Ready to innovate during a crisis? Innovation governance during the first wave of COVID-19 infections in Iceland

ABSTRACT

Previous studies on innovation governance have focused on the governance of science, technology, and innovation from a long-term perspective. In this article we focus on the short term by exploring the generation and use of new scientific and technical knowledge to address an urgent societal crisis. We empirically analyse the emergency response during the first wave of COVID-19 infections in Iceland using a conceptual framework based on three theoretical components, namely, emergency management, innovation governance, and the innovation process as a problem-solving process. The empirical analysis is built on a systematic analysis of secondary data. Based on the results, we conclude that improvisation processes using existing knowledge and capabilities and triggered by unanticipated problems during a crisis are in some cases sources of successful innovation. In these cases initial problem-solving processes characterized by improvisation can be interpreted as blind variations that are retained and diffused through a series of complementary problem-solving processes that generate and use new scientific and technical knowledge. Furthermore, we extend the concept of innovation governance readiness to include both the readiness to exploit technological opportunities and the readiness to address unanticipated problems during a crisis and propose that our extension is useful for integrating long-term and short-term aspects of innovation governance.

KEYWORDS:

• Innovation governance
• COVID-19
• emergency response
• problem solving
• innovation governance readiness

Introduction

A reoccurring theme within innovation studies is that science and technology are bodies of knowledge which are developed by quite different communities involved in different sets of activities, leading to a series of interesting debates about how to improve innovation governance (Nightingale, 1998). Much of the early literature took the perspective of national and regional innovation systems, often stressing interactive learning and networks across public and private actors as key processes stimulating science, technology, and innovation (Edquist & McKelvey, 2000, p. xviii), but with less discussion on how coordination could be achieved. More recent literature on innovation governance (Kuhlmann & Ordónes-Matamoros, 2017) puts more emphasis on the complexity of governing the interactions between science, technology, and innovation over long time periods, with some approaches applying the concept of large socio-technological systems (Borrás & Edler, 2014). Innovation governance requires coordination and rule setting by diverse actors such as policy makers, universities, research institutes, industrial firms, regulatory authorities, and user communities, which also influence which innovations are produced, introduced, and diffused in society (Borrás & Edler, 2014; McKelvey et al., 2020). Potts (2019) calls the specific collaboration over new ideas and knowledge ‘innovation commons’, which to some extent presupposes a phase of knowledge development before market forces influence the selection of innovations. Both of these approaches to science, technology, and innovation tend to focus on long-term determinants, stressing concepts such as trajectories and institutional lock-in (Arthur, 1994) as underlying patterns in the economy.

However, a major emergency or crisis involves heterogeneous actors and also requires a rapid response, which is a very different situation. Thus, a gap in this literature is the analysis of short-term responses, which in relation to innovation governance explicitly means how science, technology, and innovation can be stimulated and regulated to help solve problems during a major crisis. Note that we are not referring to crises generated by the ‘dark side of innovation’, such as failures, unintended consequences, and grey zones of innovation (Coad et al., 2020). Instead, we are interested in major societal crises, which are characterised by the need for society to rapidly identify and address unanticipated problems alongside emergency response plans and management. Hence, the purpose of this article is to improve our overall understanding of innovation governance by exploring the generation and use of new scientific and technical knowledge to address an urgent societal crisis.

We propose to use the concept of Innovation Governance Readiness (IGR) to express a national capacity to respond to specific types of problems or act upon opportunities. Technological readiness is well established in the medical field, with the notion of readiness to enable adaptation of a particular technology into clinical practice (Mankins, 2009) and with some calls to extend it to include institutional readiness (Webster & Gardner, 2019). In response to this discussion about including institutional and national capacities more broadly, McKelvey and Saemundsson (2021) define IGR as the capability to dynamically stimulate and regulate complementary types of experimentation in such a way as to ensure continued collective action by public, private, and civil society actors. Collective action as considered here is limited to activities by individuals and organisations that may have different incentives and goals but are united by a common interest in the generation and use of new scientific and technical knowledge related to a particular problem or opportunity.

In this article, we contribute by further developing the concept of IGR. We do so by explicitly focusing on the concept of ‘problem solving’ in relation to the initiation, or activation, of an innovation governance system that is put in place to solve a particular problem or take advantage of an opportunity. In doing so, we propose that McKelvey and Saemundsson (2021) conceptualisation can be enriched by current thinking about emergency management, which has moved from an emphasis on gaining command and control of a situation to an emphasis on continuity, coordination, and cooperation (Dynes, 1994; Quarantelli & Dynes, 1977).

This article focuses on an empirical context relevant for medicine and health care more generally, namely, the COVID-19 pandemic. In 2020, the COVID-19 global pandemic created a crisis in many countries, thus providing a unique opportunity to study innovation governance during a crisis. From a public policy perspective, OECD and WHO have commissioned many reports and gathered digital and written resources to promote good governance of the crisis and to investigate how similar crises should be addressed in the future (cf. OECD, 2022; WHO, 2021b). Furthermore, policy researchers have studied local and regional health system responses to COVID-19 (Saunes et al., 2022) and the general implications of the pandemic for governments, such as their competencies and dynamic capabilities (Mazzucato et al., 2021) as well as ethical aspects of crisis management in the digital age (Boersma et al., 2022). Finally, innovation management scholars have studied how various aspects of the crisis, both medical and non-medical, have been addressed around the world (Di Minin et al., 2021). These include considerations of a crisis model of innovation (Ardito et al., 2022; Dahlke et al., 2021; Hanisch & Rake, 2021) and the role of open innovation during a crisis (Bertello et al., 2022), and questions have been raised about the potential for rethinking how and where medical innovation takes place (George et al., 2020).

In many countries, government agencies had already prepared emergency management plans before the outbreak of COVID-19 that dictated how they should respond to a pandemic, including how activities across different actors should be coordinated during the response. However, at the beginning of the crisis very little was known about how quickly the virus would spread, how many of those infected would need hospitalisation, what medical treatments would be effective, and what Non-Pharmaceutical Interventions (NPI)¹ would be appropriate. Thus, to address the crisis, there was a need to quickly generate and use new scientific and technical knowledge.

The structure of the paper is as follows. Section 2 provides the conceptual framework for our analysis of innovation governance during an emergency response to a crisis. In Section 3 we describe the research methods of the study, which is based on the analysis of secondary data related to emergency management plans and response to COVID-19 in Iceland during the first wave of infections in spring 2020. In Section 4 we present the results of our analysis by describing the emergency management of COVID-19 in Iceland, the sequence of major events during the response, and innovation governance of problem solving. In Section 5 we revisit our conceptual framework, and in Section 6 we conclude and provide suggestions for further research.

Conceptual framework

The conceptual framework of the study is based on three theoretical components, namely, 1) emergency management; 2) innovation governance, and 3) the innovation process as a problem-solving process. As we focus on the context of health care, our references and empirical context draw primarily from previous work in that area.

The first component of our conceptual framework is emergency management, to capture the short-term and urgent aspects. Emergency management is a discipline that is concerned with risk and risk avoidance (Haddow et al., 2011). Emergency management is therefore an important function of government concerned with public health and safety, which, because many risks are international in nature, is coordinated through international organisations such as the United Nations (UN) and the World Health Organization (WHO). Moreover, as the COVID-19 pandemic was already perceived to be a threat to public health at its onset, we find it reasonable to assume that innovation governance during the pandemic was strongly influenced by emergency management of public health and safety.

A comprehensive emergency management model includes four types of activities or phases: mitigation, preparedness, response, and recovery (National Governor’s Association, 1979; Phillips et al., 2017). Mitigation includes activities that aim to reduce the probability of a disaster or to reduce the impact of disasters. Preparedness concerns the preparation for a potential disaster, for example, by educating the public, building warning systems, planning a coordinated response and recovery, and providing training. In the response phase, a disaster has occurred, and plans are activated and carried out. Finally, recovery includes activities to restore the situation to normal, for example, restoring services that were disrupted or infrastructure that was damaged or temporarily put to different use during the response. Of these four, our empirical study focuses on the response phase.

In this literature, a dominant approach to the governance of emergency response has emerged that has been labelled as the problem-solving model (Dynes, 1994). Traditionally, disaster response was considered in military terms, with an emphasis on gaining command and control over a situation where normalcy had disappeared, in terms of infrastructure, mechanisms of government, and social control. This was believed to require the establishment of a clear authority for making the right decisions and the ability to communicate these decisions to the relevant actors, including the public. In contrast, current thinking stresses the continuity, coordination, and cooperation model of emergency management (Waugh & Streib, 2006). By assuming continuity of behaviour, existing structures and routines should be utilised as much as possible. These need to be coordinated during the response, and emergency planning and rehearsals should mainly be concerned with the mechanisms, techniques, and facilities that promote coordination and common decision making across multiple organisations. Finally, cooperation with the affected population and organisations should be considered, as research has shown that people experiencing a disaster are more likely to be motivated to help with the response rather than being traumatised or showing irresponsible behaviour as assumed by the command-and-control model (Dynes, 1994; Quarantelli & Dynes, 1977).

Furthermore, our interpretation is that this problem-solving model of emergency management is based on two additional assumptions that are also relevant for our view on innovation governance. First, it is assumed that crises are characterised by radical uncertainty, where unanticipated new demands and problems are likely to emerge. Second, it is assumed that the most effective way to deal with this uncertainty is to take advantage of local knowledge and capabilities of the affected population and organisations, including their problem-solving capabilities. Thus, despite the important role played by prepared and rehearsed response plans, there is an awareness that the unanticipated demands and problems that emerge during the response require an innovative response, that is, a response that was not planned for (Comfort, 2007; Lagadec, 2009). Because there is often a great urgency to act during a crisis, the emergency management literature commonly adopts the concept of improvisation to characterise this innovative response, stressing its collective nature (Frykmer et al., 2018; Kendra & Wachtendorf, 2007; Mendonca & Wallace, 2007; Roud, 2021; Webb & Chevreau, 2006).

More generally, the notion of collective improvisation has been extensively studied within organisation studies under the label of organisational improvisation (Ciuchta et al., 2021; Cunha et al., 1999; Hadida et al., 2015). A central aspect of improvisation is the intentionally unplanned nature of a novel action, that is, ‘the deliberate fusion of the design and execution of a novel production’ (Miner et al., 2001, p. 314). Improvisation typically has an implicit starting point as well as continuing touchstones that anchor and constrain action (Berliner, 1994; Miner et al., 2001; Weick, 1993). Miner et al. (2001) use the term ‘improvisational referents’ for these structural elements, and there may be more than one operating at the same time, such as an unexpected problem and a pre-existing behavioural routine or a plan.

To summarise the first component of our conceptual framework, a comprehensive emergency management model includes four phases, of which we focus on the response. The response is guided by a response plan that lays out the coordination among the different actors identified as involved in the response. During the response phase, unanticipated problems are likely to emerge, and these require collective responses that were not included in the plan. Due to high urgency, the responses are characterised by improvisation, although they are guided by the nature of the problems and the overall structure of the emergency response plan.

The second component of our conceptual framework is a conceptualisation of innovation governance. To reiterate McKelvey et al. (2020) and McKelvey and Saemundsson (2020), innovation governance is here conceptualised as the stimulation and regulation of the creation and use of new scientific and technical knowledge, where heterogeneous actors, each with their own goals and incentives, participate in collective action using a shared resource pool. Furthermore, this dynamic process involves interactions and feedback loops, where the actors need to learn and adapt their actions based on outcomes.

Innovation governance involves drawing knowledge from both science and technology, and many debates link science and technology as differing types of activities performed by different types of actors. One debate focuses on what boundaries exist between scientific and technological communities. This literature argues that, on the one hand, researchers in different types of organisations such as universities and firms are responding to differing incentive structures but that, on the other hand, hybrid communities, organisation, and incentive structures emerge over time to try to combine the two communities and their incentives and bodies of knowledge (Dasgupta & David, 1994; Gittelman & Kogut, 2003; McKelvey, 1996; Murray, 2010; Nightingale, 1998; Perkmann et al., 2019). Even so, assimilating knowledge across organisational boundaries requires some complementarities, in that firms which invest more in R&D and in open innovation are also more likely to draw from university science for their innovations (Laursen & Salter, 2004). Another debate builds upon Fleming and Sorenson (2004), who argue that distant search through science leads to more radical breakthroughs, whereas close search through technology leads to more incremental steps in independent components (Fabrizio, 2009; Nickerson & Zenger, 2004). Finally, a parallel debate focuses more on users, applications, and the marketplace rather than on scientific and technological knowledge per se. This includes the concept of users leading the generation of new ideas through user innovations (von Hippel, 2005) as well as technological exaptation, for example, how already existing traits are exploited for new purposes, with economic implications for creating new markets (Dew et al., 2004).

Drawing on these perceived theoretical differences between science and technology, we interpret that in relation to our empirical study, different types of actors need to be identified as they may specialise in the generation and use of scientific and technical knowledge and therefore may specialise in different types of activities related to innovation. An illustration related to our empirical context is an expectation that universities, firms, and university hospitals have different specialisations. Universities likely specialise in research and hence in solving generic theoretical problems, which may or may not have direct connections to specific practical problems or specific sets of users. Similarly, firms likely specialise in innovation, and they do so by solving specific practical problems experienced by a set of users and may, or may not, use knowledge generated by the science community. Still other actors, such as university hospitals, may combine the generation and use of scientific and technical knowledge and be involved in a broader set of activities related to using innovation in practice.

Summarising the second component, we conceptualise innovation governance as the dynamic stimulation and regulation of collective action related to the generation and use of new scientific and technical knowledge. The collective action is performed by a heterogeneous set of actors that may differ in terms of the degree to which they specialise in various activities related to innovation processes. We divide these activities, for analytical purposes, into the generation of scientific knowledge, the generation of technical knowledge, and the use of scientific or technical knowledge.

The third component of our conceptual framework is a conceptualisation of innovation as a problem-solving process whose aim is to introduce new ways of doing things. Innovation – articulated by Nelson and Winter (1982) as a change in organisational routine and by Schumpeter (1934) as carrying out new combinations – can be perceived as a process directed towards the ‘replacement of uncertainty by more secure yet fallible understanding’ (Metcalfe, 2016, p. xiii). This is especially true for innovation based on the creation and use of new scientific and technical knowledge, which is a process characterised by complexity and uncertainty because it involves multiple actors, and the outcomes and their benefits are not known in advance (Fagerberg et al., 2005).

Thus, we conceptualise innovations based on new scientific and technical knowledge as new ways of doing things emerging from an evolutionary process of problem solving (Clark, 1985; Consoli et al., 2016; Kline & Rosenberg, 1986; Thomke et al., 1998; Vincenti, 1990). These evolutionary processes are essentially knowledge processes but become innovation processes when they include actors who also use the newly generated knowledge to do things in a new way that is beneficial for them. In such problem-solving processes, repeated trials of potential solutions to a given problem are means, in conjunction with insights about where solutions can be found, to revise or refine the solutions to be tried out, or to revise or refine the problem.

In this article, we specify three dimensions as relevant for making distinctions between different types of problem-solving processes underlying a future innovation. First, we specify that the insights that guide actors when revising or refining problems or solutions can be either theory-driven or experience-driven (McKelvey & Saemundsson, 2020). Theory-driven insights are based on generalisable scientific knowledge that provides the equivalence of a map that can be used to predict the feasibility and efficacy of a particular set of solutions. Experience-driven insights, however, are based on previous experience with similar problems and may be personal and thus difficult to articulate. Second, we specify that trials can be both direct and indirect (McKelvey & Saemundsson, 2020). Direct trials take place in a real-world setting, for example, flying a full-sized prototype of an airplane, administering a new pharmaceutical to humans, or using real-time data when developing information systems. In contrast, when trials are indirect, a simplified representation of the real-world setting is used to test the expected efficacy of the proposed solution in an actual real-world setting. This is done, for example, when performing mathematical calculations or simulations to predict outcomes of the problem-solving process, investigating a miniature version of an airplane in an artificial wind tunnel, administering pharmaceuticals that are intended for humans to animals, or using predefined datasets when developing information systems. Finally, inspired by the emergency management literature reviewed above, we specify that problem-solving processes can be either planned or improvised, depending on whether they are designed before or during execution, respectively.

To summarise the third component of our conceptual framework, we perceive innovations based on new scientific and technical knowledge to be the result of an uncertain problem-solving process with repeated trials, the aim of which to introduce new ways of doing things. Innovations are thus broadly defined here. Moreover, we specify three dimensions as being particularly relevant for characterising different types of problem-solving processes, namely, 1) the degree to which they are theory-driven or experience-driven, 2) the degree to which they use direct or indirect trials, and, 3) the degree to which they are planned or improvised.

Figure 1 summarises our proposed conceptual framework for innovation governance during a crisis based on the three theoretical components described above.

Figure 1. Conceptual model of innovation governance during a crisis.

The left-hand side of Figure 1 denotes the actors involved in the specific innovation governance process analysed, including public, private, and civil society. The right side denotes the collective problem-solving activities related to the three, partially overlapping, aspects of the generation and use of new scientific and technical knowledge that are to be stimulated and regulated. The dynamic nature of the innovation governance process is denoted by arrows and feedback received by the actors from the outcome of the problem-solving processes. During a national crisis, collective problem-solving activities are triggered by unanticipated problems and guided by an emergency response plan. The success of their outcomes is determined by the degree to which they resolve the unanticipated problems and are put to regular use during the response.

Method

This section describes our methodological approach, data collection, and data analysis. To explore innovation governance during a crisis, we analyse the response to COVID-19 in Iceland during the first wave of infections in spring 2020.

This empirical context was selected for several reasons. First, the first wave of COVID-19 was a period of radical uncertainty when it came to the consequences of the disease and how it should be addressed. Thus, there was a strong need for new scientific knowledge about the nature of the virus and the disease as well as technical knowledge about how the disease could be addressed. Furthermore, it was urgent that this knowledge was put in use as soon as possible. Second, Iceland shares the institutional settings of many European countries, especially the Nordic countries, with public information laws making official documents public, and its relatively small size (360,000 inhabitants) making it easier to analyse the response in a holistic way. Third, in Iceland there exists a generic capacity to create and use new scientific and technical knowledge in the context of medical innovation. While Iceland lacks industrial capacity that is specific to infectious diseases, such as the development and manufacturing of vaccines, protective equipment, and diagnostic equipment, it has a relatively strong public university and health care system and some industrial capacity for medical innovation. Fourth, in Iceland, a diverse set of actors were mobilised, in both public and private sectors and in civil society, to assist with the COVID-19 response, and their participation went beyond what was prescribed by emergency management plans. Finally, Iceland performed well, compared to other European countries, when responding to the threats of COVID-19 (Coccia, 2022).

One of the authors was partly involved in the response in a consultancy role, allowing direct access to many of the key people and organisations analysed here. However, we have chosen to use these insights only as background information for understanding the empirical context of the key actors and activities. Our research is instead based upon our analysis of official documents, where Iceland – like most Nordic countries – has strong public information acts so that all material can be obtained, enabling us to conduct a systematic analysis of the emergency response.

Throughout the empirical analysis, the authors collaborated. One of the authors read and coded the material, which was in Icelandic, and prepared summaries in English, presented as tables, figures, and text. These were analysed and discussed with the second author on a weekly basis during approximately one year. Multiple sources allowed triangulation of our key analysis, and in those rare cases of conflicting information, official documents were given precedence in our interpretation.

The data preparation and analysis proceeded in five steps to explore innovation governance during the first wave of infections.

The first step was collection, chronology, and identification of key actors. Official documents about the COVID-19 response were collected to identify the actors that are part the COVID-19 response governance system (see Section 4), the COVID-19 response plan, and basic information about the response during the first wave of infections. The collected material is primarily official documents and websites, supplemented with published books, scientific publications, articles in professional magazines, and the daily press. In total, excluding material from the daily press, this material consisted of close to 2000 pages of text (see Table 1).

Table 1. Data sources on the first wave of COVID-19 in Iceland.

Type of material	Pages
Laws, statues and official announcements	229
Emergency management plans and related material	754
DCPEM status reports	285
Official reports and newsletters	86
Published books	490
Scientific publications	74
Professional magazines	9
Total	1.924

The second step was to use this material to identify the unanticipated problems to be solved during the crisis. Unanticipated problems were identified based on daily status reports from the Department of Civil Protection and Emergency Management (DCPEM), which were available for the period 28.01.2020 to 20.05.2020 (the first wave of infections). In total, we identified 20 unanticipated problems (see Table 2).

Table 2. Unanticipated problems during the first wave of COVID-19 in Iceland.

Unanticipated problems	Date first mentioned	Creation or use of new scientific and technical knowledge
Keeping track of Icelanders abroad that may need help getting back to Iceland	26.2.2020	No
Insufficient capacity to answer emergency phone number	2.3.2020	No
Capacity to analyse diagnostic tests	4.3.2020	Yes
Evaluating the diffusion of virus	6.3.2020	Yes
Registering voluntary back-up health professionals	10.3.2020	No
Potential shortage of testing pins used for diagnostics	17.3.2020	No
Registering voluntary back-up welfare workers	18.3.2020	No
Predict infections and load on health care system	19.3.2020	Yes
Maintaining sufficient capacity to trace infections	23.9.2425	(Yes)
Monitoring of COVID-19 patients at home and at hospital	19.3.2020	Yes
Results from diagnostics available in health portal	20.3.2020	Yes
Help automate tracing	25.3.2020	Yes
Experimental use of existing pharmaceuticals	27.3.2020	Yes
Establishing an outpatient clinic for COVID-19 patients	29.3.2020	Yes
Awareness of the effects of the pandemic on mental health	6.4.2020	No
Measuring antibodies	7.3.2020	Yes
Securing supplies of protective gear	8.4.2020	No
Regional diagnostic testing after a large group infection	15.4.2020	No
Funding of research project to investigate public health	15.4.2020	No

The third step was to examine each of these problems in-depth, to ensure that we understood the associated problem-solving processes. We searched for and obtained further information for each of the identified problems, for example, how the problem had been solved, the degree to which the solution was used, and how the solution had improved the response. The search included the official documents that had already been collected as well as new secondary data material, such as articles in the daily press, published books, and scientific and professional publications. In many cases this material included verbatim interviews, conducted during the response, with key participants in the response. As the focus of this study is on the creation and use of scientific and technical knowledge, the unanticipated problems whose corresponding problem-solving processes did not involve the creation or use of new scientific or technical knowledge were excluded from further analysis (see Table 2 above).

In the fourth step, we further analysed the processes that were initiated to solve those unanticipated problems and involved the creation or use of new science and technology. After multiple iterations, we coded the problems into three groups, based on the high-level problem they addressed: Knowing how to apply Non-Pharmaceutical Interventions (NPI); Knowing how to treat patients, and Maintaining response at scale (see Table 3).

Table 3. Unanticipated problems during the first wave of COVID-19 infections in Iceland coded into three groups of high-level problems whose solution involved the creation and use of new scientific or technical knowledge.

Unexpected problem or opportunity	Code
Evaluating the diffusion of virus	Knowing how to apply NPI
Using prediction models to predict infections and load on health care system	Knowing how to apply NPI
Measuring antibodies	Knowing how to apply NPI
Monitoring of COVID-19 patients at home and at hospital	Knowing how to treat patients
Experimental use of existing pharmaceuticals	Knowing how to treat patients
Capacity to analyses diagnostic tests	Maintain response at scale
Maintaining sufficient capacity to trace infections	Maintain response at scale
Communicate results from diagnostic tests	Maintain response at scale
Establishing an outpatient clinic for COVID-19 patients	Maintain response at scale

The fifth step involved the more detailed analysis, reported in Section 4.3 below. Each problem in Table 3 was analysed using our conceptual framework, regarding the actors involved in the problem-solving process, stimulation and regulation, the nature of the problem-solving process, the outcomes, and the extent to which the outcomes solved the unanticipated problem and the degree to which the solutions were used. During the analysis, some of the originally identified problems were broken down into sub-problems that were addressed by different sets of actors or that made use of different types of problem-solving processes.

Results

The results are presented in three parts. First, we describe the emergency management plan on which the response to the COVID-19 pandemic in Iceland was based, focusing on its governance structure. Second, we provide an overview of the main events of the emergency response during the first wave of infections in spring 2020. Finally, we present the results of our analysis of how scientific or technical knowledge was created and used to solve unanticipated problems during the response.

Emergency response planning of COVID-19 in Iceland

The emergency response planning of pandemics in Iceland is based on the International Health Regulations (IHR) adopted by the World Health Organization (WHO) in 2005 and entered into force in 2007. This international law, the objective of which is to limit the spread of health risks to neighbouring countries while at the same time preventing unwarranted restrictions on trade and travel, requires countries to build capabilities to detect acute public health threats in a timely manner, assess public health events and report to WHO those that may be of international concern, and to respond to public health risks and emergencies. In spring 2006 the Department of Civil Protection and Emergency Management (DCPEM), on behalf of the National Commissioners Office (NCO), and the Chief Epidemiologist (CE), on behalf of the Directorate of Health (DH), started work towards complying with the revised IHR, including specifying necessary changes to existing legislation and preparing response plans (DCPEM, 2008; DH, 2007a, 2007b; WHO, 2021a).

An outline of the governance system of the national response to the COVID-19 pandemic in Iceland is presented in Figure 2. The system is based on two principles. First, the system should support the coordination of action by two branches of government: the Ministry of Health and the Ministry of Justice. On their behalf, national actions are coordinated by the Chief Epidemiologist, whose office is a part of the Directorate of Health, and the manager of the Department of Civil Protection and Emergency Management (DCPEM), which is a part of the National Commissioner’s Office. Second, the system should promote national and regional coordination. The country is divided into seven regions, each with a regional epidemiologist appointed by the Minister of Health and a police chief appointed by the Minister of Justice that are responsible for coordinating the regional response in cooperation with the Chief Epidemiologist and the manager of the DCPEM.

Figure 2. Governance system of the national response to the COVID-19 pandemic in Iceland (based on DCPEM (2020a, 2020b), PI (2020a), and PI (2020b)).

In addition to the Cabinet, three councils are responsible for strategic direction when preparing the response and providing guidance during an emergency: the National Security Council, which coordinates action between the executive and legislative branch of government, and the Epidemiology Council and Civil Protection and Security Council, which are councils of experts connected to the Chief Epidemiologist and DCPEM, respectively.

Furthermore, many public and private organisations have an explicit role in the response and are required to prepare and execute their own response plans that are aligned with the national plan. These include 32 national organisations (three health services organisations) and 16 types of organisations that operate at the regional level (five types of health services organisations). For our purposes, the most important of those organisations is the National University Hospital. Within the hospital, the response is led by a Pandemic Committee, which operates on behalf of the response management team of the hospital and is directly in contact with the Chief Epidemiologist.

In addition to spelling out the set of actors who need to prepare for an emergency and whose activities are to be coordinated during the response, the plan specifies three levels of activation. At the lowest level (‘Alert’), there is a suspicion that an imminent event may threaten public health. Surveillance efforts are increased as well as communication within the governance system. Actors in the system review their plans and update as needed. At the middle level of activation (‘Danger’), an event has occurred that is likely to threaten public health, and the process that started at the ‘Alert’ level has higher priority. At the highest level of activation (‘Emergency’), it has been confirmed that the event that has occurred will threaten public health. Within the governance system, the response now has top priority and all other activities are set aside (DCPEM, 2020a).

The emergency response plan for COVID-19 in Iceland emphasises information sharing and coordination among a wide set of actors during the response, reflecting awareness of the inherent uncertainty of the emergency and the need for an adaptive response. However, the plan includes some of the organisations that have a key role in regulating health care services, medical research, and medical innovation, but not others. Included are the Directorate of Health (regulating health services providers and physicians) and the Icelandic Medicines Agency (regulating pharmaceuticals and medical devices), but the National Ethics Committee is excluded. In addition, organisations that are traditionally involved in medical research and innovation, such as universities and technology-based firms, do not have an explicit role in the pandemic emergency response planning. Thus, even if the emergency response planning for COVID-19 has prepared actors for the inherent uncertainty of the emergency and the need for adaptation during the response, it only partially addresses innovation governance in the context of medical research and innovation.

Overview of the main events of the emergency response

In early January 2020, news spread around the world about a new virus detected in humans, later to be named Severe Acute Respiratory Coronavirus 2 (SARS-CoV-2) and whose infectious disease became known as COVID-19 (WHO, 2020). On January 23^rd, the office of the Chief Epidemiologist and the Department of Civil Protection and Emergency Management in Iceland started to prepare for the activation of emergency response plans (DH, 2020). Figure 3 provides an overview of the main events of the emergency response from January 27^th, when the plans were activated at the ‘Alert’ level, to May 25^th, when the activation level of the COVID-19 response was changed from ‘Emergency’ to ‘Danger’ (DCPEM, 2020c), signifying the end of the first wave of the pandemic.

Figure 3. Overview of main events of the emergency response during the first wave of COVID-19 infections in Iceland.

Following the activation of the response plan, daily coordination meetings were held, coordinating actions between the two branches of government (Department of Health and the Department of Justice) and within each branch. The centre of the coordination was at the level of the Chief Epidemiologist and the manager of the Department of Civil Protection and Emergency Management.

Diagnostic testing started at the National University Hospital on February 1^st. When the first infection was detected on February 27^th, 65 individuals had previously been tested negative. Once the first infection was detected, the activation level was set to ‘Danger’ and a special team was set up by the police to perform contact tracing. The planned response was to isolate infected individuals and quarantine those who had been in close contact with the infected individuals according to the results from contact tracing (DCPEM, 2020b, 2020c).

On March 6^th, the first infections in Iceland were confirmed. At that time, around 400 individuals had previously been tested, of which 43 had been positive and were in isolation. Additionally, 326 were in quarantine. On March 6^th alone, 70 individuals had been tested, and efforts were made to increase the testing capacity at the National University Hospital. On March 6^th, the activation level was raised to its maximum (‘Emergency’), and on March 7^th, the first of a series of daily briefings for the media and the public was broadcast on public television (DCPEM, 2020c).

On March 13^th, the Icelandic government issued restrictions for the general public. As of March 16^th, gatherings of more than 100 people were banned, and social distancing of at least 2 metres was to be maintained. Teaching at universities and secondary schools was suspended, but primary schools and kindergartens remained open. International travel was discouraged. On the same day, the website covid.is was launched. The website provided information and guidelines related to the pandemic, including statistics. On March 23^rd, further restrictions were imposed. Gatherings of more than 20 people were banned. Many facilities and services, such as swimming pools, fitness centres, hair salons, and museums, were closed. At this point, 659 persons had been infected (DCPEM, 2020c).

The peak in the number of infections during the first wave occurred at the end of March and the beginning of April, with around 100 new infections each day and around 1000 active infections in total. On May 5^th, some restrictions were removed; for example, teaching at universities and secondary schools was resumed, gatherings of fewer than 50 people were allowed, and elective surgeries were resumed. Restrictions were further reduced on May 18^th, and on May 25^th, the activation level of the COVID-19 response was set to ‘Danger’ and the daily briefings on public television were discontinued, which signalled the end of the first wave (DCPEM, 2020c).

Solving unanticipated problems during the emergency response

We now analyse in more detail, using our conceptual framework, the unanticipated problems that were addressed during the response through the creation and use of new scientific or technical knowledge. Table 4 summarises the results by providing information about the high-level problem being addressed, the problem owner, leader of the problem-solving activities (PS leader), other actors involved, type of activity (science, technology, use), the characteristics of the problem-solving processes (theory- or experience-driven, direct or indirect trials, planned or improvised), types of outcomes, regulatory issues, and the extent to which the outcomes were used.

Table 4. Summary of the results from the analysis of unanticipated problems.

High level problem	Unanticipated problem	Sub-problem	Problem owner	PS leader	Other actors	Activities	PS characteristics	Outcomes	Regulatory issues	Extent of use
Knowing how to apply NPI	Evaluating the diffusion of virus	Random testing	Chief epidemiologist	deCODE Genetics	University Hospital	Use-Science	theory-driven; indirect; planned	Scientific results;	Ethical approval	Temporary
Knowing how to apply NPI		Voluntary testing	Chief epidemiologist	deCODE Genetics	University Hospital	Use-Technology	experience-driven; direct; improvised	Information; Organization; IT system	Ethical approval; Data privacy	National system
Knowing how to apply NPI		Sequencing of virus	Chief epidemiologist	deCODE Genetics		Use-Technology	experience-driven; direct; improvised	Information; Organization; IT system	Data privacy	National system
Knowing how to apply NPI	Predict infections and load on health care system	Infections and load – Based on public data	Chief epidemiologist	University of Iceland		Use-Technology	theory- and experience- driven; direct; improvised	Predictions; Prediction model		National system
Knowing how to apply NPI		Load – based on hospital data	University Hospital	University of Iceland		Use-Technology	experience-driven; direct; improvised	Predictions; Prediction model	Data privacy	National system
Knowing how to apply NPI	Measuring antibodies		Chief epidemiologist	deCODE Genetics		Use-Science	theory-driven; indirect; planned	Scientific results;	Ethical approval	Temporary
Knowing how to treat patients	Monitoring of COVID patients at home and hospital		University Hospital	University Hospital		Use-Research-Technology	experience- and theory-driven; direct; improvised and planned	Information; Scientific results; Treatments	Ethical approval; Approval of treatments Data privacy	National system
Knowing how to treat patients	Experimental use of existing pharmaceuticals		University Hospital	University Hospital		Use-Research-Technology	experience- and theory-driven; direct; improvised	Information; Scientific results; Treatments	Ethical approval; Approval of treatments	National system
Maintain response at scale	Capacity to analyse diagnostic tests		University Hospital	University Hospital		Use-Technology	experience-driven; direct; improvised	Equipment; IT system	Data privacy	National system
Maintain response at scale	Maintaining sufficient capacity to trace infections	Manual tracing	Police	Police	deCODE Genetics	Use-Technology	experience-driven; direct; improvised	Organization; Facilities; IT system	Data privacy	National system
Maintain response at scale		Automatic tracing	Police	Industry consortium		Use-Technology	experience-driven; indirect; planned	Information; Smartphone application	Data privacy	Installed by many, but not helpful
Maintain response at scale	Communicate results from diagnostic tests		Directorate of Health	Directorate of Health		Use-Technology	experience-driven; direct; improvised	IT system	Data privacy	National system
Maintain response at scale	Establishing an outpatient clinic for COVID-19 patients	At hospital	University Hospital	University Hospital		Use-Technology	experience-driven; direct; improvised	Organization; Facilities; IT system	Data privacy	National system
Maintain response at scale		Monitoring patients at home	University Hospital	Sidekick Health	CCP	Use-Technology	experience-driven; indirect; planned	Smartphone application	Data privacy	Not implemented

As shown above in Table 4, we identified 14 unanticipated problems, including sub-problems, which we categorised into three groups based on the high-level problem addressed. These are: Knowing how to apply Non-Pharmaceutical Interventions (NPI) (six unanticipated problems), Knowing how to treat patients (two unanticipated problems), and Maintaining response at scale (six unanticipated problems).

For the six unanticipated problems related to knowing how to apply NPI, the innovation governance focused on the Chief Epidemiologist, who was the owner of all but one of the problems. All six problems related to how to collect information about, or predict, the state of the disease. As a result, the primary outcomes of the problem-solving activities were information and information technology systems (IT systems) as well as the organisation needed to collect and distribute information across different actors. In most cases the problem-solving processes were repeated improvised experience-driven direct trials, but such processes were occasionally complemented by planned theory-driven indirect trials, such as doing diagnostic testing of a random sample of the population.

Despite being the problem owner, the Chief Epidemiologist was not leading the problem-solving activities. Instead, they were led by the private company deCODE genetics in four cases and by the University of Iceland in two cases.

In early March, the CEO of deCODE genetics, which is a subsidiary of Amgen Inc., contacted the Chief Epidemiologist and offered to provide voluntary diagnostic testing to help estimate the diffusion of the virus. The testing would be an addition to the testing of symptomatic persons and those in contact with infected persons, which was already being performed by the primary care system and the National University Hospital. Subsequently, the company also tested random samples of the Icelandic population to provide more systematic evidence of the diffusion of the virus, supplied the facilities to sequence the virus genome of all persons infected, and performed antibody measurements. Moreover, deCODE genetics collaborated with the University Hospital, which was also involved with analysing diagnostic tests.

The involvement of deCODE genetics in diagnostic testing and handling of patient data led to some regulatory issues. Questions were raised, especially at the University Hospital, about the company’s motives, that is, whether it might be using the situation to gain access to valuable data that would be used for commercial ends. Furthermore, the Scientific Ethics Committee argued that the company’s involvement needed ethical approval from the committee as it was a part of a scientific study, and the Data Protection Agency complained about the company’s collection of personal health care data when performing diagnostic testing and virus sequencing. However, the Chief Epidemiologist argued that the company’s contribution was very important, and its participation in the response continued.

Also in early March, two groups at the University of Iceland were asked if they could help predict the progression of the pandemic in Iceland and the resulting demand it would place on the health care system. The Chief Epidemiologist, on behalf of the Directorate of Health, asked a research group at the Health Science Institute to help predict the number of individuals diagnosed with COVID-19 and the corresponding burden on the health care system. In addition, the National University Hospital asked a group of researchers in Industrial Engineering if they could help predict the flow of COVID-19 patients at the hospital. Both groups updated their models and predictions throughout the first wave.

For the two anticipated problems related to knowing how to treat patients, innovation governance was mostly within the University Hospital, which was the problem owner as well as the leader of all problem-solving activities. The primary outcome of these problem-solving activities was information and scientific results about the disease and its progression as well as treatments that were systematically applied given diagnostic information about patients. The problem-solving activities were highly influenced by scientific and clinical reports from abroad, and also by information collected through the telemonitoring operation of the hospital and clinical experience from inpatient and outpatient treatments. Thus, it included a very broad range of problem-solving processes, which commonly characterises the intersection of science, technology, and use at a university hospital. The stimulation and regulation of these processes were primarily internal to the University Hospital. The telemonitoring of patients was initiated by individual physicians and later embraced by hospital management. The use of personal health care data and the evaluation and selection of treatment alternatives were regulated by routines that are part of the daily operation of the hospital.

For the six problems related to maintaining response at scale, innovation governance was more complex in that there were multiple problem owners and problem-solving leaders. This may have to do with the nature of the problems. Two of the problem-solving activities, the establishing of an outpatient clinic and monitoring patients at home, were part of the scaling up of the telemonitoring activities put in place to help medical professionals know how to treat patients. The other four were concerned with the scaling up of the collection and distribution of information related to tracing and diagnostic testing. In all six cases, the problem-solving processes were experience-driven. In four cases they were further characterised as direct improvised trials, and in two cases as indirect planned trials. The primary outputs were information systems and the information they generated. As for any information system collecting personal information, the main regulatory issue was data privacy.

Most of the outcomes listed in Table 4 were put to regular use as parts of the national system for diagnosing and treating COVID-19 that was in place at the end of the first wave of infections. At that time, there was a centralised IT system for managing diagnostic testing, sequencing of virus genomes, and contact tracing. The system provided daily information that was used by the Chief Epidemiologist and the University Hospital to evaluate the state of the pandemic and predict its progression and associated load on the health care system. This information was also communicated to the public and used to justify decisions about the timing and extent of restrictions. The use of this system to evaluate the diffusion of the virus and predict infections and the load on the health care system was highly successful. The information and predictions were regularly communicated at the televised daily briefings and on the website www.covid.is and were used by the Chief Epidemiologist and the government to justify the selection and timing of NPIs. The success of this system and its use is further reflected by the high share of the population (96%) that reported at the end of the first wave that they had high confidence in the government’s ability to handle the COVID-19 crisis. Furthermore, the University Hospital had formally established its telemonitoring operation as a COVID-19 outpatient clinic and integrated it into the hospital’s IT and organisational structure to assist in patient evaluation and selection of treatment. The systematic collection of information on symptoms and various physiological parameters made possible by the telemonitoring operation, and the specialised outpatient clinic was judged by the medical community as instrumental in reducing the need for hospitalisation and the number of deaths during the first wave.

In four cases, outcomes were not put to regular use during the first wave of infections. In two cases (random diagnostic testing and measurement of antibodies) the outcomes were temporary and scientific in nature and provided important input into decision making at a particular point in time during the response. In the other two cases (automatic tracing and monitoring patients at home) the outcomes were intended to become a part of the national system but failed to do so for different reasons. On the one hand, a smartphone application to help automate tracing was developed in record time by a consortium of highly competent industry volunteers and passed all regulatory standards of data privacy. The application was promoted by the Chief Epidemiologist and widely used by the public but was of limited help to the tracing team due to technical limitations. On the other hand, a smartphone application to assist with the daily monitoring of patients at home was based on an existing platform created by the company Sidekick Health, which specialises in digital therapeutics. The idea was that the application would allow each patient to self-report his or her status and make it available in the hospital systems, reducing the workload of health care professionals. The application was developed in record time with assistance from the gaming company CCP and connected to the hospital’s IT system, but it was never put to regular use due to resistance from the hospital nursing staff. It is noteworthy that despite the apparent different reasons for failure, the problem-solving processes applied in both cases stand out as the only ones characterised by planned indirect trials, which indicates that access to real-time information obtained through direct trials was important for developing useful solutions.

Revisiting the conceptual framework

Having described the emergency response planning for COVID-19 in Iceland, provided an overview of major event of the response during the first wave of infections, and analysed how unanticipated problems were solved during that time, we return to our conceptual framework presented in Section 2 and visualise the updated results in Figure 4.

Figure 4. Innovation governance during the emergency response to COVID-19 in Iceland.

The left side of Figure 4 includes the actors we have identified as involved in innovation governance during the crisis. Many of them (ministries, national coordination, police, health services, public) were formally a part of the governance system of the national response to the COVID-19 pandemic (see Figure 2 in Section 3) while others (universities, firms) were not.

The right side of Figure 4 summarises collective problem-solving processes that were stimulated and regulated during the crisis. We grouped the unanticipated problems that emerged during the crisis and triggered problem-solving into three high-level problems, namely, 1) Knowing how to apply NPI, 2) Knowing how to treat patients, and 3) Maintaining response at scale. The first two problems were concerned with reducing the radical uncertainty that characterised the early phase of the pandemic, and the outcome was primarily in the form of scientific results, information, and predictions. The third problem was primarily concerned with the scaling up of an interconnected information system for collecting information that was used to monitor the state of the pandemic, learn about the progression of the disease, treat patients, and predict the development of infections and the associated load on the health care system. Hence, decision making during the pandemic (not only the first wave), both related to the application of NPIs and treating of patients, was based on the information collected and distributed by this system.

We find the focus of problem-solving activities during the crisis to be at the intersection between the generation and use of new knowledge, as visualised in Figure 4. For solving problems related to knowing how to apply NPI (A), the activities were focused on the generation and use of either scientific or technical knowledge, whereas for solving problems related to maintaining response at scale (C), the activities were exclusively focused on the intersection between technical knowledge and use. Finally, for solving problems related to how to treat patients (B), the activities were exclusively focused on the intersection between the generation and use of scientific and technical knowledge.

Our finding that problem-solving activities take place at the intersection between the generation and use of knowledge may not be surprising given the urgency of the response and its reliance on improvisation. However, despite the urgency, the response lasted several months, which allowed for repeated trials, and we found that a broad range of problem-solving processes were used, resulting in new ways of doing things (innovations) that were put to regular use. The repeated trials provided feedback in the form of outcomes or new problems that were related to the previous ones (shown at the bottom of Figure 4). Thus, we find sequential problem-solving processes, in some cases consisting of multiple improvised direct trials, but also improvisation processes that were complemented by planned indirect trials, and even new problem-solving routines that were created and repeatedly applied, as in the case of the outpatient clinic (evaluating and treating patients). Furthermore, improvisation processes were mostly using direct trials, but like the planned processes, they were based on both theory-driven and experience-driven search.

Finally, in this study, we added insights about the stimulation and regulation of problem solving in the context of an emergency response plan (see middle of Figure 4). As expected, the actors belonging to the governance system of the national response played a key role as owners of the emerging unanticipated problems while at the same time being responsible for coordinating the response. Thus, they both stimulated and regulated the problem-solving activities that were initiated to address these problems. Furthermore, they judged the efficacy of the problem-solving processes, and the selective retention of their outcomes, by how well they aligned with the objectives of the emergency response plan and the degree to which they improved the effectiveness of the response. However, we also find that other actors – especially firms and universities – that were not formally a part of the emergency management system also played an important role, even if their participation was controversial. However, their role was different from what would be expected, and the extent of the use of outcomes was mixed.

In general, firm participation in the response was contested, especially the participation of deCODE genetics, which is a science-based firm. Firms’ motives for participation were questioned, given that profits are expected to be their primary incentive, even if firms were seen to provide valuable contributions to the response. Additionally, in the case of deCODE genetics, the firm was involved in problem-solving processes normally associated with universities and research organisations (planned, theory-driven, indirect trials), the results of which they published in academic journals, as well as problem-solving processes associated with firms (improvised/planned, experience-driven, direct trials). In contrast, the University of Iceland, whose participation was not contested, was involved in problem-solving processes normally associated with firms, and not in problem-solving processes normally associated with universities and research organisations.

Finally, the success of the outcomes of the problem-solving processes led by firms and universities varied. In the case of the University of Iceland, the outcomes were well aligned with the emergency response plan and were thus beneficial for the response. For deCODE genetics, they provided important scientific results at several points during the response, and their participation played a pivotal role in creating and operating an extensive information system that was repeatedly used for diagnostic testing and distribution of information among different stakeholders. Efforts made by other companies to assist in the response were not as successful, despite impressive pooling of talent from different firms and extremely short development times. In one case, the failure was due to technical issues, and in the other case, it was due to staff resisting implementation. In both cases, however, the problem-solving processes were characterised by planned indirect trials in contrast to the more successful problem-solving processes at the intersection between technology and use, which were characterised by improvised direct trials.

Conclusions and future research

The purpose of this article is to improve our understanding of innovation governance by exploring the generation and use of new scientific and technical knowledge to address an urgent societal crisis. We provide a conceptualisation of innovation governance during an emergency response and apply it to empirically analyse the response to COVID-19 in Iceland during the first wave of infections in spring 2020.

Our first set of conclusions is concerned with problem solving related to science, technology, and innovation during a crisis. We conceptualise innovations based on the creation and use of new scientific or technical knowledge as it emerged from evolutionary processes of problem-solving (Clark, 1985; Consoli et al., 2016; Kline & Rosenberg, 1986; Thomke et al., 1998; Vincenti, 1990), aiming to introduce new ways of doing things. Furthermore, we identified three dimensions as relevant for making distinctions between different types of problem-solving processes in this context, two of which are based on previous research in innovation studies (theory- or experience-driven search, direct or indirect trials), and one of which is based on the emergency management literature (planned or improvised trials).

Based on the results from this study, we argue that even if many problem-solving processes during a crisis are improvisations based on existing knowledge and capabilities, some of these improvisations are complemented by the generation of new scientific and technical knowledge and become the sources of successful innovation. Thus, the improvisation processes triggered by unanticipated problems during a crisis can, in some cases, be interpreted as blind variations (Campbell, 1987) that are retained and diffused through a series of complementary problem-solving processes and, eventually, through organisational routines (Nelson & Winter, 1982).

Even if the generated innovations were successful in addressing unanticipated problems at required scale during the crisis, one can question whether the solutions on which they were based meet the standards set for evidence-based medicine, patient safety, and data privacy during normal times and are thus relevant and useful in the long term and once the crisis is over. An interesting avenue for further research would therefore be to better understand how innovations that are generated and used early during a crisis, and the scientific and technical knowledge on which they are based, shape the long-term response to a prolonged crisis and the degree to which they create opportunities for further innovations once the crisis is over.

Our second set of conclusions is concerned with how the results inform our understanding of the concept of innovation governance readiness. The concepts of technological readiness (Mankins, 2009), institutional readiness (Webster & Gardner, 2019) and innovation governance readiness (McKelvey & Saemundsson, 2021) have been proposed as measures of how ready a particular medical technology is for general use in clinical practice. In this paper we extend McKelvey and Saemundsson (2021) definition of innovation governance readiness by focusing on problem-solving in relation to the activation of an innovation governance system aiming to solve a particular problem. In this way we arrive at a more general notion of innovation governance readiness that can be applied to evaluate both the readiness of a new technology to be exploited in clinical practice, as in the original concept of technological readiness, and the readiness to address unanticipated problems during a crisis.

Furthermore, we argue that this extension provides a more dynamic notion of innovation governance readiness related to technological opportunities. The ability to address unanticipated problems during the evolution of a new technology is an important part, operating in the short term, of a more long-term capability to exploit a new technology. So, like Mazzucato et al. (2021), we suggest that short-term dynamic capabilities, such as those called for during a crisis, are an important part of a more long-term capacity for innovation. Unlike Mazzucato et al. (2021) we stress the role of the private sector, especially in the short term, as we see from the results of this study. This is because private companies are likely to have knowledge and capabilities that do not exist in the public sector, which may be useful to reduce uncertainty and initiate collective and complementary problem-solving processes in response to unanticipated problems that will generate successful innovation. However, we also note from the results of this study that having superior knowledge and capabilities does not guarantee successful innovation but needs to be complemented by access to direct trials, but such access, especially for private firms, may be highly contested and problematic for governments to provide (Boersma et al., 2022).

Thus, we conclude that having a high level of innovation governance readiness is based on two components. The first is the capability to activate and maintain – through stimulation and regulation – collective problem-solving activities by public, private, and civil society actors. The second is the capability to apply a broad set of problem-solving processes when addressing problems or pursuing opportunities. While innovation governance readiness is always related to a particular type of problem, such as pandemics, or a specific technology, such as regenerative medicine, an interesting opportunity for further research is to investigate how readiness in one technology or problem can be transferred to other areas. Such transfers may be related to the complementary roles that different types of problem-solving processes have in generating ideas for innovation and validating that the ideas meet the standards set by the professional community using the innovation (e.g., evidence-based medicine) and by regulatory authorities (e.g., patient safety and data privacy) (McKelvey & Saemundsson, 2021). Another opportunity for further research is to consider how the concept of innovation governance readiness can be integrated into emergency management, for example, how it can be integrated into the comprehensive emergency management model (National Governor’s Association, 1979; Phillips et al., 2017) to better prepare for recurring types of crises, such as pandemics. Given the prevalence of digital technologies and the importance of real-time information for coordination of the response, more research is needed about the governance structures that enable the sharing of information among public and private actors participating in the response.

Our final set of conclusions is concerned with what we learn about innovation governance when specifically considering short-term responses during a crisis. Previous innovation studies have mostly been concerned with the long-term aspects of stimulating and regulating innovation based on science and technology focusing on interactive learning among multiple types of actors using the lens of innovation systems (c.f. Edquist & McKelvey, 2000) or innovation governance (c.f. Borrás & Edler, 2014; Kuhlmann & Ordónes-Matamoros, 2017) and stressing path dependency and institutional lock-in. We propose that the approach taken in this paper, including the extensions to the concept of innovation governance readiness, is not applicable only to short-term responses to crisis but also to short-term responses to opportunities brought forward by breakthroughs in science and technology.

Finally, an interesting avenue for further research would be to apply the concept of innovation governance readiness to a broader set of empirical contexts to further develop and test its wider applicability and usefulness. For example, the study of the COVID-19 response is a study of a particular type of innovation process that operates in a non-market selection environment where there is limited competition among actors. Other societal crises, such as the climate change crisis, may have longer durations than pandemics. For instance, the U.S. Office of Scientific Research and Development played a major role in the governance of technical innovations during World War II (Gross & Sampat, 2020). Their impact in a relatively short period was characterised by the generation and use of new scientific and technical knowledge in response to unanticipated problems but in turn created a broad set of long-term opportunities for innovation in society.

Acknowledgments

The research reported in this paper was funded by the Swedish Research Council VR DNR 2017-03360 grant, awarded to Professor McKelvey for the Distinguished Professor Program “Knowledge-intensive Entrepreneurial Ecosystems: Transforming society through knowledge, innovation and entrepreneurship”

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work was supported by the Vetenskapsrådet [VR DNR 2017-03360].

Notes

1. Non-Pharmaceutical Interventions are measures, other than the use of vaccines, that are taken to reduce the transmission and spread of an infectious disease. These measures include personal hygiene, social distancing, and limitations on movement (WHO, 2006).