The Role of Social Media Influencers in Public Health Communication: Case COVID-19 Pandemic

ABSTRACT

During public health crises, public organizations face a variety of strategic communication challenges, and the outbreak of the COVID-19 pandemic in 2020 is an extreme example. In Finland, the Prime Minister’s Office initiated a communication campaign that utilized social media influencers to communicate timely instructions regarding the pandemic. However, it is uncertain how social media influencers adapt to briefings of public organizations given that they typically work with brands that align with their own interests and expertise, which rarely is epidemiology. We use the two-step flow of communication model and social influence theory to analyze research data that consisted of 96 Instagram posts, 108 Instagram Stories and 1097 comments. Qualitative content analysis was used to see how the influencers communicated about the pandemic and how their followers reacted. The results suggest that the influencers tried to adapt the messages to their own style, and, instead of committing to the wordings of the campaign, they shared general guidelines and, with their own example, showed how to behave during the pandemic. Their participation in the campaign helped affect social norms during the time of the crisis, which in the case of public health communication is a substantial, strategic goal.

Introduction

In a time of crisis, Finland is calling in the cavalry: social media influencers.

(Politico, 2020)

This paper studies a strategic communication campaign to reach and inform citizens during the COVID-19 pandemic. Besides the unique context of a global pandemic, the campaign was unique also because it relied on social media influencers – thus far a rare feature in public communication campaigns (Bonnevie et al., 2020; Borchers & Enke, 2021). Finland was one of the first countries in the world to include social media influencers in the urgent COVID-19 communication, as part of a communication campaign initiated by the Prime Minister’s Office. The campaign titled Corona Facts encouraged and supported social media influencers in sharing reliable information about the coronavirus, combating the spread of misinformation, and reaching citizens who were outside the reach of legacy media and public organizations’ communication channels (Ping Helsinki, 2020; Reinikainen et al., in press). These efforts are perceived as strategic social media influencer communication – the utilization of social media influencers in strategic goal-attainment (Enke & Borchers, 2019).

However, most social media influencers that authorities might hope to engage in public health communication campaigns are not experts in medicine, and their strategic influence is based on a number of interactive follower relationships and capability for content production and distribution (Enke & Borchers, 2019). Therefore, including social media influencers in strategic communication of urgent public health issues might be risky for organizations: public health-related posts most likely contrast with usual influencer content and undermine content coherence, a key factor of social media influencer effectiveness (Pöyry et al., 2019). As a result, the content does not necessarily create the wanted attention, engagement, and interaction.

Building on strategic communication, media studies and social influence theory, we analyze the Corona Facts campaign using the two-step flow of communication model. The model explains how information flows from the mass media to opinion leaders and then to the general public (Lazarsfeld et al., 1948). Later research has questioned the model, particularly because social media allows refined targeting and personalized communication (Bennett & Manheim, 2006). While most research agrees that opinion leaders play a role in disseminating information in social media, research has largely focused on organically emerged opinion leaders with pre-existing interest and knowledge of the topic in question (Choi, 2015; Nisbet & Kotcher, 2009) – features that rarely describe social media influencers in the case of contagious diseases and epidemiology.

We analyze whether the intended campaign messages go through unchanged or whether the influencers shape the content or add their own interpretation of it. Further, according to the social influence theory, we analyze whether the influencers follow the informational or normative route of persuasion. The theory posits that people’s decisions and behaviors are influenced by either information on the desired behavior or by observations on others’ behavior and the social norms attached (Kaplan & Miller, 1987; Wood, 2000). In computer-mediated communication, normative influence disappears if there is no group saliency (Perfumi et al., 2019) but social media influencers can pose normative influence over their followers. This allows us to understand the persuasion strategies the influencers used and how their followers reacted.

Literature review

Strategic social media influencer communication has developed into a prominent domain in strategic communication (Borchers & Enke, 2021). Many organizations have found that collaboration with social media influencers can advance their strategic goal attainment, as endorsements made by influencers have been found to increase favorable attitudes, intentions, and behaviors towards organizations (J. E. Lee & Watkins, 2016; Munnukka et al., 2019; Reinikainen et al., 2020; Sokolova & Kefi, 2020; Trivedi & Sama, 2020).

From the point of view of strategic communication, social media influencers have been defined as “third-party actors that have established a significant number of relevant relationships with a specific quality to and influence on organizational stakeholders through content production, content distribution, interaction, and personal appearance on the social web” (Enke & Borchers, 2019, p. 267). This definition highlights the interactive and intimate relationships that social media influencers are able to tie with their followers in high numbers. Studies have found that follower engagement is an important factor contributing to the success of influencer endorsements (Munnukka et al., 2019; Reinikainen et al., 2020) and thus to the effectiveness of strategic social media influencer communication.

Interestingly, existing academic literature on social media influencers’ role in public health communication concentrates mostly on the influencers’ potential harmful effect on their followers. For example, Jenkins et al. (2020, p. 11) argue that social media influencers “often promote damaging fad-diets and share misinformation without consequence” and Leader et al. (2021) interview vaccine-hesitant and vaccine-refusing social media influencer-mothers. While these studies generate important knowledge on how disinformation and otherwise harmful information spreads, it is worthwhile also to look at the opportunities that relate to influencers as identifiable opinion leaders and communicators of urgent and critical information. From organizations’ point of view, working together with these opinion leaders might be essential to attain their strategic communication goals.

Information flows and the role of social media influencers

A seminal communication theory, the two-step flow of communication model, highlights interpersonal relationships in the mediation of information in the society. The theory argues that, instead of getting information directly from the news (one step), most people are persuaded through the so-called opinion leaders, who mediate information from news outlets to their friends and family (two steps; Katz, 1957; Lazarsfeld et al., 1948). At the time of the introduction of the theory, opinion leaders were thought to be individuals who have relatively large circles of contacts, are influential in specific domains, and more exposed to formal mass communication than other group members (Katz, 1957). Personal beliefs and values were seen to affect what kind of information opinion leaders pass on, thus acting as gatekeepers between public organizations, corporations, news organizations, and the masses (Laughey, 2007).

While some elements of the two-step flow of communication model are less pressing in the present time and age, such as people’s access to news and the mass media, there are other elements that make the theory still relevant. In the time of online news and social media, the role of various gatekeepers and opinion leaders is as critical as ever. For example, in the case of social media platforms, people are more likely to read online news that are liked and recommended by other people (Choi, 2016; Wang & Li, 2016). It has also been shown that the so-called weak ties (interpersonal relationships that connect dissimilar and distant networks of people) are key in disseminating information in the digital sphere (Bakshy et al., 2012), a phenomenon that is further affected by the logic of the algorithms (Soffer, 2019).

Most social media channels were originally created for the purpose of connecting with friends, family and acquaintances. Later, the importance of organizational and non-personal connections on social media grew. In the current social media environment, one can subscribe to the content created by brands, media outlets, politicians, celebrities, and social media influencers, which inevitably decreases the relative visibility of personal, “real-world”, connections (Pöyry et al., 2019). Because of this phenomenon, Thorson and Wells (2016) propose the concept of curated flows to be an important feature in what kind of information gets consumed. According to this view, one’s personal communication network is affected by processes of curation by different actors, such as journalists, strategic communicators, social contacts, algorithmic filters, and the person herself.

A social media influencer’s decision regarding the topic areas they post about represents the process of curation (Thorson & Wells, 2016). The existing categorization of the curated communication flows does not however account for the curation undertaken by social media influencers. Social media influencers are not journalists as they are not bound by similar ethical guidelines. They typically post about matters related to their personal life and are not expected to consider any wider perspective. Neither are they friends with all their followers, which means that social curation is different from the curation performed by the influencers. Finally, influencers might curate content based on financial or other interests, which makes it resemble strategic curation, but the goals of the influencers might heavily fluctuate over time, which is rarely the case of corporations or public organizations. “Influencer curation” is therefore an important and distinct process when considering the information that gets produced, circulated, and consumed today.

Persuasive capabilities of social media influencers

In principle, there are two types of opinion leaders: monomorphic, who employ expertise in one specific, narrow area, and polymorphic, who employ expertise in several, broader areas of knowledge (Merton, 1957). In practice, however, opinion leaders usually possess a mixture of monomorphic and polymorphic features, thus exerting both narrow and general expertise (Wang & Li, 2016). The same multifaceted nature of expertise can be seen to apply also to social media influencers; many social media influencers have a distinct brand and are profiled through a specific domain (Duffy & Hund, 2015). This does not mean that they only publish content in their domain, or that they would not have an effect on other domains or more general issues.

Based on the theory of weak ties (Bakshy et al., 2012), it could be assumed that individuals who typically publish content on a certain domain (e.g., beauty), can have a disproportionately large effect on information dissemination and persuasion when they publish content on other domains (e.g., a public health crisis). Indeed, there are numerous examples of celebrities who have taken upon various social causes that go beyond what they were originally known for (Bennett, 2014; Williams, 2020), as well as private citizens who possess disproportionately large impact in their online social networks because they are active and influential in several different sub-networks (Choi, 2015; Nisbet & Kotcher, 2009).

The persuasive capabilities of social media influencers rely on other factors than in the case of other types of communicators. Authenticity (Pöyry et al., 2019) and intimacy (Berryman & Kavka, 2017) are features that describe persuasive social media influencers and differentiate them from politicians, officials, and other formal authorities. Social media influencers are often perceived as peers to their followers, which differentiates them from traditional celebrities (Enke & Borchers, 2019). Nevertheless, influencers are often adored and admired in the same way as celebrities are. Thus, a certain kind of fan behavior can be discerned – followers sometimes support and defend their favourite social media influencers, pursue interaction with them, seek to conform to their endorsements and follow their advice (Berryman & Kavka, 2017; Marôpo et al., 2020).

Research has shown that it is rather their emotional contagion (e.g., excitement, passion) and linguistic style that affect social media influencers’ ability to persuade than content or production expertise (M. T. Lee & Theokary, 2021). The relatively insignificant effect of influencer expertise on endorsement effectiveness has been also shown by other research (Kim & Kim, 2021; Lou & Yuan, 2019). A unique feature in what makes social media influencers persuasive is also their relationship with followers. Reinikainen et al. (2020) show that parasocial relationship – an imaginary relationship with media performers (Brown, 2015, p. 275) – affects influencer endorsement effectiveness. This suggests that persuasion of social media influencers is notably different from the traditional meritocratic formation of expertise and authority.

Finally, an influencer’s ability to persuade his or her followers might follow the routes of social influence theory. Social influence refers to how people are influenced by the behavior of others to conform to community behavior patterns (Venkatesh & Brown, 2001). The theory suggests that there are two types of social influence: informational and normative. The former refers to information as a form of persuasion while the latter to persuasion based on observations on others’ behavior and the desire to maintain group harmony (Kaplan & Miller, 1987; Wood, 2000). Normative influence is thus based on perceptions about social norms.

Online environments typically support informational persuasion because information can be so easily shared in a digital format, even anonymously. However, normative influence can take place in online environments if there is group saliency (Perfumi et al., 2019), that is, when people identify with a group that defines membership in terms of certain behavioral and attitudinal norms (Terry & Hogg, 1996; Wellen et al., 1998). Social media influencers could, therefore, utilize normative persuasion if their followers perceive to belong in the same social category with the influencer. This is a notable opportunity particularly when considering the format and style of most social media content – story-driven and emotion-laden images and videos (M. T. Lee & Theokary, 2021).

Based on the reviewed literature and the particularities of the Corona Facts campaign, the research questions of this paper are: 1) How do social media influencers interpret the campaign’s suggestions for communicating about COVID-19 pandemic? 2) How do their followers react to the campaign messages? 3) How do influencers interact with their followers on the topic? Next, we describe how these questions are empirically scrutinized.

Methodology

The Corona Facts campaign started in March 2020 quickly after the pandemic hit. The campaign invited all Finnish social media influencers to participate but also selected some influencers for increased collaboration. None of the influencers however received financial compensation for posting for the campaign. The marketing agency responsible for the campaign created an information center for the influencers, sent out sixteen newsletters and organized two webinars on the pandemic and the imposed safety measures and restrictions. The influencers were asked to post about this fact-checked content and use a hashtag #coronafacts.

We collected all Instagram posts that mentioned the campaign hashtag and their comments on May 4, 2020. For data collection, we used 4CAT, a tool provided by Digital Methods Initiative of University of Amsterdam, that can be used to scrape incoming posts on various forums, image boards and web platforms (Peeters & Hagen, 2018). At that time, the tool allowed scraping feed data from Instagram, which has however been disabled since. Thus, we have data ranging from March 7 to May 4, 2020, during which nine newsletters were sent out. Instagram Stories – short videos or photos that by default disappear after 24 hours – were collected manually if they had been saved to Instagram Highlights by the influencer. In the case of the Corona Facts campaign, three influencers had saved their Corona Facts Stories.

We sorted out posts by public organizations and posts by individuals with less than 1000 followers. This resulted in 96 Instagram posts, 108 Stories and 1097 public comments made to the posts. Multiple story photos or videos were considered one story if they formed one clear entity (e.g., a longer video cut in shorter bits). Figure 1 presents the timeline of the Corona Fact posts (excluding Instagram Stories because their exact date is not provided).

Figure 1. Timeline of Corona Facts posts and comments and the publication of the newsletters (dotted lines).

We used qualitative content analysis to see how the influencers interpreted the communication suggestions (according to the content and schedule of the newsletters) and how their followers reacted to these posts. Posts and comments were analyzed separately. In case the influencers interacted with their followers, we also analyzed the content of these interactions. We followed the procedure of inductive content analysis (Krippendorff, 2004; Kuckartz, 2014). After having collected the data set, a classification scheme of thematic categories was created, discussed and revised by two researchers using a subset of the data. While we were interested to see how strictly the influencers followed the campaign suggestions for communicating about the COVID-19 pandemic, the classes were allowed to generate from the data. Then, the full dataset was classified using the scheme by one researcher. The final classifications were cross-checked and discussed by the research team, and they are presented in Table 1. While the main goal of the analysis was to qualitatively understand the nature of the data, the share of each class is displayed to give the reader an understanding of their relative prevalence.

Table 1. Categories of qualitative content analysis

Influencer posts (feed, Stories)		N	%	Example message
Instructions, general facts	General instructions about how to behave during the pandemic, and general facts about the virus and related issues	73	36%	”Hey, we have a message to you Let’s stay home. Let’s wash our hands. Let’s respect the professionals who are fighting for us. Let’s save lives. We will get through this together. #flattenthecurve #stayhome #covıd19 #corona #coronafacts”
Direct sharing of campaign messages	Direct sharing of campaign messages	45	22%	”Finnish Food Authority informs: There is no research evidence that dietary supplements would protect against the corona virus or cure the disease it has caused. --- #coronafacts #foodsupplement #health #medicine #bloggers #influencer #influencermarketing #healthblog”
Showing an example	Demonstrating or describing how to behave during the pandemic and how to cope with it. Peer support.	32	16%	”Tomorrow we’ll go outdoors only with our own family, like we did today. Luckily we’re many, we live close to nature, and the spring is pretty. We’ve decided that we don’t meet any friends, even outdoors. I wrote a blog post about this decision and I included some tips for how to communicate this to kids’ friends!”
Campaign	Information about the Corona Facts campaign, informing about the reponsibilities of social media influencers	28	14%	”Hey you, we’re now needed @finnishgovernment has put its faith on social media influencers and asks us all to take part in relable communication about the corona virus. --- Regardless if you have 200 or 20k followers – all voices are now important to speak for urgent matters.”
Emotions	Sharing of emotions, such as concern, humour, hope, respect	26	13%	”If something about these dreaded times is to stay, I wish they were the sense of community and respect for healthcare professionals. #flattenthecurve”
		204	100%
Comments (either by a follower or the influencer)
Positive reaction	Supportive messages, positive or supportive emojis, giving thanks, relationship work	687	63%	”More of these, please ”, ”Have a nice May Day and remember to eat lots of donuts ”
Information sharing	Sharing the same opinion, sharing information, answering questions	124	11%	”Mortality rates vary widely between countries and are largely affected by how badly the health care system is saturated. Washing your hands should be a basic thing, but yes, even I have paid more attention to it now.”
Advice	Giving advice, tips, sharing new ideas	71	6%	”You should look closely at what model to buy. A mere rectangular strip of fabric with rubber bands fits poorly. At least a couple of pleats should be at the edges to allow for stretching.”
Concern	Concern about the disease or its effects on oneself or other people, or about the restrictions affect the society or the economy sad reactions	42	4%	”I’m worried because my mediaction lowers my immunity”, ”That is truly irresponsible and selfish”
Question	Question about the virus, how to behave in certain situations, or about something that the influencer has posted (sincerely)	28	3%	”What is your opinion as a doctor, should meetings for less than 10 people be cancelled? What about two-person meetings?”
Debate	Contesting, pondering about the truth, irritation towards others	26	2%	”We cannot afford to keep everyone alive forever. According to the oath of the doctors, every person who dies is a failure. Should we start looking at the situation differently?”
Unrelated	Comments that do not fit to the context, links to other users, only hashtags	119	11%
		1097	100%

The unit of analysis was one post or one comment. In case of posts, the analysis focused particularly on textual or oral content as opposed to visual content. In case of comments, it was sometimes needed to go back to the original context to understand its (likely) meaning. While influencers can be considered public figures and their posts are publicly open for anyone to see, we hid the identity of the commentators in the analysis phase. The only exception were comments that were from the influencer who had posted the original photo or video. To illustrate the findings, we present example messages from the data. All messages are however translated from Finnish to English, which works as fabrication particularly in case of the comments (Markham, 2012).

Results

Social media influencers’ means of communication

In total, there were 204 posts by social media influencers on Instagram that mentioned the hashtag #coronafacts (96 posted to the feed and 108 to the Stories section). Sixty-six different influencers took part in the campaign and, notably, a large share of them were women. While it is not possible to identify the gender of all the influencers (some are not personified at all), it could be estimated that at least 85% of them identified as women. There were both big influencers with tens of thousands of followers as well as small influencers with a few thousand followers.

Instructions, general facts

The most typical posts by the influencers concerned general facts and instructions on how to deal with the virus, the pandemic, and the following restrictions (37% of the posts). Many posts reminded followers about proper hand and coughing hygiene and asked everyone to stay at home and avoid unnecessary physical contacts. A common hashtag in these posts was “flatten the curve”, which suggested that the healthcare system should be spared from overflow of corona patients. While many of these posts aligned with the information shared by the Corona Facts campaign, they were more general and did not relate to the weekly or even daily updates from the government or the health officials.

Among the photos that attracted the most likes and comments were a photo of 16 influencers arranged in a collage. In the photo, each influencer held a paper with a word, and, together, those words said: “Let’s stay at home, let’s wash hands. Let’s respect the professionals who fight for us, that way we’ll save lives. Together we’ll go through this. Everything will be alright. #flattenthecurve”. The same picture was shared by the participating influencers, but not all of them used the campaign hashtag. The concept of a collage photo was recreated also by travel and pet influencers.

Direct sharing of campaign messages

The next most common category of posts was posts that directly shared the campaign messages (20% of the posts). They were identifiable based on the ready-made visuals and texts provided by the marketing agency on their Instagram account or in their newsletters or webinars, or they were screenshots or shared posts from the website or Instagram account of the Finnish government or the Finnish institute for health and welfare. Unlike usual, these posts did not highlight the posting individual, but they merely forwarded and reminded about important public information. Content-wise, these posts concerned timely information about the virus, instructions on how to prevent an infection, and newly imposed restrictions to prevent the spreading of the virus. A share of these posts were also direct campaign messages and pictures about “Virtual May Day”, an event that helped people celebrate May Day virtually and apart from other people. One influencer for example, shared a campaign photo and wrote: “May Day has not been cancelled – it will be celebrated this year at home and virtually!”, see, Figure 2.

Figure 2. An example post by a social media influencer (direct sharing of campaign messages). Caption shortened. Reprinted with the permission of the influencer.

Showing an example

One way of communicating about the COVID-19 pandemic and the restrictions was showing an example on how to behave and cope (16% of the posts). Influencers for example, told that they had been staying at home, maintained an improved standard of hygiene, or had avoided seeing other people, for example, grandparents or friends.

A strong element in these posts was peer support and consideration of the particular issues one’s followers might have; family-themed influencers posted about things to do with kids while staying at home, and travel influencers discussed how to find adventure in nearby locations. One influencer for example, posted a picture of two kids using a smartphone alongside a caption that said:

Keeping in touch while social distancing Although it’d now be possible to travel to grandma, it’s not yet time for that long-awaited visit. Let’s all still stay at home, keep physical distance and avoid all unnecessary contacts #socialdistancing #stayhome #coronafacts #facetiming #twins #mymamahood

A few influencers also shared their own experiences with the virus, either directly or through someone they know. One influencer had got a severe COVID-19 disease and she shared multiple, detailed stories of her case. With plain black text on white background, she began to describe her condition on her Stories as follows:

I feel insecure about being sick now. Because so few are admitted to corona tests. – I begin to think of this as corona now that I’ve been sick for 14 days and I still have shortness of breath and lung pain. – I think this needs to be treated as coronavirus only in case that’s what it is.

Campaign

Some posts discussed the Corona Facts campaign itself or encouraged influencers to use their voice and influence to share reliable information about the coronavirus (14% of the posts). These meta-level posts were more often Instagram Stories than posts of other categories, and many of them were from influencers who were somehow affiliated with the marketing agency that was responsible for the campaign, albeit did not receive a direct compensation for posting about the pandemic.

For example, an influencer, who was involved in coordinating the campaign, posted several Instagram Stories of how the campaign unfolded and how it was put together. In this set of videos, after having appeared on television to talk about the campaign, she says how lovely it was to meet a few people and “get the sensation that maybe the world will be saved, and life goes on and, what we do, matters”.

Emotions

As opposed to sharing instructions or facts about the pandemic, some influencers posted emotion-laden pictures or texts (13% of the posts). These posts were either somber, hopeful, or humoristic by nature. Some paid respect to doctors and nurses and other essential workers who were not able to practice social distancing. An influencer who had recovered from the coronavirus disease, posted a picture of her with a donut and wrote:

I got 2 boxes of donuts for Nurses who took care of me when I was in the hospital - in the ward and in the ICU. And a box for home as well. – I wanted to do this as I truly appreciate the good care and empathy I received at [the name of the hospital].

Some expressed frustration with people who did not obey the restrictions or with conflicting rules and practices. One influencer, for example, described in her Stories how some of her friends had flu symptoms but were unable to get tested for COVID-19. She frustratedly described their situation and had captioned the Story “It’s a big problem that not everybody is eligible for a test!”. She also shared her followers’ messages regarding lacking testing resources and its worrying consequences. Despite the critique, the influencer pointed out that even with the smallest flu symptoms one should stay at home, thus complying with the campaign’s key messages.

Humoristic posts were typically memes. For example, one influencer posted a picture of a word search puzzle where all letters created only one phrase, “at home”. The influencer captioned the picture: “The first word you see tells you where you are in April. ” These posts were distinctively not in line with the original idea of the Corona Facts campaign (sharing reliable and factual information about the pandemic), but none of them tried to counter it, either.

Follower reactions and influencer-follower interactions

To the 96 Instagram feed posts, there were 1097 comments made by either followers (76%) or the influencers themselves (24%). In total, seven comment categories were created based on the inductive qualitative content analysis: positive reaction, information sharing, advice, concern, question, debate and unrelated. When comparing the distribution of different kinds of comments to different influencers, it is notable that the three medical doctor-influencers, who participated in the campaign, received 305 comments to nine posts, which is much more than other types of influencers on average. These comments were also more varied by nature: concern, debate, information sharing, questions, and advice were relatively more numerous in the comments of medical doctors than other influencers. Positive reaction was the most typical comment to other than medical doctors.

Positive reaction

By far the most typical comments to the campaign posts were positive or supportive comments by the followers (63% of the comments), which reflects the generally positive online culture on many Instagram accounts. Most often, followers showed their support with positive emojis like hearts, smiley faces, clapping hands or thumbs up. In this category, there were also short supportive comments like “Lovely ” or “Amazing photo and wise words ”, but some put more effort to notice the work the influencer was doing for the cause.

Most influencers also interacted with followers in their comment section, and most of the comments by the influencers were also positive reactions, such as thank you messages or positive emojis like hearts or smiley faces. In some cases, followers shared their own experiences with the coronavirus, whereby the influencers typically responded in a supportive manner.

Information sharing

In addition to the positive reactions, there were more conversational comments that were labeled as information sharing (11% of the comments). These comments were also typically positive but more informational, with the follower wanting to add to what the influencer had said. For example, an influencer, who is a medical doctor by profession, had posted a photo of herself at work. With the aim to relieve anxieties related to the pandemic and fears about the disease, she detailed in the caption the Finnish healthcare system’s ability to patients’ needs. In the comments, a follower – possibly a professional, too – added:

That is a great remark that in the intense care they know how to manage these situations. Also operating rooms and control units can be converted into intense care units. We have started to prepare for this a long time ago and the situation is being monitored closely.

Some influencers responded to comments by sharing information about their personal life and thoughts. These mini conversations (usually only two or three subsequent comments) rarely related to the pandemic but drifted to exchange of personal news. For instance, a travel influencer asked in her photo, what kinds of travel dreams her followers have. One follower commented that she was hoping to travel back to Finland for Christmas. The influencer commented back:

Fortunately, there is still time for Christmas. Hopefully then it will be ok to travel longer distances so you can get to Finland. We are invited to a wedding in Thailand in January, so we should probably buy flights before all the planes are fully booked

Advice

In the comments, both followers and influencers gave advice, shared tips and corrected potentially false information (6% of the comments). Corrections were made for example, if someone (either an influencer or a follower) expressed views that were not in line with the official information provided by the government or health officials. For instance, as a comment to an influencer’s post where she had reminded proper hand washing and the use of hand sanitizer, a follower commented:

If you can properly wash your hands, then there is no need to use hand sanitizer. Hand sanitizer is specifically designed for situations where hands cannot be washed, this was the advice of a health professional even in case of corona

Tips were particularly answers to influencers’ questions. For example, a medical doctor-influencer asked her followers how they coped with stress during a crisis, and she got answers, such as: “Exercise, knitting, reading, baking and spending time with the family ”, and “Meditation, yoga, stress observation (not manic but observational), intuitive writing, balance of rest and exercise, as well as regular, individual and clean diet”.

Concern

Some followers expressed concern in their comments (4% of the comments). These comments reflected a variety of issues that the pandemic had caused, concern towards people who do not obey the restrictions or other negative feelings related to the situation. In several cases, the followers commented about a specific group’s inability to follow the restrictions and the following concern or fear. For example, to an influencer, who reminded her followers to avoid meeting elderly people, a follower commented:

My parents are in their eighties and can’t really use any online stuff and even their 20-year-old phone broke down just right. They are missing their grandchildren And it seems they cannot stay at home, they feel like they must go to the store even though their neighbor offered help.

Question

Questions directed to the influencer comprised 3% of the comments. A few questions were not coronavirus-related (questions about how an Instagram live event is organized, questions about travel destinations) but most questions were specific questions posed to the medical doctor-influencers. There were three medical doctors that had participated in the Corona Facts campaign and some of their followers asked rather detailed questions about the virus and how to interpret restrictions. A follower of a medical doctor-influencer for example, asked: “What is your position, should one keep a child at home from kindergarten if one parent is pregnant and the other has asthma?” The doctors diligently answered these questions and encouraged followers to ask questions in the comments.

Debate

Debate was broadly speaking the only category of comments that consisted of comments that somehow challenged an influencer’s argument or view (2% of the comments). These comments represented debate in that they either expressed doubt about the virus or the restrictions, or they firmly supported them in response. Notably the share of these comments was very low and, typically, the doubtful comments were sporadic – followers with similar views did not come out to comment. For example, when a medical doctor-influencer posted a photo that said: “Did you know? Marketing food supplements as a cure against the coronavirus is ILLEGAL!”, she debated with a follower in the comments:

Follower: You can’t be for real

Follower: I guess you can advertise and say they improve the immune system without mentioning the horrible c-word?

Influencer: Well, you can’t advertise like that. I recommend familiarizing yourself with the law! This law has been in place for a long time, but it is being constantly broken and consumers are being misled.

Discussion

This study set out to explore strategic social media influencer communication that the Finnish Prime Minister’s Office executed through a campaign called Corona Facts in the spring of 2020. The campaign aimed to get social media influencers to share urgent information about the COVID-19 pandemic. We wanted to find out how social media influencers interpreted the campaign’s messages, how their followers reacted to the messages and how the influencers and the followers interacted with regards to the messages.

Summary of results

According to the results, approximately one third of the influencer posts originated directly from the Corona Facts campaign or concerned the campaign itself. While this might seem successful sharing of campaign messages, the key tenets of strategic social media influencer communication suggest otherwise. These posts do not utilize all the factors that make influencers effective. Most importantly, these posts do not utilize the influencer’s own content production capabilities and do not fully reflect his or her persona (Enke & Borchers, 2019). Arguably, the posts that displayed the influencer’s own interpretation of the campaign (their own photos and wordings) were better able to speak to the audience, were truer to their visual and narrative style, and more clearly reflected their persona (Enke & Borchers, 2019).

In these types of posts (instructions, general facts), influencers typically avoided going into too much detail and reminded their followers about things that quickly became common knowledge (e.g., wash hands, stay at home). Many of these posts displayed a sense of duty and desire to contribute to common good. This was also the intended ethos of the Prime Minister’s Office when it launched the campaign. Some explicitly pointed out that even though they do not usually post about things like this (serious or sad topics, news topics), they felt that now they can use their platform to share vital public information. This can be seen as a strategy to maintain content coherency and logic (Navarro et al., 2020).

Other approaches in the influencer posts were to show an example and to express and relieve emotions. These posts resembled most typical social media influencer content, with least contrast to their other posts (Pöyry et al., 2019). Showing example was the strongest way to impose normative influence over followers (Kaplan & Miller, 1987; Wood, 2000), which is a unique capability of social media influencers – something that existing research within computer-mediated communication and social influence theory has not considered earlier (c.f. Perfumi et al., 2019). Exhibiting relatable in-group behavior is how the strategic communication practice can distinctively benefit from social media influencers in persuading citizens during a time when social norms need to be drastically affected.

A significant exception to the typical influencer posts in the campaign was the style of influencers who were medical doctors by profession. They shared very detailed information about the virus and precautions for protection. Notably, their followers also reacted differently by commenting and asking questions in great numbers. Here, the influence was based on a more traditional meritocratic basis, and these influencers would have, most likely, posted about the pandemic even without the campaign. Nevertheless, they chose to use the campaign hashtag, which supported and validated the campaign. Otherwise, most followers reacted positively but shortly to ‘regular’ influencers’ posts. It seems that followers also wanted to pitch in and support the cause, which highlights the followers’ role in co-creating the influence of the influencers (Munnukka et al., 2019; Reinikainen et al., 2020).

Theoretical implications

While research literature has theorized the role of social media influencers in strategic communication (Enke & Borchers, 2019; Luoma-aho et al., 2019), there has been a lack of knowledge regarding the dynamics of such communication campaigns. A first-in-the-world example of engaging social media influencers in pandemic communication sheds light on the future trends of strategic communication theory and practice. The present research shows how social media influencers can cater their followers information and social norms that are of strategic relevance to an organization through their authentic voice and intimate follower relationships (Berryman & Kavka, 2017; Enke & Borchers, 2019; Luoma-aho et al., 2019; Pöyry et al., 2019).

From the perspective of the two-step flow of communication model (Katz, 1957; Lazarsfeld et al., 1948), the role of most of the participating social media influencers was the opinion leader who participated in forming the social norms regarding the pandemic. Even though they most often only passed on general information that was also shared in the news and elsewhere (that most people could not avoid getting exposed to), their role was to show that, in their social context, it is the norm to avoid social contacts, work at home, avoid traveling, and maintain an elevated level of hygiene. Most of the participating influencers did not typically post about public health matters but, by posting about them, they made their point. This shows that the process of curation (Thorson & Wells, 2016) is critical not only for what information people get exposed to but also by whom – the concept of “influencer curation” is thus proposed as a distinct way how information flows from a public organization to citizens, and what kinds of social meanings it contains.

The results therefore highlight that even though most influencers used informational influence in their posts (sharing general information and advice regarding the pandemic), the normative aspect is present through their persona and brand (Enke & Borchers, 2019). Through the associations attached to them, they can show that people, who identify with their in-group (Terry & Hogg, 1996; Wellen et al., 1998), should pay attention to the issues they post about – even if the post content is far from their usual expertise and domain. The brand of the influencer was visible through their usual esthetics and linguistic style (Enke & Borchers, 2019; M. T. Lee & Theokary, 2021) even if the topic was unusual. Strong and explicit informational influence was exercised by medical doctor-influencers who can act as an important information source to many people. These individuals portray a natural and context-fitting social media presence for a medical doctor (c.f. Brown et al., 2014).

The distinction of these two types of social media influencer groups, comparable to opinion leaders of narrow and/or general expertise (Wang & Li, 2016), has not been identified in the literature of strategic social media influencer communication before. We argue that the distinction affects how an organization can enable strategically important conversations that help it reach its goals (Zerfass et al., 2018); Influencers with narrow, topic-specific expertise can legitimize a strategic communication goal of an organization and provide a venue for informational interaction. Influencers with general expertise, together with their followers, can help mold social norms around the topic by taking part in the pursuit of the goal.

Practical implications

The results carry significant implications for organizations with urgent communication needs that aim to reach niche audiences through influencers. Two groups of influencers emerge: those with topic expertise and credibility (medical doctors in this case), and those who can affect relevant social norms and provide emotional peer support. The latter group is most likely hesitant to participate in public health communication campaigns, which should be accounted for in campaign design.

While the gender distribution of Instagram users globally is fairly even, young people between 18 and 34 years use the service more frequently than others (Statista, 2021). In this study, the influencers who voluntarily took part in the Corona Facts campaign were more typically women than men, and some of their topic areas (beauty, family) are more popular among women. It thus seems that public health campaigns with Instagram influencers are most effective in cases where young women and families need to be addressed.

It is still worthwhile to ponder whether the campaign succeeded in reaching “those who are difficult to reach through traditional channels”, as the Prime Minister’s Office hoped (Henley, 2020), but did it act more as an amplifier of the desired message attached with the social norms the influencers are able to create? In the current case, the influencers were able to promote the “correct” way of thinking and talking about the coronavirus and showed how staying at home could be presented as the desired mode of being. Thus, influencers might be the most effective in communicating about crises and health-related topics that require fostering a new social norm. If true hard-to-reach audiences need to be addressed, it is advised to collaborate with carefully selected social media influencers to attain that goal.

Limitations and future research

As with any research, there are limitations that need to be addressed. First, due to changed data source principles of the tool used for data collection, our data does not cover campaign messages posted at the end of the campaign (5.5.–18.6.2020). This is however not deemed problematic as the great majority of the campaign messages were posted in the earlier months of the campaign, which was also the time of the most intense pandemic restrictions. Second, Instagram Stories have become a popular way of consuming Instagram but data on Stories need to be collected within 24 hours from their publication. This was not done in this study, and we have data on Stories only if they were saved to the Highlights section. There are some ethical considerations attached to collecting Instagram Stories that are not saved. If a person intends the content to disappear after 24 hours, should a researcher collect such data without explicit permission to do so? Bainotti et al. (2021) however suggest that public, yet ephemeral (disappearing) content could be treated in the same way as any other public social media content; The platform policies do not guarantee the permanent disappearance of content anyway.

Another limitation relates to the way people interact on Instagram. Besides commenting on a photo or a video on the feed, people can also interact via direct messages that are private. These messages are naturally beyond a researcher’s reach, but the ability to communicate also via direct messages might affect what kinds of comments are being posted publicly. In this study, positive reactions and other positively oriented comments were numerous while negative and critical comments were rare. It might be more typical to post a positive comment publicly but send a negative comment privately. Future research could collaborate with influencers and study anonymized direct messages influencers receive from their followers (comparable to fan mail) to understand how followers perceive their content without the pressures of public visibility. Another limitation with regards to the comments relates to the possible content moderation performed by the influencers (c.f. Ruckenstein & Turunen, 2020). This is another factor that might create a bias towards more positive comments in the data.

Finally, the context of study is likely to affect the results. When the COVID-19 pandemic hit Finland in March 2020, the consensus was that excess illnesses should be avoided, and the functionality of the healthcare system should be protected by social distancing. The government swiftly posed several restrictions for example, on traveling, businesses, and schools, and the government was very popular during this time (Yle, 2020). The contrast to the pandemic restrictions in neighboring country Sweden was, for example, sharp. Thus, it could be that in Finland influencers were unusually willing to take part in a government-initiated communication campaign. Research is therefore needed to understand more thoroughly the cultural, social, and political aspects that affect influencer participation in public health campaigns.

Acknowledgments

The authors are grateful for the financial support from Helsingin Sanomat Foundation (project Unconventional Communicators in the Corona Crisis) and Academy of Finland (decision number 320373).

Disclosure statement

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

Additional information

Funding

This work was supported by the Academy of Finland [320373]; Helsingin Sanomat Foundation [Unconventional Communicators in the Corona Crisis].