relationship to the concrete task instead of only considering the knowledge stocks per se. Based on this distinction, it appears that interaction partners who have different task-related knowledge and are otherwise alike are best positioned to successfully accomplish the complex innovation tasks [34–36]. However, it is also reflected here that in the case of innovation projects, it is highly unclear what knowledge needs to be combined to be successful. Therefore, a wider range of differences is also considered more fruitful here.

The presented literature is characterized by a high level of abstraction. In research on IKSI, the causal relationships can be simplified through greater contextualization. This concerns both the optimal heterogeneity measure and the empirical identification of heterogeneity. It follows from the foregoing that IKSI promise to contribute to the success of innovation projects when the knowledge bases of the interaction partners have both similarities and differences. In this context, it is significant to reflect that IKSI take place first and foremost among colleagues within an organization. Since specific organizations attract only certain individuals, and only certain individuals are hired and remain, the employees of an organization are typically characterized by a comparatively high degree of homogeneity [37]. This is further enhanced by the fact that employees in organizations share knowledge interactively with each other, thereby assimilating. IKSI therefore tend to benefit from a relatively high heterogeneity of interaction partners rather than risk suffering from it. In research, therefore, the value of heterogeneity is emphasized foremost [1, 36, 38].

A particular challenge for research on heterogeneity of IKSI partners –also as a result of the comparatively high homogeneity of employees in an organization –is to empirically capture differences in knowledge stocks. Due to the high complexity, it is not possible to directly determine the knowledge stocks of interaction partners using established methods. Therefore, it has proven useful to approximate heterogeneity by means of more easily ascertainable factors. Three approaches have proven to be particularly valid.

First, more easily accessible characteristics of individuals such as the disciplinary classification of their degrees are seen as markers of specific knowledge stocks. Apart from the fact that individuals received a corresponding imprint through their disciplinary education and thus possess subject-specific knowledge, competencies, and perspectives, it is pointed out here that even the choice for a subject implies a strong preselection. Accordingly, students from different disciplines differ systematically even before they start their studies, and this is also true with regard to their knowledge and perspectives. Unsurprisingly, research proves that categorical heterogeneity is related to the heterogeneity of knowledge stocks and that interaction relationships between individuals who are heterogeneous in this sense show the expected innovation-promoting effects [38, 39].

Second, the heterogeneity of knowledge stocks correlates with their embeddedness in social structures. This argument was first formulated by Granovetter [40]. He argues that non-redundant knowledge can be expected to be more likely when interaction partners do not maintain close relationships but are connected by weak ties. This is based on the assumption that individuals who maintain strong ties are also likely to have similar knowledge and perspectives because they probably interact more frequently and have the same or similar sources of information. The basic argument has been refined by Burt [41] in terms of network analysis. In contrast to Granovetter, however, he looks not only at individual ties or networks of individuals, but at the totality of relationships of specific circles of individuals (e.g., members of an organization). He finds that clusters are formed within these circles. Constitutive for clusters is that individuals who maintain a relationship also maintain relationships with the same third persons. Accordingly, the ego networks of individuals in a cluster are to some extent alike. Because they have the same contacts, the individuals’ knowledge and perspectives are also similar. Furthermore, the tendency to align is promoted by dense relationship networks in clusters, since social control and subsequent social pressure to conform is stronger on each individual in such. Divergent views are therefore more likely to be held back and the development of different perspectives is impeded [42–44]. Within clusters, therefore, there is a relatively high degree of homogeneity with respect to the knowledge and perspectives of the actors. The formation of clusters is accompanied by the formation of structural holes, which are characterized by the absence of relationships between individuals belonging to different clusters. Burt thus notes that individuals belonging to different clusters are likely to have reasonably different knowledge and perspectives, and that bridging structural holes is therefore likely to lead to non-redundant knowledge [6, 45].

Third, the heterogeneity of knowledge stocks correlates with their formal structural embedding. Comparable to the argumentation on the importance of social structures, it is argued here that knowledge is shared to a greater extent within organizational boundaries (e.g. in departments) than between different organizational units. Here, too, both a higher concentration of interaction, incentive mechanisms rooted in hierarchies, and employee (self)selection processes ensure a homogenization of knowledge stocks. The interactive bridging of formal boundaries, referred to as Boundary Spanning, therefore leads to non-redundant knowledge and is therefore particularly fruitful for innovation projects [30, 46, 47].

In summary, research shows that IKSI between individuals who differ categorically, socio-structurally, and/or formally are particularly fruitful for innovation projects because they are likely to connect individuals who have different knowledge stocks. There is therefore the potential to learn from each other and recombine disconnected knowledge to solve complex problems and create innovations.

4Who turns to whom for IKSI?

While research clearly shows that IKSI promise to provide crucial impulses for innovation projects especially when the interaction partners differ from each other categorically, socio-structurally and/or formally, it is unclear how exactly such constellations are formed. Although a rich body of research exists on the formation of interactions, it does not distinguish between the formation of homogeneous and heterogeneous interaction partner constellations. In the absence of this distinction, it is not possible to explain how the particularly beneficial IKSI between heterogeneous partners come about. However, it seems extremely desirable to resolve the question of how they come about, because this would enable more far-reaching explanations of innovation processes and stimulation of the fruitful interactions in practice in a more systematic manner than before. Stimulation has great potential, mainly due to the fact that (in contrast to research-based recommendations and partly contrary to the conviction of the practitioners) mainly homogeneous IKSI partner constellations are formed in practice and heterogeneous ones are rather the exception.

Following the literature on serendipity, I hypothesize that heterogeneous interaction partner constellations are formed primarily through serendipitous informal knowledge sharing interactions (SIKSI). SIKSI are characterized in that the interacting individuals did not plan them and just make them happen, but that their occurrence was linked to conditions beyond the control of the individuals. PIKSI are for example such IKSI, which result from a colleague, who was chosen as interaction partner in advance, being deliberately approached or a corresponding meeting being arranged with this colleague.

In line with the concept of serendipity, I do not assume that SIKSI strike people completely without their involvement [12, 48]. Serendipity research has highlighted that unplanned discoveries are usually accompanied by the discoverers being on the hunt. Accordingly, I assume that although SIKSI are not concretely planned and in a sense arise, there is certainly a fundamental interest in impulses about a particular problem that occasionally becomes motivating for action when an opportunity arises [11, 12]. For example, an interaction is still a SIKSI when a developer already had the idea that a certain colleague could help him or at least that it might be helpful to talk to someone about it, but the concrete interaction resulted from an unplanned encounter. Furthermore, serendipity research emphasizes that events do occur unpredictably and unplanned for the discoverers. However, they are not random in the sense that they are completely unexplainable. Rather, it is assumed that they do occur systematically and that the corresponding patterns can be worked out scientifically. This is also true for SIKSI. As I will discuss further, the formation of these is influenced, for example, by spatial structures that explain who is likely to encounter whom unplanned. Even if these encounters appear random to the individuals involved, they arise systematically and can be explained on the basis of the corresponding scientific analyses [12, 48].

I argue that SIKSI, rather than PIKSI, connect individuals who are categorically, social-structurally, and/or formal-structurally distinct from each other. Previous research on the formation of ISKI has not distinguished the partner constellation in this sense. To develop the argument, I will apply the conceptual distinction between SIKSI and PIKSI introduced here to the state of research on the formation of IKSI. To do so, I hypothesize that PIKSI are highly attributable to partner selection decisions, while SIKSI are less attributable to such decisions and more to opportunities. I will show that the decisions shaping PIKSI have a strong bias toward local search. Specifically, colleagues prefer each other as PIKSI partners when they are categorically, social-structurally, and formally close. In contrast, I will show that the mechanisms explaining SIKSI are decoupled from these biases to a greater extent, thereby favoring the involvement of colleagues who are less close categorically, social-structurally, and formal-structurally. In what follows, I will discuss the different mechanisms and shed light on the extent to which they contribute to the homogeneity/heterogeneity of interaction partners. I will first discuss the choice-based mechanisms that explain PIKSI and then the opportunity-related mechanisms that are central to the explanation of SIKSI.

4.3Summary

In this section, I have presented the mechanisms that explain the formation of IKSI and asked in each case to what extent they bring together heterogeneous interaction partners in terms of social categorization, social structure, and/or formal structure. I posed the question in light of the finding that heterogeneous interaction partner constellations have proven to be particularly fruitful for innovation projects, as they are more likely to provide non-redundant knowledge and open up new perspectives. The analysis has shown that all mechanisms that affect partner selection decisions and thus explain PIKSI constitute local search and thus lead to the involvement of a few homogeneous colleagues. In contrast, opportunity-based explanations are particularly powerful in the formation of SIKSI. Accordingly, the logic underlying their emergence appears to be highly decoupled from categorical, social-structural, and formal-structural constraints (see Fig. 1).

Fig. 1

PIKSI typically connect individuals that are categorically, formally and socially similar (local search) while SIKSI connect individuals that are different in this respect (boundary spanning).

SIKSI therefore promise more heterogeneous interaction partner constellations that are more likely to provide new perspectives and thus important impulses for innovation projects. In the following, this argument will be tested in a first empirical investigation.

5Method

The derived argument claims that SIKSI are more likely than PIKSI to open up new perspectives. The link between the formation and impact of IKSI has not yet been adequately considered. One reason for this is that it is a particular challenge to validly capture the formation of interactions. Accordingly, the study presented in the following aims not only to empirically test the theoretically substantiated link, but also to propose and exemplarily apply an adequate procedure for validly detecting the formation of IKSI.

5.2Data collection

The particular challenge of the empirical study was to validly record the occurrence of the IKSI. The two procedures established in the research field appeared unsuitable for this purpose. In the network-analytical studies on IKSI, it is typically asked retrospectively with which colleagues such interactions took place. However, critical studies of this procedure have demonstrated recall biases [79, 80]. In particular, it has been shown that participants are worse at recalling interactions that tend to be less common. This may be tolerable for many studies. Here, however, it would mean that especially SIKSI would not be adequately represented. A comparison would therefore lack robustness.

More robust methods are used by some studies that focus on the spatial embeddedness of the formation of interactions [e.g. 76, 81]. In these, the interactions themselves are captured through observations. However, this approach also seems unsuitable here, since these studies only insufficiently succeed in differentiating forms of interaction. Based on the observation of an interaction, it is only possible to a very limited extent to determine whether it is an IKSI and whether it is planned or serendipitous. The information that is crucial for the question focused on here could therefore not be reliably recorded in this way.

We therefore decided to use event-based diaries [82]. This method is characterized by participants themselves recording events according to predefined rules. Namely, the participants were instructed to document interactions whenever they discussed topics related to current development projects with people formally not involved in them. To structure the documentation, participants were supplied with a digital diary template. To determine whether the recorded interaction was a SIKSI or a PIKSI, participants were asked how the interaction occurred (see Fig. 2).

Fig. 2

Question on the formation of IKSI from the diary template.

If the fourth box was checked, the interaction was classified as SIKSI. If the third box was checked, further consideration was given to how spontaneously the intention evolved. If the question about this (Fig. 3) was answered with “Only during the encounter... ”, the interaction was also classified as SIKSI.

Fig. 3

Question on the intentionality of IKSI from the diary template.

Finally, it was asked whether the interaction had opened up new perspectives on the project. As an additional quality control, interviews were conducted with all participants to reconstruct how the interactions occurred and to verify whether they were indeed SIKSI or PIKSI. IKSI were recorded in this way for one month in each of the three departments. Data collection was conducted in all cases in 2017 and a total of 132 IKSI were recorded.

The data basis allows a first empirical test of the conceptually elaborated correlation. However, it is limited in two respects. First, the number of cases is comparatively small. Second, new perspectives are recorded and interpreted here as an indicator of heterogeneous partner constellations, while the heterogeneity of the partners themselves is not captured independently.

6Results

The count of the 132 cases initially shows that 65 cases are SIKSI and the remaining 67 cases are PIKSI, so that the two types of IKSI are statistically well comparable. Counting the frequencies concerning the outcomes of the IKSI reveals a clear picture (Table 1). While 75% of the SIKSI open up new perspectives, only 43% of the PIKSI do so.

The two-sided chi-square test shows that the differences found here between SIKSI and PIKSI are not random and thus significant (significance level 0.01). Hence, the analysis shows that SIKSI significantly more often lead to new perspectives than PIKSI.

7Discussion and conclusion

Informal knowledge sharing is of fundamental importance for numerous knowledge-intensive processes in organizations and especially innovation projects. Research on the effect of such interactions in organizations has shown that particularly heterogeneous partner constellations open up new perspectives that give important impetus to innovation projects. By facilitating the success of the projects, such IKSI enhance job satisfaction and well-being of employees. However, research also shows that, contrary to this evidence, IKSI in practice primarily connect individuals who are categorically, socio-structurally, and formal structurally proximate to each other. There is a bias toward local search. Against this background, the question arises as to how IKSI between heterogeneous partners nevertheless emerge.

The main contribution of this paper is to provide an elaborate answer to this question. Following the literature on serendipity, I have argued that serendipitous informal knowledge sharing interactions are more likely than planned informal knowledge sharing interactions to connect heterogeneous partners and lead to new perspectives. To substantiate this argument, I intertwined the two unconnected strands of literature on the formation and impact of informal knowledge sharing interactions. Drawing on this rich knowledge base, I elaborated the different mechanisms by which SIKSI and PIKSI are formed and which categorical, social, and formal partner constellations form in consequence. Hence, I showed in detail why SIKSI are more likely to provide valuable impetus for innovation projects. Complementing this, the argument that SIKSI more often lead to new perspectives than PIKSI has been confirmed in a first empirical investigation. The paper thus offers a solidly substantiated argument that is relevant in the three referred research fields and in practice.

A starting point for further research results from the limitation of the presented empirical study. Due to the high plausibility, the heterogeneity of the interaction partners themselves was not recorded. However, it seems possible that the higher fertility of SIKSI has other causes besides partner heterogeneity. For example, it seems plausible that the interaction situations in which SIKSI typically take place encourages more open-ended interactional behavior because interactions in such situations are less focused on specific questions and the mutual expectations are less elaborated and constraining [77, 78]. A closer look at the causes of the correlation therefore seems worthwhile.

The basic argument was derived from the research on serendipity [10, 81]. In this literature, it is emphasized that serendipitous events can provide important impulses for innovations in research and development. So far, however, only anecdotal references to serendipitous interactions have been made. This paper has systematically considered such interactions and reconstructed them in detail on the basis of the related literature. Thus, an important subject of this emerging field of research has been profoundly explored. A special feature is that not only a single serendipitous phenomenon was considered, but it was systematically compared with the planned counterpart. The comparative analysis enabled to precisely determine the significance of serendipity for the specific case of IKSI. In light of the findings, it seems extremely promising to pay further attention to serendipity phenomena, systematize it more strongly and to compare the underlying processes with planned processes.

Concerning research on the formation of interactions and the relations that emerge from them, the findings of the paper primarily call for a further distinguishing of partner constellations [8, 9]. Typically, in this research, the formation mechanisms are brought into focus, while the categories for interactions and relationships remain fuzzy. Here, the distinction between heterogeneous and homogeneous interaction partner constellations, which is found to be extremely important from research around the effect of interactions, was introduced and revealed that they follow different formation logics. This analytical distinction significantly extended the explanatory power of the known mechanisms. For further research, it seems extremely promising to leverage even more potential by distinguishing more precisely between partner constellations or types of interaction in explaining their formation. This promises a more accurate description of the underlying formation mechanisms. Further, comparative analysis of specific interaction types and their formation promises generalizable insights on this connection. These could be the basis of a general theory on which characteristics of interactions are associated with which formation mechanisms. Since interactions and relationships are of fundamental importance in numerous fields, such a theory would have a significant impact.

With regard to research on the effect of interaction partner constellations [45], this paper expands the focus to their formation. Although this research deals with dynamic processes, interactions are often conceptualized as static channels and reduced to the characteristics of the individuals involved. Although this conception of the object was adopted here and the explanation of heterogeneous and homogeneous partner constellations was explored, with their formation at least a part of the phenomenon was considered in its dynamics. This has proved to be extremely fruitful here. Subsequently, it seems promising to consider further parts of interactions as a process. Specifically, we should ask about the scope of action individuals have in the respective constellations, how varying agency can be explained, and what consequences they have respectively. A seminal example is the contribution of Obstfeld et al. [83], who show that individuals in broker positions can act differently and that the structural constellation has different effects depending on their behavior. However, further research is needed, for example, on which practices make the same heterogeneous partner constellations result in innovations or in conflicts.

Finally, the paper underlines the importance of informal and unplanned interactions in the context of innovation projects and other knowledge-intensive activities in organizations. This point seems to be particularly important against the backdrop of digitization - which has been significantly accelerated by the corona crisis. Digitization is often accompanied by a strong formalization of processes. In consequence, the opportunities for informal and unplanned interactions, are increasingly disappearing. The paper can therefore also be read as a call for a stronger consideration of informal and unplanned interactions when organizing work and designing (especially technical and spatial) working environments. Indications for appropriate design measures are provided by studies that reveal the general situational prerequisites for such interactions [77, 78] as well as those that focus on the effect of specific medial or spatial work environments on interactions [84, 85]. However, further research is needed here to provide a reliable basis for the careful design of such measures.

Ethical approval

Not applicable.

Informed consent

Not applicable.

Conflict of interest

None to report.