“We Do Not Have Any Further Info to Add, Unfortunately” – Strategic Disengagement on Public Health Facebook Pages

ABSTRACT

During the COVID-19 pandemic, communication with the public has been a central concern for state actors. One important question has been how to best use social media to ensure the sufficient uptake of their advice and recommendations to the public. With regard to such strategic communicative aims, a significant amount of attention has been previously devoted to the engagement, interaction, and dialogic forms of strategic communication on social media. This paper, however, focuses on an aspect that has not been discussed much in the literature: the need an organization might have to disengage due to a lack of resources or when a conversation has stalled. Using the communication that Scandinavian public health authorities carried out through their Facebook pages as cases, this paper employs a thematic analysis of the associated posts and qualitative interviews with employees to argue that these institutions use three disengagement strategies: 1) contradiction, 2) meta-discursive disengagement, and 3) disengagement through sympathy/empathy. Based on this, we consider the strategic potential of disengagement and discuss whether disengagement strategies can be considered legitimate tools for public health organizations’ crisis communication that can allow them to achieve the dual aim of ensuring citizens’ support for and compliance with authorities’ recommendations.

Introduction

The COVID-19 pandemic presents many concerns for the public health institutions (PHIs) involved in the official crisis response, which include challenges related to their communication efforts. Efficient communication with the public has been highlighted as a core component of the state’s response to the pandemic (Hyland-Wood et al., 2021) and is also a primary challenge. Within the context of strategic communication as well as crisis and risk communication, researchers have pointed out that social media platforms have become central communicative arenas for organizations, offering potential for dialogic, symmetrical, and/or interactive forms of communication with the public (Grunig, 2009; Kent & Lane, 2021). Using the interactive potential of new platforms to engage with the public has also been highlighted as a way of approaching strategic communication in a socially responsible way. Accordingly, PHIs grapple with the question of how to maximize the potential of social media when communicating with the public. This issue becomes particularly pressing when communicating about the COVID-19 vaccines to ensure the sufficient uptake of the same.

In addition to providing an ethical framework for strategic communication, communication engagement has been heralded as highly beneficial for organizations in general and for public sector organizations, specifically (Canel & Luoma-aho, 2018; Johnston & Taylor, 2018). In this context engagement can be defined as a social and relational exchange between an organization and one or more stakeholders, where communication plays an important role in creating, nurturing and influencing the outcomes of such exchanges (Johnston & Taylor, 2018a).

Most of the literature on engagement focuses on its positive value as well as how it can be understood and/or improved. Some attention has also been directed at negative engagement (Lievonen et al., 2018), understood as those incidents where negative issues related to an organization are discussed publicly. Scholars have also pointed to the learning potential of “undesired comments” in, for instance, social media and how such comments can provide valuable insights into how stakeholders think (Coombs & Holladay, 2018, p. 277). Seeking to supplement existing knowledge, this paper focuses on an entirely different dimension of engagement that, to the best of our knowledge, has not been discussed much in the literature, namely strategies for disengagement that could be used when it has become more expedient for the organization to end rather than maintain the exchange. If engagement is seen as an ethical imperative and an important strategic function in the current media environment, research on strategic communication ought to also consider how organizations can end engagement ethically and effectively. We argue that disengagement can contribute to achieving the dual goals of organizational communication, ensuring both legitimacy and efficiency in times of crisis and increased pressure. Identifying and discussing such strategies can contribute to our understanding of how organizations operate in times of crisis while also allowing for new applications of the normative perspectives suggested by theories of engagement. Theories associated with disengagement could provide organizations with tools that allow them to limit the duration of interactions with the public while still recognizing and deepening such potential future interactions.

In making this argument, we focus on a particular type of organizational actor – PHIs. As Luoma-aho and Canel (2020) argue, public organizations rely on public opinion and legitimacy to a greater degree than do private organizations. The ethical imperative of engagement is likely to be even stronger for public than for private organizations as public organizations in a democracy are instruments for governments that in turn are elected by citizens.

In practical terms, however, creating and maintaining engagement, can be resource intensive. In particular, many organizations experience limitations in their ability to engage actively with all communication taking place on their social media channels over time. Furthermore, upholding communication engagement in situations that offer little dialogic “movement,” or when participants are repeating themselves might seem futile. Thus, PHIs must practice disengagement strategically and should be particularly concerned about the ethics of such disengagement.

During the COVID-19 pandemic, Facebook has been the most central social media channel for the Scandinavian PHIs, and has proved to be especially fit for communication with the public as a whole. An important question, then, becomes how these organizations manage the discussion taking place on their Facebook pages; how they go about engaging in but also disengaging from – or shutting down – conversations. The issue of vaccination has added to this already strained situation, as it is especially characterized by the existence of vocal and active counter-publics who are not necessarily open to changing their mind (WHO, 2016).

Accordingly, we examine how PHIs use disengagement on Facebook as part of their strategic communication about vaccination during the COVID-19 pandemic. We study the Facebook pages of three Scandinavian PHIs: the Norwegian Institute of Public Health (NIPH), the Public Health Authority of Sweden (PHAS), and the Danish Health Authority (DHA) during the beginning of their vaccination programs in January and February 2021. The launch of their vaccination programs was accompanied by a rhetorical challenge: the need to actively engage with the public, explain and give information in order to, ultimately, ensure sufficient uptake among the public for the vaccination program to be effective. Through a close analysis of Facebook posts, combined with qualitative interviews with employees in all three organizations, we investigate how these PHIs go about disengaging from interactions as a particular strategy for realizing the above aims with the resources at hand.

In the following section we will briefly review and discuss existing literature on engagement, strategic communication, and social media, before suggesting a definition of disengagement. Based on this review, we then present the research questions for our exploratory study of how strategies for disengagement have been used during the pandemic. This section is followed by a short presentation and discussion of our methodological approach before we turn to the analysis itself. In the latter part of the paper, we first provide a short overview of the national contexts of the three studied PHIs. Then we detail how the PHIs relied on strategies for disengagement that can be placed in three main categories – 1) contradiction, 2) meta-discursive disengagement, and 3) disengagement through sympathy/empathy. The paper ends with a concluding section that considers theoretical and practical implications, limitations of the study, and, finally, possible avenues for further research.

Literature review

Engagement and strategic communication

As public institutions, PHIs find themselves at an intersection of several goals and responsibilities. These different and sometimes conflicting organizational goals also lead to different and sometimes conflicting communicative goals, for instance, by combining public sector goals of openness with commercial goals of efficiency (Fredriksson & Pallas, 2016). While goals and responsibilities may be conflicting, the fact that public organizations seek to achieve their aims communicatively places such communication within the remit of strategic communication as defined by Hallahan et al. (2007).

Various existing research has examined the increasing focus on forms of strategic communication undertaken by public organizations as well as the particular conditions for such organizations’ communication work (e.g., Fredriksson & Pallas, 2016; Lai et al., 2020; Luoma-aho & Canel, 2020). For instance, public organizations face increased expectations of transparency and a politicized communication environment (Luoma-aho & Canel, 2020).

As mentioned in the introduction, engagement can be understood as a social and relational exchange between actors, relying on communication and with potential benefits for all involved parties (Johnston & Taylor, 2018b; Kent & Taylor, 1998; Romenti, Murtarelli, & Valentini, 2014).

In discussing social engagement as a strategy, Johnston (2018b) argues that organizations, by adopting engagement as an orientation, may improve both strategic and social outcomes (Johnston, 2018b, p. 28). Previous research has also argued that engagement should be seen as a central concept within strategic communication (Smith & Taylor, 2017) and pointed to the benefits of focusing on engagement, both in the form of beneficial outcomes for organization and as an ethical imperative (Chen et al., 2017). Particularly the latter has been seen as an important corrective to an organization-centric paradigm, in that it sees a core function of strategic communication efforts as reaching mutually beneficial understandings between organizations and stakeholders (Heath, 2018). Again, such arguments about the normative function of engagement can be even more important when it comes to public organizations, such as the PHIs included in this study.

Social media have been identified as particularly important arenas for engagement, often seen as a form of dialogue (Kent & Taylor, 1998). Relatedly, the use of dialogical strategies on social media have been identified as a central form of stakeholder management in times of crisis (Romenti et al., 2014). Examining the current use of dialogue in studies of public relations, however, several scholars have raised criticism of how broadly the term is conceptualized and applied. Kent and Lane (2021), for instance, argue that current research does not seem to operate with a consistent understanding or use of the term dialogue; hence, what is observed in research could often be better described as simple talk or as engagement (Kent & Lane, 2021, p. 3), here understood as a more limited form of interaction than true dialogue. Of particular relevance to the present study, these authors identify areas where dialogue is unlikely or impossible. They specifically mention situations where persuasive communication occurs, situations where issues of fact are at stake, and situations that are defined by power or fear (p. 6) as situations in which dialogue is unlikely to occur. All of these features are defining of public health authorities’ communication of their pandemic response, generally, and of communication about vaccine programs, more specifically.

If we accept the argument that most interactions on social media fall short of the definitions and requirements for proper dialogue (Lane & Bartlett, 2016), the question becomes how we should understand the observable interactions taking place. Here, we argue that several existing streams of research point toward the usefulness of the concept of strategic disengagement. Roberts-Miller, for instance, argues that some rhetorical acts, particularly those taking place in social media, are mostly about establishing one’s position as a member of an in-group as opposed to a bad out-group (Roberts-Miller, 2009). Within such a perspective, dialogic strategies from organizations towards members of the public making critical posts on social media may not be the most fitting response, but there might still be a need for some form of engagement. The same interpretation can be obtained from the work of Karlsen et al. (2017), who argue that debates on social media can be considered to operate according to a trench warfare dynamic,where both opposing and supporting views in a debate reinforce existing beliefs. Persuasion or conversion might not be possible in such cases, but some form of engagement and recognition of the opposing view could still be beneficial.

The use of disengagement strategies can also be understood in light of a related study by Smith and Taylor (2017), who examined the motivations and explanations for the use of social media and found that surprisingly few respondents described having faith in the possibility of influencing or changing organizational behavior by interacting with them on social media. Nevertheless, certain respondents reported how they would still post or comment with the motivation of simply communicating, and receiving recognition for, their perspective or position (Smith & Taylor, 2017). Such recognition may be communicated by organizations as responses to criticism and could play an important role as strategic ways of engaging with critics who the organizations assume are unlikely to change their mind.

In addition to these observable responses to disagreement and repetition, we argue that another aspect of organizational communication on social media can be seen as a strategy of disengagement: organizations’ moderation and curation of their social media pages. Such efforts can also be considered part of their strategic engagement efforts since they shape how social media is used by the organizations as well as how the public interprets their presence on these platforms. One way of understanding the visible and practical work of moderating social media is as a form of metadiscourse (Buttny, 2010), wherein actors attempt to establish the borders and lines for the conversational space. Metadiscourse, referring to the communication itself, can be undertaken by all participants when they address aspects of the conversation (Buttny, 2010).

Previous research has investigated the concrete effects of such moderation. For instance, Buhmann et al. (2021) found that moderation generally had a negative impact on attitudes towards organizations from stakeholders. They place moderation within the framework of dialogue, arguing that proper dialogical orientations from large organizations may demand increased resources in the form of personnel and new infrastructure that is hard to justify, particularly since, they argue, stakeholders do not generally seem interested in engaging in dialogical relationships on social media (Buhmann et al., 2021). Supplementing this view, we will place moderation within the framework of engagement, arguing that it may be a strategic way of shaping and limiting the interaction that occurs on organizations’ social media pages.

Having discussed previous research on engagement and dialogue, their relation with strategic communication, and the various ways in which existing research highlights the limits and challenges of engagement, we now suggest a definition of what we refer to as strategic disengagement, which, in our opinion, contributes to how we can understand these limited and contested ways of engaging with the public.

Defining disengagement

When arguing that the term is useful and describes concrete behavior by organizations, our starting point is that disengagement is not the same as silence or ignoring replies. For disengagement to occur the organization has to rhetorically, e.g., through intentional communication, attempt to shut down or limit their engagement with claims, replies or statements made by members of their publics. As such, it can be helpful to consider disengagement as a specific type or limited form of engagement, rather than as the negation of engagement. If engagement is considered an exchange between the organization and one or more stakeholders (Johnston & Taylor, 2018a), then what we describe as disengagement certainly falls within the definition. Where it separates itself sufficiently to warrant discussion as a topic in its own right is in its goals of ending or limiting the exchange. A preliminary definition of disengagement, then, would be rhetorical acts designed to a) recognize and interact with concrete utterances of stakeholders, while b) limiting the scope and duration of the exchange.

This is necessarily a descriptive definition that does not attempt to incorporate the motivations that lead to the strategy being applied. However, as we have discussed, there are several possible motivations for and limitations to prolonged, repeated interactions with stakeholders. These limitations can range from interaction being resource intensive to considerations about the value of continuing such exchanges as well as organizational goals of controlling conversations that take place on their own pages and platforms. Organizations may also be faced by repeated interactions with the same participants where they know from experience that there is little chance of reaching any consensus through the exchange. All of these conditions are apparent in the contexts of PHIs’ communication about the pandemic; many members of the public have become highly engaged in conversations about various dimensions of the causes of and responses to COVID-19. In this context, vaccination is a particularly contested topic about which some are keen to voice opinions without being willing to listen to the views of others (Ihlen et al., 2021; WHO, 2016).

Taken together: while interaction, dialogue and engagement in particular have become core concepts in strategic communication, little research has been conducted on the counterpart of the latter – disengagement. In the following sections we first present out research questions and then present the methods by which we investigate the phenomenon of disengagement empirically.

Research questions

Having briefly indicated our empirical focus, reviewed and examined existing literature, and suggested a definition of strategic disengagement, we pose three research questions:

RQ1: How do PHI social media personnel disengage with stakeholders on PHI Facebook-pages?

RQ2: When do social media personnel disengage with stakeholders on PHI Facebook pages?

RQ3: How can we understand such disengagement as strategic communication?

Answering these research questions contributes to our understanding of strategic social media communication by organizations in general and public organizations in particular; of the potential limits of engagement and interaction with the public during a pandemic like COVID-19; and of engagement as an ideal for strategic interaction in such instances.

Method

In this paper, we investigate the social media activity of the official accounts of PHIs in three Scandinavian countries: Norway, Sweden, and Denmark. These Nordic countries are in an interesting geographical region for this study due to several factors – both structural and specific to the COVID-19 pandemic. With regard to the former category, Scandinavian countries are fitting areas to study communication activities on social media partially due to the high levels of use. Facebook, for instance, is consistently reported as the most used social media platform in all three countries by the Reuters Institute Digital News Report. In the same survey, the numbers for those who use social media for news are also high, at slightly less than 50% in all three countries (Newman et al., 2021). Since a core component of the engagement paradigm involves issues related to trust, it seems pertinent to study disengagement as a strategy in countries that are generally characterized as high-trust societies. Nordic countries, such as the three included in this study, consistently obtain high scores when measuring trust in institutions as well as social trust, which leads to discussions of trust as a core component of their societal model (See for instance; Andreasson, 2017). When using COVID-19 as a case, the countries selected also serve as interesting points of comparison; despite their many similarities, they have differed significantly in their pandemic response and in the measurable aspects of the pandemic effects such as deaths and number of infected individuals. Sweden has been an outlier in this sense, with higher numbers of fatalities and infections throughout the pandemic. Despite this, support for the government and the public health authorities in Sweden has remained high (Andersson & Aylott, 2020; Esaiasson et al., 2021; Ihlen et al., 2022). Having explained the logic behind our case selection, we will now describe the two datasets with which we have worked before proceeding to an account of our methodological approach to the thematic analysis.

The first dataset consists of comments made by the PHIs as replies to their own posts on Facebook during January and February 2021. The period of January and February 2021 was chosen in order to cover a time when the organizations’ communication work revolved around the issue of vaccination, based on the assumption that this would also be a period with increased interaction with members of the public. Some basic traffic figures are presented in Table 1. As is evident, the DHA’s Facebook output was larger than the output of the other two combined, while the PHAS had an average number of comments on their posts three times larger than NIPH. This disparity might be caused by several factors outside of the scope of this paper, such as the varying national divisions of responsibilities and roles of the PHIs during the pandemic, or even variations in the number of inhabitants between the three countries.

Table 1. Descriptive statistics of PHI Facebook use.

	NIPH	PHAS	DHA
Page registered	2010	2013	2020
Total number of posts	16	11	53
Total number of comments:	3175	6826	27,925
Average number of comments:	198	621	527
Followers:	125,241	248,536	162,355
Likes:	68,536	85,600	81,164

Note. Data from the Facebook pages of the Scandinavian public health authorities during January and February 2021, collected 04.03.21.

In order to examine the material qualitatively and in depth, a selection of the eight most commented posts from each organization was made, on the basic assumption that posts with more comments had a larger likelihood of containing responses and interaction from the PHIs. For these 24 posts, replies from the PHI to their posts were saved and included in the data set. The data material included in the study consisted of 1681 replies from the three PHIs (NIPH:443, DHA:748, PHAS: 490).

Second, we conducted six qualitative, in-depth and semi-structured interviews with staff members from the three PHIs involved in their social media communication, two from each, during the Fall of 2020 and the Spring and Summer of 2021. The interviews were carried out as part of larger waves of interviews in the organizations, aiming to gather data concerning all functions of the communication departments. The interviews lasted about one hour each. An interview guide was created in advance, focusing on different aspects of communication work during the pandemic. Of the wider data corpus, those interviews containing relevant discussions concerning social media use were included in this study. The interviews were transcribed in full and the quotes included in this article were translated by the first author and presented to the informants for final approval. For excerpts from the interviews, the organization is indicated by the first letter of its name, which is followed by a number that represents the individual interviewees.

Since little existing research deals explicitly with disengagement as a part of organizational communication, we have chosen to rely on an inductive approach to the thematic analysis of the material. In our study, we have applied parts of the so-called “Gioia methodology” (2013): First, we identified the first-order concepts related to disengagement among the Facebook posts included in our analysis. In other words, we identified elements in posts that could be described as fitting our suggested definition of rhetorical acts designed to a) recognize and interact with concrete utterances of stakeholders while b) limiting the scope and duration of the exchange. These can be thought of as individual examples of disengagement in practice, as identified by us as researchers (Braun & Clarke, 2006). When conducting our thematic analysis of the 1,681 total posts, various themes of disengagement were found in a total of 297 posts, with the remaining posts typically containing concrete answers to questions posed or other forms of open-ended engagement. To provide some idea of the prevalence of these themes within the material, we have included a table that illustrates the distribution of the themes in the responses of each of the PHIs (see Table 2).

Table 2. Coding of themes.

	TOTAL	NIPH	PHAS	DHA
Contradiction	154	76	23	55
Metadiscursive	95	17	42	36
Sympathy/empathy	54	24	2	28

Note. Table shows the distribution of qualitative codes for each PHI in order to illustrate the prevalence of the qualitative themes.

Through our thematic analysis (Braun & Clarke, 2006), we then synthesized these examples and concepts into three thematic categories or second-order themes, which are all relevant to the aggregate dimension of disengagement: 1) contradiction, 2) meta-discursive disengagement, and 3) disengagement through sympathy/empathy. These thematic categories can be considered generalizations of the individual examples into broader categories that are meant to represent overlapping and related trends in the material. We discuss these categories as potential strategies for disengagement later on in this paper. In the results section, each of these three second-order themes will be analyzed and illustrated using relevant examples from the Facebook material. While this approach is inspired by certain elements of the Gioia methodology, it differs in some important regards – primarily, in treating communicative acts, in our analyzed Facebook comments, as sources of first-order concepts. The rationale behind the application of this approach is based on the central assumption of the methodology–that the informants are knowledgeable actors. Similarly, we assume that the communicators are knowledgeable actors who act in identifiable and strategic ways that we, as researchers, can identify within the research data. We also assume that this, in combination with the context established through our interviews, can help us understand how disengagement may be used in strategic ways. We believe, following the points and perspectives discussed by Gehman et al. (2018), that such customization of the methodology according to a concrete piece of inductive qualitative research is essential to make it fit the various contexts in which the qualitative research is being conducted. We argue that while such an inductive approach is not sufficient to generalize the frequency of the various strategies, it does lay the ground work for a typology that can be applied and tested in later work.

Results and discussion

As described in the previous section, we have grouped our results as three second order themes related to disengagement as a dimension of the PHIs’ social media work as identified in their Facebook posts. In this section we will describe and exemplify, one by one, how these themes present themselves in the data material. Before moving to the concrete analysis of the various themes, it is necessary to provide some background and context about the different PHIs, their use of Facebook in general and during the specific period included for research in our study.

National contexts

In Norway, the NIPH communication department had a well-established presence on social media prior to the pandemic. When asked about the communication goals for the social media work, one informant described them as being largely the same as for the organization as a whole, while pointing to dialogue as a central ambition:

To be open and available and to have a dialogue with our target audience. To be visible, not for the sake of it but to share our knowledge and give advice, and to receive feedback on our work in order to improve. To be present in the channels where we need to be present and to say what we know and what we don’t know in order to build trust in our knowledge and our advice (N3, NIPH)

Interviewees from the various organizations share this ambition, concurring that their organizations should try to respond to all inquiries. When asked about their work on Facebook, respondents from PHAS highlighted how challenging managing their pages during the pandemic had been, largely due to the amount of interest and inquiries:

There have been all sorts of threats, and we remove those of course; but still, it has turned out well because the employees have made an enormous effort. We have been available around the clock. It is not like we have been adhering to office hours; that has not been the case for us since February [2020]. (P1, PHAS)

The informant specifically mentions threats as something they have chosen to delete, and the issue and extent of threats towards staff of the Public Health Agency has been discussed and debated in Swedish media (Wong, 2021).

A different informant from the same organization described how answering all questions on Facebook was a conscious decision, stating that they, during the pandemic, decided to prioritize engaging with the public on Facebook while not necessarily doing so on Twitter. The informant also tied this strategy to increased amounts of inquiries:

We decided quite early on that we would answer all questions that we received on Facebook, so we have used a lot of resources in doing so. But that has also resulted in an enormous number of people visiting and commenting on the page. (P2, PHAS)

DHA only registered their Facebook page in 2020, specifically in February a few days after the first reported case of COVID-19 in the country. In interviews, respondents from DHA clarified that the process of actively using Facebook in their communication had in fact been started earlier, and included the hiring of employees specifically tasked with running their social media (D1). Discussing the differences between social media and other forms of communication, the same respondent mentioned the interactive nature of the platforms:

It can be a lot more involved […] It might be that we set the premise for a conversation, but then it might go in any possible direction, and that is significantly different than when we send out press releases and so on. In these cases [on social media] we are hosting the conversation, it is taking place on our platform, and we have to participate, that is kind of the premise of it. (D1, DHA).

The value of interaction, and, the ideals of dialogue and engagement, can be seen in the responses from all three organizations. In the following sections we will examine what forms this interaction take in practice, but first let us detail how the organizations view their terms of engagement.

All three organizations have a set of stated rules and guidelines posted to the “about” section of their Facebook page. While there are some differences in emphasis and focus, many traits are similar between all three. They all mention that while they appreciate discussion and disagreement, they intend for their page to be a welcoming place for everyone. All three also state some concrete rules for behavior they will not tolerate in order to ensure this welcoming atmosphere. While the Swedish rules are quite detailed, mentioning specifically that such things as racism, cursing or pornography are banned from the page, the Norwegian page simply states that participants are expected to follow common decency. The Danish page specifies that it will not accept offensive or harassing comments.

The Swedish page specifically mentions that it does not allow posts that appear to be misinformation or information that cannot be verified. The same is true for the Danish page, which in addition mentions that breaking rules will lead to a formal warning, and that three warnings will result in the commenter being banned for six months. During interviews, a respondent from PHAS described how they preferred to enforce their rules through “hiding” comments rather than outright deleting them and explained that the decision of what to consider misinformation was made based on reports from The Swedish Civil Contingencies Agency (SCCA):

We collaborate with SCCA when it comes to disinformation and such things. And every week they send out a report about what is being spread and so on. And we help out with that as well of course. But the things we consider to be disinformation can be hid, so that it is not visible, but we do not delete it. So it is visible for the ones who [posted it]. […] but it is not something we have decided on our own, we follow existing routines. (P2, PHAS)

Although the difference between hidden and deleted comments is emphasized here, this difference is only clear to the administrators. General users do not see a hidden comment, just as they do not see a deleted comment, while the author sees the comment and does not know that it has been hidden from everyone else

The Norwegian page does not specifically address misinformation, but mentions “irrelevant comments” as something that can be deleted; they also mention that repeated violations of the stated policy can result in being banned from the page.

As we will discuss in further detail later, the stated rules also serve as a reference point for metadiscourse when engaging with comments that cross the line with regard to what organizations consider acceptable on their page.

Disengagement through contradiction

The first strategy of disengagement that we identify consists of the PHI’s contradicting information in posts made to their page. One concrete example is an effort by DHA to challenge what they consider misinformation about tactics for strengthening the immune system as a way of combating COVID-19..

There is no doubt that a well-functioning immune system is essential in the body’s fight against diseases, but it is important to emphasize that there are no superfoods that can defeat the corona virus […] The best way to defend yourself against the corona virus is to follow our recommendations on good hygiene, coughing into your sleeve and limiting physical contact with others. On social media there are a lot of “good advice” about how you can eat your way to a strong immune system and through that avoid being infected by the corona-virus. This is misinformation that risks misleading the Danish people. (DHA, Facebook comment, February 2021)

The answer has a ‘concede and refute’ structure which makes the answer both pleasant and definitive. It recognizes some value in the question and the benefits of good nutrition, so as to concede to some extent, but then it presents a refutation with strong truth claims regarding certain recognized measures that assuredly work during the pandemic. It is a counter that does not beg further questions. This approach of contradicting users’ claims clearly suffices to tackle some of the communicative challenges presented by the comments that the organizations are replying to; while it does not explicitly shut down further attempts at interaction, the approach does not actively seek to further engage with the commenter.

This approach to disengagement can also be seen in cases where the post being replied to consists of accusations, rather than misinformation. In one such case, NIPH’s recognition of dissenting views goes so far as stating “thank you for your opinion. When it comes to our recommendations, they are based on an updated and professional assessment of the current knowledge, and definitely not on apathy. Best regards NIPH” (NIPH, Facebook comment, February 2021). In the case of a different critical post, the organization takes a more combative stance, stating that: “This is a pretty unreasonable statement with no real basis [in facts]” (NIPH, Facebook comment, February 2021).

A subgroup within the same theme is the combination of contradiction with the recognition of the right of the commenter to disagree with the organization. In one such exchange an employee of NIPH engages in what seems like a back and forth exchange with one user, before seemingly deciding that the interlocutor is unlikely to change their position. They then end the exchange by stating:

Yes, immunology can be complicated stuff, so it’s a good thing that we have qualified experts in the field. We would also remind you that vaccination is voluntary, and wish you a good weekend! Best wishes, NIPH (NIPH, Facebook comment, February 2021)

While the post includes references to superior knowledge, it is the most explicit in disengaging when this is read in combination with the emphasis on the voluntary aspect of vaccination and the conclusive well wishes at the end. A similar tactic is employed in the case of a commenter accusing NIPH of wanting people to suffer, to which the organization replies:

Your impression does not match reality, NIPH wants the opposite: as little suffering as possible for as many healthy inhabitants as possible. But you are allowed to think what you want. We do not have any further info to add, unfortunately. (NIPH, Facebook comment, February 2021)

Both the statements of “have a good weekend” and “we do not have any further info to add, unfortunately”, clearly mark the end of the exchange. As part of the organizational strategic communication effort, these statements serve two different functions: they directly address and contradict what organization employees consider to be misinformation, while simultaneously showing a clear intention of shutting down the discussion, rather than continuing to spend organizational resources on engaging with a particular issue or commenter. In a similar example, NIPH responds to a comment stating that:

… [I]t is not a poison, it is a vaccine that protects against serious disease and death, and the list of ingredients is available and well known. But we know that you disagree with us about this, and you will probably continue doing so. Anyway, have a good Friday and a nice weekend. Best regards NIPH (NIPH, Facebook comment, February 2021)

Recognizing that the organization does not expect or demand mutual understanding where the commenter they reply to changes their opinion removes the need for further communication. A similar tactic is at times used by PHAS, although their version tends to focus on the organization itself rather than the commenter. One example of this is when the organization states that “We maintain our position on this, children and young people spread infection less.” (PHAS, Facebook comment, January 2021). That is not to say that the organization does not use flat out contradiction, such as in a reply where they state: “No, that is not what it means. The vaccine cannot give you COVID-19.” (PHAS, Facebook comment, February 2021).

A similar dynamic, but with a slightly different articulation, can be found in the active use of the voluntary nature of vaccination in the countries included in the study. In these examples, the communication employees generally contradict commenters’ positions or give information about the positions and recommendations adopted by the organization, before underlining that vaccination is voluntary and that no one will force anyone to subject themselves to the vaccine. In one typical example from NIPH the organization starts out by contradicting claims made about the m-RNA technology being new and experimental. They then give a selection of technical information and refer to a separate public organization for technical information about how the vaccine works before ending the engagement by stating that “you are correct that it is up to the individual to decide whether they wish to be vaccinated, all vaccination is voluntary in Norway” (NIPH, Facebook comment, February 2021). In a similar case the DHA asks “Why do you think it is made up? Choosing to be vaccinated is completely your own choice”. Here we see a combination of turning the table and questioning as a way of showing willingness to engage while simultaneously disengaging by highlighting the citizens’ freedom of opinion and action as vaccination is voluntary.

In sum, what we have defined as disengagement through contradiction can be understood as a strategy where organizations end an interaction by contradicting statements made by citizens in reply to posts on their page. This strategy for disengagement seems to be used when the organization needs to clearly respond to misinformation but does not wish to delete the same or get caught up in a longer exchange.

Moderation and metadiscourse as disengagement

The second strategy for disengagement that we will discuss is what we have chosen to call disengagement through moderation and metadiscourse. This strategy relies on the active use of the rules set by the organizations as a way of disengaging with members of the public. However, not all metadiscourse functions as a form of disengagement. In one example, the NIPH replies to accusations that they have removed a comment by stating that they do delete comments if they break the stated rules for the page, but that they have not deleted anything in the concrete case being discussed. While this is a form of metadiscourse, it does not attempt to disengage from a specific interaction.

In an example of how instances of metadiscourse can function as a form of disengagement, the PHAS refers to their stated policy against misinformation: “You are welcome to comment here [on this page], but it is not ok to spread false information about vaccination, if you continue you will be blocked” (PHAS, Facebook comment, February 2021). While this is clearly an example of metadiscourse, referring to the rules and regulations by which the organization’s comment section is run, the reply also directly addresses and comments on a concrete case of misinformation being posted. Accordingly, it can also be understood as a way of disengaging with a commenter while still rebutting the content of their post.

In a different case, the Danish authorities refer explicitly to the general tone of the debate and their stated policies:

Hi everyone in this thread. Our Facebook-page should be a nice place to participate for everyone. Therefore, we encourage you all to write in a nice and respectful tone. Criticism is completely fine, but the tone cannot be too harsh or too personal. See also our stated guidelines in our “about” section. (DHA, Facebook comment, February 2021)

Importantly, this instance of metadiscourse seems to be preemptive rather than justifying specific acts of moderation – that is, the organization does not seem to have deleted comments occurring earlier in the thread, at least they do not state that they have. Rather, they are reminding participants of the rules and expectations so as to stop further posts with inflammatory content or at least prevent the thread from escalating.

A separate type of metadiscourse that can serve as a form of disengagement is the reiteration of the conditions necessary for comments to be responded to. In one such case PHAS responds to a commenter saying: “Hi, we are happy to answer questions, but we are not able to engage in speculative discussions” (PHAS, Facebook comment, January 2021).

Similarly, metadiscursive disengagement has the potential to simply dismiss posts as a form of spam; one clear example of this can be found in a case where the DHA responds that “you have made this post many different places, and we consider that spamming. See our guidelines under our “about” section. You can consider this your first warning, if you get three warnings, we have to block you from our page for the next six months” (DHA, Facebook comment, February 2021).

We also see metadiscursive disengagement applied through active disengagement from discussions and questioning that fall outside the responsibilities of the individual organizations. In one such case, the DHA states that “it is our job to make decisions regarding public health, but it is the politicians that make decisions when it comes to the shut-down of society (DHA, Facebook comment, February 2021). As a method for disengagement, this approach underlines that the organizations do not make decisions about certain areas of policy, and that debating the issue with them is unlikely to change anything.

Lastly, there are cases in which metadiscourse overlaps with other themes of disengagement. In one example from the Swedish PHAS, the employees combine metadiscourse with contradiction. Opting to explain that since the information is incorrect, it can be considered misinformation and therefore in violation of the organization’s stated rules for their Facebook page:

Your claim is not correct. We do not allow misinformation in our comment section. The Corona-virus has not only led to the spread of a disease, but also to the mass spread of rumors and misinformation. WHO has called this epidemic of rumors an infodemic. We take this seriously and we will not allow our comment section to be a part of it. You can assume that if you continue to spread misinformation you will be blocked from this page. (PHAS, Facebook comment February 2021)

While this post clearly contradicts and brands the original post as incorrect and could have been discussed in our previous segment about contradiction, it also goes further by labeling it as a violation of the rules and therefore subject to removal. The function of the contradiction, then, becomes to use meta-communication as a way of disengaging.

This form of metadiscursive disengagement, where organizations attempt to end an interaction through referencing the rules and guidelines for their page or platform or the scope of their responsibility can be considered a strategy for disengagement, as it functions as a way of shutting down conversations that are outside the accepted topics or include aspects that the organization does not tolerate. Simply deleting the post would not count as a form of disengagement according to our definition but providing a reply and an explanation both controls and limits an exchange while acknowledging the poster of a comment.

Sympathy, understanding and empathy as disengagement

A third rhetorical strategy of disengagement is the active use of understanding, sympathy and empathy as a way of recognizing critical comments while disengaging from the interaction. While this form of reply takes various forms, they share the explicit recognition of the feelings and humanity of the person commenting as well as the employee formulating the reply. One example from the DHA is a reply opening with “I can understand that it is frustrating that there are no free bookings to get vaccinated against COVID-19,” followed by a general explanation of how responsibility for vaccination, scheduling and supply is divided between various levels within the Danish health system (DHA, Facebook comment, February 2021). Similar tactics are employed in posts from the NIPH, who often attempts to recognize and empathize with commenters expressing frustration and fear, one clear example of this stating:

We understand that you are getting tired of this, we are getting tired as well. Now that we are beginning to make good progress with the vaccine, we are approaching the point of being able to open society again, step by step as time goes by. But unfortunately, it will probably be a while before we can give each other hugs again:’(Best wishes, NIPH (NIPH, Facebook comment, February 2021)

In one of the more telling examples of how this strategy can work as a form of recognition while disengaging, the DHA opens their comment by stating “I am sad to hear that you are furious”, before moving on to providing links to a separate public agency responsible for the distribution of vaccines (DHA, Facebook comment, February 2021).

We did not find many examples of the Swedish PHAS using expressions of understanding, but in one of the few examples in the material the agency expressed how “that is sad to hear. It is not true that many people get sick from the vaccine” (PHAS, Facebook comment, February 2021). Interestingly, this use of sympathy is combined with the use of direct contradiction, with sympathy functioning to assuage the contradiction whilst not detracting from its force.

This strategy for disengagement, where the organization takes care to express an understanding of and/or commonality with the commenter while not attempting to resolve any criticism or disagreement or inviting further engagement in any other way, appears less frequently in the data and is often combined with other functions such as replying to factual inquiries. As a strategy of disengagement, this can be related to ideas about the value of acknowledging the user (Canel & Luoma-aho, 2018) and is more about the recognition of emotional states and the expression of a form of solidarity than the recognition of dissenting views.

Concluding discussion

This paper has demonstrated that while engaging, replying and interacting with the public are common approaches to PHI social media work, there are frequent instances where the goal of messages seems to be to shut down rather than to open up conversations to further interaction or dialogue. As an answer to RQ1, we have categorized these forms of disengagement into three distinct strategies. These strategies often seem to function as a compromise between principles and goals of the organization and constraints posed by a lack of resources and a need for control. While the wish to engage with and reply to members of the public is visible, both in answers given during the interviews and in the sheer amount of replies the organizations do post to their own Facebook posts, scarce resources and time pressure mean that they cannot possibly respond to all inquiries or statements. This is to be expected during an ongoing crisis, such as the COVID-19 pandemic, where increased pressure is placed on the organization. Rather than simply ignoring posts, or deleting them, which could hurt their legitimacy, the organizations attempt to clearly communicate the end of the interaction while recognizing their counterpart by using disengagement strategies.

Concerning when these strategies are used (RQ2), the motivation for using the different strategies could vary. One possible interpretation, particularly for the use of disengagement through contradiction could be tied to the perspectives suggested by Roberts-Miller, where the purpose of the original comment may not be to reach some form of consensus, but rather to state one’s group identity (Roberts-Miller, 2009). If the PHI employees, either through repeated interaction with the same commenter or through clues from the language in the post, feel confident that the poster is not in reality looking for any form of dialogue, they might be less likely to prioritize responding to them in a way that opens up or furthers interaction. We see examples of this in the previously quoted post by NIPH where they state that they know that the commenter disagrees with them, and that they do not find it likely that they will change their mind.

To explain this strategy, it is also possible to draw on the idea of recognition, as suggested by Luoma-aho and Canel (2020) and the related findings of Smith and Taylor (2017). According to these scholars, recognition might not only be the most important part in responding to a post, it might also be exactly what the poster hoped to receive when posting it. Here it is also possible to speculate that recognition is even more essential during a pandemic; when uncertainty and fear are common emotions in the public, simply being ‘heard’ may offer some reconciliation.

As for RQ3, our results indicate that when conceptualizing disengagement as a form of strategic communication we should take care in emphasizing that disengagement is not the same as silence or silencing. It is an active effort to end an interaction in a way that recognizes the counterpart, while minimizing negative responses to the fact that the interaction is being ended. It is also important to understand disengagement in the context of the strategic use of social media as an arena for engagement, where the PHIs operate with an ambition to answer and engage with members of the public. This interpretation is also strengthened by our finding that strategies of disengagement are more frequently used when organizations do not delete what they consider to be misinformation.

When it comes to the legitimacy of disengagement, the concrete forms of disengagement we have discussed cannot be separated from important normative and ethical discussions. Throughout this paper, we have referred to disengagement as motivated and justified through the context that the PHIs find themselves in, and as responses to, for instance, what the organizations perceive to be misinformation. While we argue that the inclusion of disengagement contributes to the development of engagement theory within strategic communication by including observable strategic behavior used by organizations, this should not be taken to mean that disengagement is an ethically neutral strategy that we condone for organizations in any context. As the pandemic has shown us, even the concept of misinformation is contested, and it is not a given that public institutions’ perceptions of what is correct actually match reality (see for instance, Hansson et al., 2021).

We do argue, however, that disengagement describes a concrete observable form of strategic behavior, particularly on social media, that is quite common and to a certain extent necessary during periods of increased pressure. Studying, theorizing and developing a framework for such behavior can help us understand and discuss to what extent disengagement can fit into strategic communication in practice, and potentially, how it can fit into a larger framework of ethical strategic communication work by (public) organizations.

Theoretical and practical implications

In this study, we have conducted an exploratory investigation into a concept we have called strategic disengagement, as identified in PHI communication during the COVID-19 pandemic. We define disengagement as rhetorical acts designed to a) recognize and interact with concrete utterances of stakeholders, while b) limiting the scope and duration of the exchange. While our qualitative approach is not designed to generalize or quantify the use of this strategy, we do argue that such acts can be frequently found in our material. This is not surprising, as the potential for social media as an arena for engagement and dialogical interaction may be widely recognized (Kent & Taylor, 1998, 2002), but it is equally clear that such engagement is resource intensive and costly for organizations (Lane & Bartlett, 2016). During the COVID-19 pandemic all three PHIs included in the study experienced tremendous increases in the number of comments received, as well as in followers on their Facebook pages. NIPH for instance, has described how a post to their Facebook page could receive a thousand comments within a few hours of it being made (Fylling, 2021).

We argue that when scholars of strategic communication herald engagement as a normative and ethical approach for practitioners in their attempt to communicate with the public (Chen et al., 2017; Heath, 2018), we also need to devote time and attention to how organizations should attempt to end and limit such interactions.

Conceptualizing and analyzing such interactions contributes to the field in several ways. First, it provides tools for practitioners in the form of theories and strategies that can be applied in situations where there is a need for disengagement. Building theory surrounding such acts also contributes to critical reflection about when and how strategic disengagement can be applied. The latter is especially essential in order to retain the normative and ethical foundation that engagement theory is built on, while recognizing its limitations and the concomitant need for disengagement. The application of disengagement strategies can thus function as a bridge between the practical constraints of an organization and the normative ideal of engagement, where interactions can be limited while the organization recognizes and respects the positions and disagreements expressed by critical publics. Developing a functional conceptualization of disengagement can also contribute greatly to crisis preparedness, since crises in particular are characterized by external factors that may increase the need for organizations to limit the form of their engagement with the public due to increased pressure and constraints in time and resources.

The theoretical implications of our findings lie primarily in providing nuance and depth to theories of engagement within strategic communication (e.g., Johnston & Taylor, 2018b). Incorporating disengagement as a highly limited form of engagement, and critically studying its use both experimentally and theoretically would broaden the theory to cover more of the actual practice of organizations in their communication and thus increase the likelihood of its further adoption and use by practitioners.

The reflections above point to two significant limitations in our study, but also possible paths for further research. First, we do not study interactions between the PHIs and stakeholders as a whole, but focus on identifying the PHIs’ disengagement strategies. How members of the public respond to the attempts at disengagement is, therefore, an open question for further study. Second, while our study identifies strategies and approaches, it does not test the effect and consequences of these strategies. We focus on the choices and strategies employed by the organizations, but further research is needed to investigate the relationship between organizations and individual commenters on Facebook. It should also be noted that there are some characteristics of the current situation that may be particular to long lasting pandemics with large societal impacts. The organizations included in our study had been under significant pressure for close to a year at the time when our data was gathered and their strategies may have been different during a more intense and short-lived crisis. While more research is needed, we argue that disengagement is a central but overlooked aspect of strategic communication, generally, and that further studies may find that most if not all organizations use disengagement strategies in certain situations. Further identification, categorizing and testing of these strategies, as well as their potential effects, can contribute to a better understanding of the forms strategic communication take in practice.

Disclosure statement

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

Additional information

Funding

This work was supported by the Norges Forskningsråd [312731].