Pandemic preparedness systems and diverging COVID-19 responses within similar public health regimes: a comparative study of expert perceptions of pandemic response in Denmark, Norway, and Sweden

Strengthening pandemic preparedness systems is central to reduce the impact of pandemics on vital societal functions [28]. In many countries the COVID-19 pandemic has unsettled belief in the ability of existing preparedness principles and institutions to properly mitigate global health emergencies. Recent studies have highlighted significant variations in how governments have understood and acted upon the pandemic to minimize its consequences, resulting in large national differences in mortality rates and economic repercussions [10]. Importantly, such policy divergences have crystallized despite the World Health Organization’s (WHO) efforts to strengthen and harmonize national pandemic preparedness since the late 1990s [55]. The inadequacy of prior assessments of pandemic preparedness in predicting COVID-19 efficacy has already been documented, such as in the case of the 2019 Global Health Security Index (GHSI), which ranked the US and the UK number one and second in the world in relation to health security capabilities prior to the pandemic [33]. The predictions of rankings such as the GHSI have since received widespread criticism as a result of the US and the UK’s initial COVID-19 responses [18], displaying, in turn, a number of issues in the organization of established pandemic preparedness systems [19].

A starting point for exploring these issues is to compare how globally configured pandemic preparedness systems governed and coordinated through the WHO’s 2005 International Health Regulations were activated in countries with similar public health regimes [27, 35]. Recent studies have shown how countries with the same public health regime responded differently to the COVID-19 pandemic during its first wave despite similarities in health expenditures and health infrastructure [7, 10]. Such within regime variation has fertilized the ground for discussions about what types of public health regimes (democratic or authoritarian), and formal political institutions (federalism or presidentialism) best determine COVID-19 response and performance [27, 37]. While these determinants are highly useful for discussing the macro-politics of global health emergency response, they, however, provide little guidance for understanding COVID-19 policy divergence between countries belonging to the same public health regime.

A useful point of departure for understanding such divergence among most similar countries are the Nordic welfare states’ different responses to the COVID-19 pandemic. Denmark, Norway, and Sweden engaged differently with the pandemic during its first wave despite sharing Social Democratic and Scandinavian welfare state models, health care organization, geography, corporatist traditions, levels of institutional trust, and high rankings in the GHSI (Denmark 8; Norway 16; Sweden 7) [11, 48]. Differences in the use of lockdown measures, face masks, testing, and COVID-19 response strategies raises the question of how countries belonging to similar public health regimes could respond so differently to the same biological threat.

In studying the policy process, policy studies have long recognized expert-based information as one of the most important factors for explaining policy outcomes [64]. More than just information itself, research in the tradition of Policy Advisory Systems (PAS) have underlined the importance of understanding the sources of policy advice, both within and outside bureaucracies, as always operating within particular ‘systems’ of policy advisors that vary between jurisdictions and change over time [14, 31]. In this framework, policy influence has been conceptualized as a function of location within the system and the type of advice being offered [15]. Recent theoretical developments in the ‘second wave’ of PAS has transposed this framework to better grasp the ‘context’ and ‘content’ of policy advice [16]. Particularly, issues of ‘context’ have led to an incorporation of key insights from the policy subsystems literature, shifting, in turn, analytical emphasis to the policy subsystem context in which advisory systems are situated. The basic idea is that the structure of policy subsystems, defined as the decision-making networks structured around policy issues, affect the influence and access of various sources of policy advice in important ways (ibid., [64]).

The importance of policy advice is particularly salient in the policy subsystem of pandemic preparedness, where research on public health security have long emphasized the centrality of experts in the design and implementation of pandemic preparedness systems [39]. Here, effective policy making in response to complex health crises is highly dependent on expert-based advice and information [53]. Experts occupy influential positions within the subsystem, such as in public health agencies and advisory committees where they are involved in framing biological threats and defining policy measures to respond to them [46, 47, 54]. This makes them important gatekeepers for understanding variation between the otherwise esoteric, technical, and siloed national policy subsystems activated during the COVID-19 pandemic [39, 52].

Studies of responses to prior pandemics such as H1N1 in 2009 have highlighted the importance of expert norms for understanding different national responses to health emergencies [3, 4, 52]. Here, expert advisors enacted ideas about the pandemic threat, settled during preparedness planning and from past experiences, which shaped policymaking. Despite public health policy in the Nordics traditionally being described as what might be called a ‘unitary subsystem’, with centralized authority in public health agencies, dominant policy images coordinated through the WHO and high intra-coalition belief compatibility [64], responses to the first wave of the pandemic saw conflicting expert visions and advice for the most appropriate national response strategy to the COVID-19 pandemic. In Sweden, in particular, these disagreements spilled out onto the pages of The Lancet [12, 25] and in public signature collections [17], pitting experts sympathetic or critical to the policies of the Swedish health authorities against each other. Globally, similar diverging positions of policy advice further crystallized with the publications and signature collections of the ‘Great Barrington Declaration’ and the ‘John Snow Memorandum’, which promoted opposite strategies to the application of non-pharmaceutical interventions like societal-wide lockdowns [9]. The existence of conflicting COVID-19 policy advice, in both the general public and in specialized research outlets, raises the question of the extent to which pandemic response divergences between countries belonging to the same public health regime are the result of differences in the ideas, attitudes, and policy preferences of national expert groups.

In light of these developments, this paper asks how countries within the same public health regime diverged in policy response to the COVID-19 pandemic and, given the central role of experts and expert-based information as a potential source of policy divergence, to what degree this divergence reflects important differences in the policy preferences within and between national expert groups?

To answer the research question, the paper employs a structured case comparison of the three Nordic countries Denmark, Norway, and Sweden, drawing on a mixed methods design. After presenting the methods in more detail, the first part of the paper compares the three countries’ politico-administrative pandemic preparedness systems and their responses to the first wave of the COVID-19 pandemic. The second part of the paper investigates to what degree these responses had support among and reflected dominant ideas held by national experts varying in disciplinary backgrounds and locations in the policy advisory system of pandemic preparedness. To do so, the paper presents and analyzes the results of an interdisciplinary and cross-country expert survey completed by 232 respondents from November–December 2020. It does this by, first, analyzing expert perceptions of national responses to the pandemic using a multidimensional response measure. Second, it applies an ordinal regression model to analyze to what degree experts’ attitudes to their respective governments efforts to save lives of citizens are associated with ideational commitment among experts to ‘focused protection’ or ‘community protection’ while controlling for the locational characteristics of experts in the policy advisory system. The analysis is followed by a discussion of the relationship between dominant ideas in national expert groups and perceptions of COVID-19 responses.

The paper makes a substantive contribution to the existing literature on public health security in general, and the COVID-19 pandemic in particular, by examining and comparing attitudes and policy preferences for national pandemic responses among national experts belonging to similar public health regimes. When controlling for locational positions in politico-administrative systems, disciplinary affiliations, as well as demographic and sampling factors, we find that experts present in countries pursuing a suppression strategy dependent on non-pharmaceutical interventions such as societal-wide lockdowns like Denmark and Norway systematically perceived their country’s COVID-19 response more favorably in terms of saving lives compared to experts from a country pursuing a mitigation strategy without the use of lockdown measures like Sweden. In light of our comparison of the three countries pandemic preparedness systems and the result of our analysis of expert attitudes to pandemic responses, we suggest that policy divergence is more related to politico-administrative factors than policy preferences among national expert communities.

Pandemic preparedness systems are defined as the set of institutions established to prepare and respond to infectious disease emergencies to mitigate their impact on systems critical to the functioning of society [13]. ‘Preparedness’ itself captures the idea that societies cannot prevent catastrophic events, like pandemics, but can prepare for them to mitigate their consequences. This requires the identification of vulnerabilities through the enactment of imagined scenarios, the development of plans to handle potential emergencies, and investments in response capacities and equipment. Despite the fact that preparedness is often treated ‘in potentia’, the circumscribed systems are part of an emergency management cycle of anticipation, response and recovery that was activated during the initial response to the spread of the SARS-CoV-2 virus [50]. The ‘policy issue’ of pandemic preparedness thus also demarcates the policy subsystem from which experts were sampled.

Since the late 1990s, the WHO has been instrumental in diffusing and standardizing the concept of pandemic preparedness across the world (Kamradt-Scott and McInnes [34]). The WHO has published several reports advocating pandemic influenza preparedness and has pushed its members to develop pandemic preparedness systems. Denmark, Norway, and Sweden established pandemic preparedness systems in the early 2000s, heavily influenced by the work of the WHO. As the WHO declared COVID-19 a “Public Health Emergency of International Concern” on the 31st of January 2020, these systems were set in action and later called into question when the three countries diverged on COVID-19 policy.

This study employs a structured comparative most similar case design to explain variation and similarity in pandemic preparedness systems and first wave COVID-19 responses in three similar countries, Denmark, Norway, and Sweden, with different COVID-19 response policies [24]. The design consists of two parts, combined in a mixed methods research strategy, to avoid the risks of selective evidence sometimes associated with qualitative research, while yielding nuanced accounts that supplement statistical associations [59]. The first part systematically compares pandemic preparedness systems and COVID-19 response strategies in the three countries. The second part subjects an expert survey on perceptions of COVID-19 response to quantitative analysis.

Denmark, Norway, and Sweden all produced pandemic preparedness plans in the early 2000s with reference to the WHOs growing emphasis on preparedness [21]. Their plans have since undergone a number of transformations, in particular, as a result of the H1N1 pandemic, which early on exposed weaknesses in health crisis management in the Nordics [3]. The COVID-19 pandemic has exposed new divergences between the Nordic countries and the importance of the politico-administrative organization of pandemic preparedness for pandemic response.

All three preparedness systems are constructed on the principles of sector responsibility (administrative units are responsible for preparedness within their purview), equality (operations should be organized similarly during emergencies as under normal conditions) and proximity (emergencies are to be handled at the lowest possible administrative level). Yet, they also differ in the responsibilities and tasks given to different layers of government and whether central authorities and political leaders can intervene in the affairs of lower levels of government and overrule the advice of public health agencies.

This study has, in a structured comparative way, investigated how otherwise similar countries belonging to the same public health regime like Denmark, Norway, and Sweden responded differently to the COVID-19 pandemic and to what degree their policy divergence reflect significant differences in the attitudes and policy preferences within and between national expert groups.

Our comparison suggests that the main differences between Denmark, Norway, and Sweden are not related to significant differences in the attitudes and policy preferences within and between their national expert groups. Experts present in countries pursuing a suppression strategy like Denmark and Norway systematically perceived their country’s COVID-19 response more favorably in terms of saving lives compared to experts from a country pursuing a mitigation strategy like Sweden. While showing the existence of a significant minority supporting ideas tied to a mitigation strategy, our results support findings made in other studies that point to the development of an emerging transnational expert consensus on the inadequacy of herd immunity by natural infection as a strategy [10, 58]. To take into account the politico-administrative position of actors, we control for location in the policy advisory system in the three counties through three parameters – part of national public health institution, seniority, and whether the expert has advised governmental and state institutions on COVD-19 related matters – but find that none of these significantly explain variation in how experts perceive COVID-19 policy response nor support focused protection. In relation to the literature on policy advisory system, our results show how experts’ perceptions and dominant ideas are not structured along locational lines related to either inside-outside or close-far positions in the system [14]. As such, we find no significant evidence to suggests that the policy divergence observed between Denmark, Norway, and Sweden reflect significant differences in the policy preferences of different national expert groups. Based on the first part of the results section, this leads us to suggest that other factors besides expert advice dissensus better explain COVID-19 policy divergence between countries belonging to the same public health regime such as the politico-administrative organization of national pandemic preparedness policy subsystems. One important difference identified by the structured comparison of Denmark, Norway, and Sweden’s politico-administrative organization of pandemic preparedness systems is the varying degree of autonomy of public health experts and agencies in the three countries in terms of defining and implementing policy measures to respond to the pandemic and the possibility of political leaders to overrule expert advice. This showcases the importance of politico-administrative factors for pandemic response and supports finding made in other cross-national studies of policy responses during COVID-19 such as Jasanoff et al.’s [33] comparison of 16 countries, which emphasized how policy response is conditioned by pre-existing structures in health, economic, and political systems.

Although our results indicate that experts’ perceptions of COVID-19 response and their support for policy preferences of either focused- or community protection are not structured along neither locational nor disciplinary divides, the results should be interpreted with caution as our measures of expert location in the policy advisory systems focus mainly on the structural aspects of location, thus overlooking potential other locational factors relevant for experts’ perceptions of response. One group of relational factors that the study is not able to control for are social network factors related to experts’ location in specific policy and expert networks in the policy advisory system [63]. Differences in experts’ perceptions of COVID-19 response and support for dominant ideas may be correlated with embeddedness in specific networks that cut across the locational, disciplinary, geographic, and demographic lines analyzed in the study. We suggest that the importance of social networks could be one fruitful avenue for future research to explore in order to further our understanding of the ideas and ties experts draw upon to understand health emergencies.

Further, our analysis identifies a statistically significant relationship between experts’ perception of COVID-19 policy response and dominant ideas such as the efficacy of focused protection strategies. While support for suppressive non-pharmaceutical interventions are widespread in the sample population, we identify a smaller group of experts whose support for focused protection reverses their perception of pandemic responses. Experts from countries pursuing a suppression strategy were more likely to perceive their national response as an overreaction in relation to the dimension of saving lives if they supported the idea of focused protection whereas experts from countries pursuing a mitigation strategy were more likely to perceive their response as adequate if they supported focused protection. The opposite pattern is visible for experts supporting the idea of community protection. As such, the results of the analysis highlight how support for dominant ideas like focused protection is significantly associated with how actors perceive and assess pandemic response. This finding echoes the general importance of ideas and beliefs for experts’ perception of pandemic response, which has also been highlighted by other studies [2, 3]. This study contributes to this literature by not only emphasizing the importance of ideas, beliefs, and preferences for how experts think about pandemic policy but also by adding how ideational support is correlated with other factors such as disciplinary background and location in policy advisory systems. As already mentioned, we find neither of these factors to be statistically significant. Interestingly, this contrasts with established theories of policy advice, which have identified the presence of shared norms, beliefs, and preferences in locational and disciplinary communities [32, 40]. This opens up space for hypotheses about what other factors are associated with ideational support like the aforementioned network factors.

While the study does not provide evidence of either locational nor disciplinary factors for experts’ perception of COVID-19 response nor support of either focused- or community protection, it does raise a puzzling question about the transnational diffusion and national embeddedness of dominant ideas such as focused protection, given the fact that its proponents are mainly located in Denmark and Sweden. This geographical concentration may be a result of sampling bias but it also opens up for thinking about how transnational ideas and advice about pandemic response travel and the transnational institutions that shape how experts perceive and think about pandemic response. It is beyond the scope of the present paper to provide a genealogy of both focused and community protection and their historical-institutional trajectories, but two areas for future research stand out in light of the results of the analysis. The first is related to the role and credibility of transnational expert authorities like the WHO in disseminating dominant ideas about pandemic preparedness and response. Despite recurring periods of crisis like the West Africa Ebola Outbreak in 2014–2016, the WHO has been central in shaping the transnational agenda on pandemic preparedness through the dissemination of new concepts, procedures, and vocabularies [39]. Following the publication of the Great Barrington Declaration, which strongly advocated for focused protection, the WHO’s director-general Tedros Adhanom publicly warned against the idea of herd immunity through natural infection, framing it as scientifically and ethically problematic due to the mortality rate of the disease, the possibility of multiple infections and the long-term health problems associated with infection [61]. Despite such clear indication from the WHO, a quarter of our sample supported the idea of focused protection one month after Adhanom’s statement. Combined with the high degree of policy variation observed during the COVID-19 pandemic, which occurred despite the WHO’s historical efforts to harmonize national pandemic preparedness systems, future research should focus on how experts across borders perceive the WHO and its epistemic legitimacy to better understand the sources and transmission of policy advice. The second is related to what alternative and potentially competing transnational institutions to the WHO that are able to shape experts’ support for dominant ideas like focused- or community protection. While it is not possible to compare the signature databases of the Great Barrington Declaration and the John Snow Memorandum with the results of our cross-Nordic expert survey, we identify similar contrasting policy preferences in the sample as in the two signature collections. While the majority of experts support non-pharmaceutical interventions, the existence of a significant minority of 25% experts in the survey still raises the question of how these experts came to support this idea and through what transmission channels. Here, future research could focus on exploring how these different policy preferences are related to epistemic disagreements and competing institutions within the field of epidemiology and their relationship to different political and moral values and beliefs.

Comparing how experts in similar public health regimes perceive pandemic response in situations of pandemic policy divergence can provide only partial accounts of how and why countries respond differently to pandemics because different cultural, historical, and economic policies can impact pandemic response in various ways. Future academic inquiry can extend the present study to other clusters of countries with similar public health regimes while keeping in mind that public health regimes are more fragmented and multi-level when it comes to pandemic preparedness than usually considered [38]. For example, the research design developed in this study can be extended to public health regimes outside of Europe such as East Asian welfare states [35] to compare expert perceptions of different public health regimes and interventions; this can provide inspiration for similar approaches tracing the impact of different dominant ideas on diverging pandemic responses and their support among national experts.

This study has examined how countries with the same public health regime diverged in policy response to the COVID-19 pandemic and to what degree this divergence reflects important differences in the policy preferences within and between national expert groups. Drawing on pandemic preparedness plans, national health care systems, COVID-19 policies, and a cross-Nordic expert survey, three main conclusions can be drawn from the study. First, Denmark, Norway, and Sweden responded differently to the COVID-19 pandemic. In Denmark and Norway, the governments enforced suppression strategies highly dependent on non-pharmaceutical interventions like societal-wide lockdowns during the first wave of the pandemic. In Sweden, a mitigation strategy was used that focused on voluntary measures, highlighting the importance of individual responsibility and seeking herd immunity in the population while only protecting individuals at high risk. These divergences crystallized during the first wave of the pandemic despite the three countries similarity in health care and welfare state models. Second, we find little evidence to indicate that this COVID-19 policy variation between Denmark, Norway, and Sweden reflect significant differences in the policy preferences of national expert groups. When controlling for locational positions in politico-administrative systems, disciplinary affiliations, as well as demographic and sampling factors, we find that experts present in countries pursuing a suppression strategy like Denmark and Norway systematically perceived their country’s COVID-19 response more favorably in terms of saving lives compared to experts from a country pursuing a mitigation strategy like Sweden. This finding points to an emerging consensus on the inadequacy of herd immunity by natural infection as a strategy. In light of our comparison of the three countries pandemic preparedness systems and the result of our analysis of expert attitudes to pandemic responses, we suggest that policy divergence is more related to politico-administrative factors than policy preferences among national expert communities. Third, perceptions of pandemic response are associated with dominant ideas such as belief in focused protection. Experts supporting the idea of focused protection perceived their country’s COVID-19 response as an overreaction if they lived in a country pursuing a suppression strategy whereas they perceived it as adequate if they lived in a country pursuing a mitigation strategy. We find that support for a focused protection strategy is associated with a consistent policy preference across locations in the policy advisory systems, disciplinary affiliations as well as gender, age and metropolitan proximity.

The study makes a substantive contribution to existing scholarship on public health security, and the growing litterature on the COVID-19 pandemic, by comparing attitudes and policy preferences for national pandemic response among different national experts belonging to a similar public health regime. Recognition of the importance of experts in pandemic response should be part of current and future research agendas that focus on understanding variations in how global ideas of pandemic preparedness are materialized in public health regimes and politico-administrative systems. What is certain is that experts and expertise will have many and complex effects on pandemic preparedness systems and pandemic response in the future [5]. Closely documenting the effects of expert networks tasked with managing and directing systems able to respond effectively to pandemic not only enhances their accountability but can also inform future policy prescriptions.