COVID-19 telephone contact tracing in Flanders as a “contested” new genre of conversation: Discrepancies between interactional practice and media image

1. Introduction

During the COVID-19 pandemic in Belgium, the media played a substantial role in communicating information to the public. Mainstream media were the main channel through which the population was informed about figures and numbers relating to, for example, infection rates, but also about the government's strategies for containing the pandemic, ranging from micro-level safety measures such as mandatory face masks, to macro-level measures such as instilling limited “social bubbles,” contact tracing endeavors and vaccination strategies. In a number of respects, the media not only determined which information reached the general public and when, but also influenced how relevant information and reported practices were framed, and by implication interpreted by the general public.

This paper considers the media's portrayal of one aspect of the government's risk management strategy: centrally organized telephone contact tracing. We report on analyses made as part of a 1 year applied conversation analytic and pragmatic study conducted at Ghent University and the University of Antwerp in collaboration with the Flemish Agency of Health and Care. In doing so, we examine the image generated by the media of such telephone contact tracing and how this image evolved throughout the pandemic. Most importantly, we demonstrate how this image contrasts with the interactional reality.

This paper starts with an outline of how centralized COVID-19 contact tracing in Flanders was organized, and the role of telephone contact tracing in the government's strategy for containing the pandemic. We then briefly illustrate the important role of the media by referring to relevant literature. Next, we briefly outline the various (interactional) functions we identified in the COVID-19 contact tracing call. Following this outline, this paper analyzes how precisely contact tracing was reported on in the media, specifically in larger Flemish newspapers. Based on this analysis, we then discuss how the elements of contact tracing covered by the media relate to the identified (interactional) functions. A comparison of these two analyses demonstrates how the media have cast a distorted and one-sided picture of contact tracing. While our interactional analysis points to a complex set of communicative functions, including more practical functions such as giving and receiving information as well as the contact tracing telephone conversation being a care conversation with empathy taking up a central role, media coverage on this type of interaction is characterized by a lack of reporting on the supportive nature of these conversations and a stark focus on the more information-processing and organizational elements. This image may have had consequences for the efficacy of contact tracing, given how for large parts of the population the contact tracing telephone call was a new genre of conversation which was introduced to citizens mostly through popular media and press coverage in Flanders, and which depended to a large extent on voluntary participation and trust by the public.

2. COVID-19 telephone contact tracing

Contact tracing came into existence in the early 20^th century, within the specific context of Syphilis and other STDs (Green et al., 2001). It is a commonplace practice to trace contacts in the context of HIV/AIDS, where it is sometimes referred to by means of the term “partner notification” (Hyman et al., 2003; Tomnay et al., 2005). Another medical context in which contact tracing is a common endeavor is tuberculosis, where contact tracing constitutes “in-depth interviews” with the infected person as the first stage (Fortuinet al., 1998; Begun et al., 2013). COVID-19 contact tracing, then, appeared in various national contexts during the pandemic. Among other countries, Belgium, the United Kingdom, France, Sweden, and the United States used it as a strategy to mitigate and contain the risk of the rapidly spreading illness (Jacob and Lawarée, 2021). Belgium's contact tracing strategy for COVID-19 has arguably been a unique case, as it entailed an automatic coupling of contact tracing activity to a central database of infected persons, as well as a massive switch from in-person interaction to “telephone contact tracing.” The practice was implemented across the three regions of the country.

2.1. Contact tracing endeavors

Part of the Belgian government's strategic approach to containing the early COVID-19 pandemic was establishing a contact tracing system through which all infected citizens and their contacts at risk of being infected could be traced. In addition to the implementation of a contact tracing app, networks of COVID-19 contact tracing call centers were established in May 2020 and governed at the level of the Flemish, Walloon and Brussels Capital Regions. These contact tracing call centers varied in size and capacity because of continual fluctuations in infection rates that called for volatile downscaling and upscaling operations. Such a contact tracing system not only allows one to track and trace the spread of the virus, it also allows citizens to be informed that and for how long they need to isolate or quarantine, and what precisely this entails. Contact tracing initiatives were also developed at a local, city level. Although there was some collaboration between central and local contact tracing, most local initiatives were developed separately and functioned independently from the federal and regional initiatives.

The focus of this paper lies on the type of contact tracing organized at the regional level, more specifically in Flanders. Such centralized contact tracing mostly took the form of systematic telephone conversations conducted in call centers; in a limited number of cases, it included field agents who performed home visits. The procedure for telephone contact tracing was as follows: when an infected person tested positive for the COVID-19 virus, this person was contacted by a contact tracer (CT) over the telephone, asked about their contacts and provided with instructions and other information regarding prevention measures (De Timmerman et al., 2022). The infected person is what we call the index patient (IP). The contact tracer asks the index patient about recent contacts, when and where the encounters took place, and so on. In cases where the index patient is not reachable via telephone, a field agent is sent to their home to carry out the contact tracing in person. In the next stage, the index patient's listed contacts are each contacted in their turn. The next step is for these contacts to self-quarantine and get tested. When a contact tests positive, the entire process is repeated, now with the contact as the index patient. Our study is limited to the first-stage telephone calls.

The contact tracing phone calls are conducted on the basis of a script, which is designed differently for calls with index patients, calls with low-risk contacts, and calls with high-risk contacts. This script has been integrated in a computer program with slots to be filled out, path dependency, and set questions. We specifically focus on the telephone conversations of one script type, “1A,” viz. those with IPs.

2.2. A genre of conversation that was new to the general public

Arguably, COVID-19 contact tracing calls can be considered a new genre of conversation. Even though contact tracing had been around in various forms and variants, internationally and within the country, the COVID-19 contact tracing telephone conversation as a genre emerged more or less overnight. It emerged as a centralized practice which differed from pre-existing contact tracing conversations to contain infectious diseases such as tuberculosis. One novelty was its implementation on a large, population-wide scale in response to a global pandemic. This matter of re-scaling and the urgency which marked its introduction sets it apart from its more low-scale predecessors. The genre was also new because it was still ‘unknown' to large parts of the general public.

Yet, at the same time, contact tracing in fact turned out to be a genre characterized by various affinities with already existing forms of discourse/interaction, such as patient-centered front line medical consultations. As an anticipated practice, it had even already been sketched in policy papers and was waiting to be activated. More specifically, the genre of COVID-19 contact tracing telephone conversations has ties with medical and institutional interactions, but also with call center interactions, given that the task of contact tracing was assigned to commercial call centers. In particular, regional centralized contact tracing in Flanders emerged as a collaborative endeavor of a consortium of government institutions including the Flemish Agency of Health and Care, call center companies, and the health insurance organizations.

The Belgian case of contact tracing telephone conversations has some notable features. Partly as a result of its urgent and rapid organization, and unlike in some other countries, contact tracing agents were not required to have a (para)medical schooling or background. Moreover, whereas one might expect the telephone conversations to be similar to medical interactions, because of its outsourcing to commercial call centers, the genre is influenced by elements of commercial call center conversations as well. Since contact tracing was a new concept to most Belgian citizens – i.e., a new genre of conversation with which most citizens were confronted only a limited number of times over the course of the pandemic – the public introduction and mass mediated image of contact tracing endeavors arguably played a crucial role in its overall reception, functioning and success rate.

Mainstream media in particular arguably had a substantial amount of influence on the public image and opinion of contact tracing in this regard. Interestingly, even though telephone contact tracing was a crucial measure taken by the government to reduce and contain COVID-19 infections nation-wide, only limited efforts were made to direct this and communicate the function of contact tracing to the public or inform/educate the public in a positive way. Moreover, even though the Flemish Agency of Health and Care stressed the care-oriented nature of these telephone conversations, no explicit campaigns were conducted to promote this preferred stance.

According to O'Connor et al. (2021), the COVID-19 pandemic can be considered an opportunity to positively highlight science communication in society. However, as their analysis of the Irish context points out, such promotion is not without risk. Specifically, the risks they identify are “feeding public alienation by purveying deficit model assumptions, reinforcing stereotypical images of scientists, and intensifying the politicization of scientific statements” (O'Connor et al., 2021, p. 19). Reminding of these risks, a study of the role of the Canadian print media in the public portrayal of essential health care services found that print media were predominantly “descriptive and uncritical” in their portrayals of public debate and institutional policy-making (Ogbogu and Hardcastle, 2021, p. 3). A case study by Filardo-Llamas and Perales-García (2022) on representations of the EU in Spanish media during the pandemic identified three frames used to both implicitly legitimize the EU and to do the opposite for certain actions by European institutions: a moral frame, including “calls for solidarity between member states of the EU” (293); an economic frame, including “potentially harmful economic consequences” (244); and a frame of conflict, including war metaphors (245). One study by Mroz et al. (2021) specifically targeted media representations of remote GP consultations in the UK. Their findings illustrate the presence of various themes such as technology and war metaphors, yet specific functions and interactional affordances of remote consultations are not represented in their analysis. Moreover, they conclude that more positive communication regarding the shift from face-to-face to remote medical practice is necessary to restore the trust of the public. These findings from different contexts illustrate a wide array of risks or negative effects linked to media representations during the COVID-19 pandemic. These examples suggest that the study of contact tracing telephone conversations in Flanders can only benefit from an analysis of media representation to understand how contact tracing was introduced, framed and perceived by the public. It is also important to consider risks or flaws tied to the frames that were used to represent contact tracing by the media.

To get a concrete sense of the image that was communicated to the public by Flemish media, this paper examines the image of COVID-19 telephone contact tracing as constructed in mainstream press and compares this to actual interactional practice. To conduct the comparative analysis of media image and institutional practice, we rely on our analysis of a large dataset of recorded contact tracing interactions and a dataset of newspaper articles published during the pandemic. The next section outlines these data and covers the relevant methodological steps that were taken to facilitate the analysis and discussion in Sections 4 and 5.

3. Data and methodology

This paper reports on data collected within the context of a 1-year COVID-19 research project funded by the Research Foundation Flanders (FWO). Even though the main focus of the project was on interactional practice, i.e., to identify and optimize the interactional dynamics in contact tracing telephone conversations in Flanders, the project was transdisciplinary in nature through a collaboration between a team of (socio)linguists, medical experts, epidemiologists, a moral scientist and a representative of the Flemish Agency of Health and Care, and one of the private call center companies responsible for COVID-19 contact tracing in Flanders. The project involved a collaboration between academics, practitioners, and government representatives, all of whom were involved from the start in the formulation of the research questions, methodological approach and desired project outcomes.

The starting point of the project was a number of interactional problems which hinder the effective functioning of contact tracing in Flanders, and which emerged in the project team's early conversations with the stakeholders: e.g., calls remain too short; the talk is script-dominated; reluctance exists to divulge necessary information; there is a lack of rapport between interlocutors. In meetings in the early phases of the project, medical professionals from the Agency of Health and Care also voiced the need to accomplish call center conversations as “care” conversations in which interactional challenges are managed more adequately. As public support and public perception of call center contact tracing were rather negative in 2020 and COVID-19 contact tracing depended on voluntary participation, the project was thus premised on the idea that one pinnacle of success to remedy some of the problems reported by the stakeholders is a call center agent who is able to establish trust and maintain rapport with the index patient during the contact tracing interaction so that the experience is more positive, effective participation is ensured and vital information can be collected and transmitted. The project's central goal was then to diagnose the “interactional” state of telephone contact tracing on the basis of an interactional analysis of a corpus of recorded calls and to formulate evidence-based recommendations to improve contact tracing practice in Flanders.

The data and results reported on in this paper relate to three datasets (1) a corpus of 100 contact tracing calls (from the first phase of the project) conducted in Dutch between a contact tracer and index patient recorded between late 2020 and early 2021; (2) a series of interviews conducted between contact tracers and a researcher working on the project and (3) a corpus of Flemish newspaper articles published between March 2020 and May 2021 and between late 2021 and early 2022. Audio-recorded data (calls and interviews) were collected with informed consent and subsequently pseudonymized by transcription; all (pseudonymized) transcriptions were analyzed using NVivo. The analytical focus of this paper lies predominantly on the third dataset: the corpus of newspaper articles. To compare the analysis of this dataset to the interactional reality, we draw on a brief outline of findings from our interactional analysis. A more elaborate account of this interactional analysis can be found in an earlier publication: De Timmerman et al. (2022).

As the above suggests, these three datasets were analyzed separately. The analytical approach for the analysis of the recorded contact tracing calls was broadly discourse analytical in nature (Candlin and Candlin, 2003) and relied on both interactional sociolinguistic (Rampton, 2019) and applied conversation analytic methodologies (Schegloff, 2007; Hutchby and Wooffitt, 2008; Antaki, 2011; Slembrouck and Hall, 2011, 2019) with a focus on the turn-by-turn unfolding of linguistic interaction. Such sequence-based micro-interactional analysis allowed us to reveal relevant conversation-technical aspects of the phone call interactions and to identify why and how particular sequences count as “strained” or “fraught” as well as sequences which mark degrees of rapport, trust and efficient and relevant information exchange. This micro-interactional analysis relied on a list of constitutional determinants of conversational interaction as a point of departure, oriented to both the speaking and listening behaviors of tracers and patients, with particular attention to reciprocity and responsiveness. The determinants were: (i) turn-taking dynamics (incl. turn signaling devices, overlapping talk, interruptions, a typology of question-answer sequences, follow-up questions), (ii) topic management (incl. the role of the script, topic initiations/digressions by the patient, narrative turns in the talk), (iii) aspects of formulation (incl. the formulation of intrinsically face-threatening acts, responses to signals of reluctance, anxiety; with specific attention to “delicate meanings” and “sensitive topics”) and (iv) face work (respect for the index caller, reassurance of patients, avoidance of face loss, positive face work which stresses the importance of contact tracing). Detailed iterative coding of these interactional foci allowed us to arrive at qualitative and quantitative analyses of the contact tracing practice, and led to the identification of the key functions fulfilled by a contact tracing call.

Semi-structured interviews were conducted by the second author with 22 different contact tracers during the first phase of the project. Questions concerned the participating contact tracers' experiences with the conversations and their opinions on topics such as the adoption of a care stance during the calls. The interviews were analyzed by the third author by means of iterative coding though content and thematic analysis (Fereday and Muir-Cochrane, 2006; Corbin and Strauss, 2008). Themes were distilled across the different conversations by focusing on similarities and differences between the contact tracers' opinions and reported practices represented in the interviews.

The corpus of Flemish newspapers was collected by the first author and contains 76 Flemish newspaper articles published during the pandemic (March 2020 – February 2022), all of which mention COVID-19 telephone contact tracing. The corpus contains articles from three main quality papers. As the fourth estate, the (political) stance typically taken by these papers is generally supportive of official government policy and approach, but they also provided space for skeptic and libertarian/anti-big state voices (especially during the pandemic outbreak). The corpus is limited to Flanders and its three quality newspapers: De Standaard (DS), De Morgen (DM), and De Tijd (DT). Because of scarce coverage of COVID-19 contact tracing during the summer of 2021, we primarily consider the first year of the pandemic (i.e., between March 2020 and May 2021 – 54 articles) and more recent developments (i.e., late 2021 up until early 2022 – 22 articles). The corpus was analyzed using document analysis (Bowen, 2009). This entails that the articles were subject to iterative content and thematic analytical coding (Fereday and Muir-Cochrane, 2006; Corbin and Strauss, 2008), and specifically relevant excerpts were highlighted after which all highlighted elements were then compared so as to generate clusters of themes which appear across the different articles. These thematic formulations became more sophisticated and nuanced as the data was skimmed through for relevant elements three consecutive times. Lastly, then, the relationships between the themes were identified and mapped visually (cf. Section 5).

In this paper, the empirically identified interactional and conversational functions and features are compared to the depiction of COVID-19 telephone contact tracing in Flemish press coverage of the pandemic (i.e., the image civilians were regularly confronted with). Section 4 below reports on the interactional functions identified in the contact tracing call, with a predominant focus on the interactional manifestation of empathy or care in our corpus of contact tracing conversations. It is followed by Section 5, in which we compare these findings to the results of our analysis of the portrayal of contact tracing in the media.

4. COVID-19 contact tracing in practice: Multiple communicative functions and the caring stance

The interactional analysis carried out as part of the larger research project uncovers several different interactionally achieved functions of the contact tracing telephone conversation, which the CTs are tasked with (Slembrouck et al., 2021; De Timmerman et al., 2022). A visual overview can be found below (cf. Figure 1). Based on our analysis, we discovered that the interactional practice of contact tracing covers five functions in these telephone conversations: Contact tracers are not only expected to (i) gather information on an index patient's symptoms and contacts and (ii) provide instructions regarding quarantine, isolation and other safety measures; they are also expected to (iii) perform the two “core” functions efficiently: (iv) while transversally maintaining an individual, patient-centered, caring stance and communicate empathetically throughout the interaction. Finally, (v) they need to approach these functions in ways which align with their role as representatives of government policy. The latter is less straightforward than may appear at first sight, as government policy itself was heavily debated and contested during various stages of the pandemic.¹

FIGURE 1

Figure 1. Functions of the contact tracing call, as identified through systematic interactional analysis.

Arguably, the transversal function of empathy or adopting a caring stance was one of the functions on which much, if not the most, importance was placed. The Flemish Agency of Health and Care's stress on the importance of a caring orientation in contact tracing conversations was repeatedly emphasized in various meetings with the project team. Moreover, we do not only see this care function reflected in the interactional dynamics noted in the corpus, but also, for example, in the interviews with the contact tracers and in the focus on empathy in the contact tracer's training program. One of the most striking examples from our interactional corpus of the contact tracing conversation not only being information-focused, but just as much being care-oriented is Excerpt 1 found below. This particular episode spans from turn 34 of the conversation until turn 145 (which is, especially in comparison with the other calls in our corpus, strikingly long) is held between two women. The IP is between 70 and 80 years of age. Below we have included two brief segments from this lengthy episode which highlight the CT's care orientation.

EXCERPT 1

Excerpt 1. IP = woman, aged 70–80.

In this case, the IP introduces several topics not included in the CT's script to elaborately describe her current situation and how miserable she feels. Rather than immediately deflecting the topic or listening only briefly and then returning to the script, the CT allows the IP to take all the time she needs to voice her issues and feelings. Many of the discussed issues do not even relate to the COVID-19 infection at all. Yet, the older woman needs support in the form of someone listening to her troubles and responding with affirmations, which the CT picks up on in this case and provides first, before returning to the more information-oriented part of the script.

The project's overarching analysis of concrete contact tracing conversations underlines the complexities of communicative work that is responsive, is accomplished sequentially and which is partly to be understood in affective and care-centered terms. Yet, when we compare the results of our analysis of the corpus of actual contact tracing interactions with the image of contact tracing in the mainstream press, we find a vastly different picture overall.

5. COVID-19 contact tracing in press coverage

Over the course of the pandemic, the mainstream press was one of the main channels through which the public was informed about anything related to contact tracing – up until then a genre of conversation and type of contact with the government unknown to most. This section covers our analysis of how Flemish newspapers portrayed telephone contact tracing throughout the first two years of the pandemic in Belgium. That is, what type of image they generated and how certain issues were framed (Lakoff, 2006). This section is divided into an overview of the media image in the early pandemic on the one hand, and more recent developments on the other.

5.1. Early pandemic outbreak (March 2020 – May 2021)

Overall, the vast majority of COVID-19 contact tracing coverage in Flemish newspapers reported on the more practical, organizational and more narrowly information-processing aspects related to contact tracing via telephone in the early pandemic. For a schematic overview of the interrelated connections between these topics, see Figure 2. Privacy issues and the gathering of personal information received much attention. Other issues reported on were the lack of medical schooling required for contact tracers and how much the operation would cost. One further aspect which received a considerable amount of attention was various technical-organizational issues with the contact tracing system and software, short-lived or not. From our analysis, it also becomes clear that in the early pandemic, most of the responsibility regarding contact tracing and keeping civilians safe was placed on the actions of regional and federal governments, rather than on individual citizens (cf. Figure 2).

FIGURE 2

Figure 2. Overview of themes in newspaper articles – Early pandemic.

In the early pandemic, most newspapers published articles in which contact tracing was contested because of the necessity to gather personal information. One article published in De Standaard (DS) discussed this issue at length: it reports on peoples' concern that, for as long as there is no clear legal framework, there is no guarantee of privacy being sufficiently safeguarded. Moreover, in one article, the ability of contact tracing to do so is questioned (DS 7 May 2020, p. 33). In this article specifically, contact tracing is framed as “violations” of privacy. The article concludes with a highly critical comparison of the government's reaction to the 2016 terrorist attacks in Brussels, which is argued to have been similarly “inadequate” and “rushed,” causing “negative effects” years later still. There is no following explanation as to which negative effects they are referring to. Particularly interesting here is the use of battle [in Dutch “(in het heetst van de) strijd”] as a metaphor for the government's strategic reaction to both COVID-19 and the 2016 terrorist crisis. This is reminiscent of Mroz et al.'s (2021) analysis of media depictions of remote consultations in the UK and Filardo-Llamas and Perales-García's (2022) analysis of the representation of the EU during COVID-19 in Spanish media, which both demonstrate the prevalence of war and revolution metaphors in COVID-19 media discourse.

Another criticism expressed in a number of articles is the fact that contact tracers are not required to have more than a secondary education degree or to have any medical training or experience. For this reason, journalists and columnists alike critically address the very limited training (a few hours, it is claimed) contact tracers receive prior to fully starting the job. This criticism is backed for example by the argument that “people will be hesitant to share the necessary personal information – especially via telephone – if they are not sure their privacy will be safeguarded” (DS 7 May 2020, p. 33 – translated).

In articles addressing issues pertaining to privacy and contact tracer training, some attention is in fact paid to interaction, and even to the need for empathy. What should be noted, however, is that these discussions on interaction are largely positive, as criticism often focusses on quantifiable results such as the number of contacts shared or the duration of the call. One of the handful of articles in which interaction is explicitly attended to, introduces contact tracing in early May:

It's not a simple task for a complete stranger to acquire, sometimes intimate, information from a patient, or to convince people to quarantine for two weeks. ‘It's important that the contact tracers build a sense of trust though a relatively short conversation', says Dhaeze [Agency of Health and Care representative]. In the training, the importance of qualities such as empathy, openness and ‘navigating between supportive and guiding listening' is insisted on. (DS 5 May 2020, online – translated)

A second mention of interactional practice can be found in the introduction of this same article. It commences with a list of “do's”: “introducing yourself, staying calm, showing empathy, confirming correctness of information,” and “don'ts”: “eating or drinking during a call, improvisation, sighing” (DS 5 May 2020, online – translated). Although such an interactional description stands out in our corpus, the fact that improvisation is framed as not-done could be understood in terms of a need to remain focused or in other words, “strictly scripted.” Such mentions of interactional practice lacking sufficient nuance could arguably cause issues with rapport, and subsequently even threaten people's proclivity to share personal information. One article published in April uniquely defends contact tracing explicitly, stating that it is not a “Chinese totalitarian technique” or similar to “Stasi practices.” Remarkably, this article also stands out from the rest of the corpus through its explicit framing of empathy as a key function: “[CTs] don't need to be doctors or nurses, they do, however, need to be able to offer advice empathetically” (DS 22 April 2020, p. 7 – translated). Note that the articles we referenced here were all published before or around the onset of telephone contact tracing practice, and that they took the form of promotional interviews with stakeholders and Agency of Health and Care representatives.

Apart from privacy issues, most other articles reflect on the financial aspects of the government's contact tracing endeavors. This financial lens casts light mostly on the ever-growing amount of money invested in contact tracing. A documentary aired on “Pano” – a national critical documentary television program – at the end of 2020, even revealed that one of the call center companies was already on the brink of bankruptcy when it was hired by the government. This piece of information was later echoed in many newspaper articles.

Various articles also addressed financial concerns related to obligatory quarantine for at-risk contacts. As was written in De Morgen (DM): “Nobody likes to be the cause of somebody else needing to self-quarantine for 2 weeks” (DM 1 July 2020, p. 9 – translated). It is argued that index patients might be reluctant to share information on their contacts because obligatory quarantine is unpleasant and inconvenient, and because it may even have dire financial consequences for people in certain professions or financial positions (DM 19 May 2020). Such financial frames remind us of the economic frame Filardo-Llamas and Perales-García's (2022) analysis identified in the Spanish media coverage on COVID-19. In contrast to some articles considering privacy, interaction or empathy is surprisingly not mentioned in relation to these financial issues. This is quite striking, as the arguments provided nearly all relate to personal, possibly even emotional consequences.

Similarly, many articles discussing contact tracing's financial shortcomings critically evaluated it by referring to the many technical difficulties which characterized its first 3 months. One article sums up the issue as follows: “Don't forget that [the contact tracing system] was built in 2 weeks. It was good enough to get started and to train contact tracers, but there is still much work to be done” (DS 30 June 2020, p. 8 – translated). In this article, contact tracing is framed as a “complex machine,” which requires “lots of dragging and pulling.”

Overall, what these articles arguably (implicitly) reflect is a strong sense of responsibility being placed on the government for combatting the COVID-19 pandemic. The expectations of how the government should be carrying out its duties seem to be very high. As such, less or even no pressure is placed on civilians. This is mostly reflected in discussions of people's (possible) reluctance to share information. For the most part, this is ascribed to issues related to the “complex machine's” malfunctioning at various levels. Yet, a handful of articles point at how this can be mitigated in contact tracing phone calls through interactional practice, and point out the importance of empathy. However, whilst privacy concerns in an era of increased (biomedical) surveillance are surely legitimate (Jones, 2015), one could equally argue from a governmental perspective that such media discourses critiquing this issue might in fact influence citizens and render them even less inclined to share their personal information with a contact tracer over the telephone, thereby undercutting the efficacy of contact tracing practices as a public health tool to fight a pandemic outbreak in Belgium.

5.2. Later developments (late 2021 – early 2022)

During the summer of 2021, contact tracing was hardly mentioned in the press. For the most part, this was because of limited news concerning the contact tracing endeavors at that time and incredibly low infection rates, most probably because 70% of the population had just been vaccinated. This changed by the end of November 2021, when allegations of fraudulent practices by one of the call centers in the consortium came to light. As opposed to the other issues mentioned across newspapers, this is the only incident which was covered by about every Flemish newspaper within a three-day timespan. Collaboration with this call center was terminated quite quickly (see e.g., DS 28 November 2021, DM 29 November 2021, DT 30 November 2021). In March 2022, one article reported that the official investigation into the matter ‘identified that the call centers submitted nearly 25 million euro in “deviating” billing' but “due to the way contact tracing is structured, it remains unclear whether these are in effect fraudulent cases” (DS 26 March 2022, p. 8 – translated).

This one call center being suspended ties in directly with a different crisis unfolding at the time. In the final months of 2021, the infection rate among Flemish citizens was extremely high, causing contact tracing call centers to be under severe amounts of pressure. The removal of one call center from the consortium operations added oil onto the fire of “crisis in the management of crisis.” This crisis was partly ascribed to the suspension of specific COVID-19 safety measures. Citizens were allowed to see a larger number of people (while remaining cautious) and restrictive limits on restaurants' and cafés' opening times were also lifted. On the one hand, the newspapers reported how it had become impossible for people to keep a record of their contacts (DS 23 November 2021). On the other hand, a government voice included in this same article stressed the underreporting of contacts during the calls: “The people who contact tracing is eventually able to reach, provide only 2,7 contacts on average.,” but “[w]e all know that everyone has more than three contacts in the current social context” (DS 23 November 2021, p. 8 – translated).

The overburdened state of the call centers prompted the consortium tasked with contact tracing in Flanders to take drastic measures. At a certain point, a representative of the Flemish Agency for Health and Care is cited in De Standaard, stating that “those who have been infected are the priority now” (DS 21 November 2021, online – translated). Concretely, this referred to the decision to restrict calls to index patients, and no longer their contacts. At one point, not every index patient was even called but received the following text message: “We are unable to reach you via telephone, but self-isolate for at least 10 days if you tested positive for the Coronavirus” (DS 21 November 2021, online – translated). One spokesperson for the contact tracing consortium stated that “It is evident that we prefer to call people, so that we can provide them with additional information and address their questions. [..] But those who do not receive a call can still contact the contact tracing center with their questions” (DS 21 November 2021, online – translated). The focus here does not so much lie on the contact tracing itself, but more so on addressing people's concerns. The quote is followed by a statement that calling is preferred over texting because a text can be less compelling for people, who may not be as convinced of the need to get tested or to go into quarantine. This is the only excerpt in the corpus which implicitly reflects some form of a caring stance as a defining part of the contact tracing conversation in this period after November 2021. Again, it is mentioned by an Agency of Health and Care representative and with the goal to promote or mitigate.

Our thematic analysis of the quality press articles published more recently indicate more event-driven rather than aspect-driven accounts, i.e., less about characterizing aspects of the practice of contact tracing and more in response to developing newsworthy events related to contact tracing. The specific events in this period which received attention have been visualized in Figure 3.

FIGURE 3

Figure 3. Overview of themes in newspaper articles – More recent developments.

Interestingly, in 2020 and early 2021, there were a few mentions of interactional practice or empathy in the newspapers, of which most were promotional in nature (cf. supra). Compared to this, in late 2021 there were distinctly fewer mentions of this dimension of contact tracing. Yet, the latter period was arguably the most crucial period in which to do so. Because of all the negative events reported on, the public image of contact tracing was severely damaged. Highlighting the interactional and empathetical quality of contact tracing calls could have been a great form of damage control and securing civilians' compliance in both the contact tracing system and the government's risk management strategy at large.

From our analysis of the newspaper article dataset, it becomes clear how the press in the studied period mostly tended to portray telephone contact tracing as a “system” designed to collect information about people's contacts and their whereabouts during their infectious period. Newspapers overwhelmingly reported on various types of matters related to the almost “mechanical” workings of this system: issues regarding privacy, financial concerns, but also practical issues and software issues. In doing so, newspapers even explicitly link this to contact tracing's efficacy by regularly stating that people may be reluctant to collaborate or share information as a direct result of such (persisting) issues. Even though contact tracing representatives explicitly mention that one purpose of contact tracing is to address people's concerns or to provide information alongside requesting it, the media image of telephone contact tracing under the period of scrutiny is mostly centered on the system itself and its perceived “main purpose” of gathering information from infected people and instructing people to go into quarantine or self-isolate. Even though a small number of articles mention the importance of interactional affordances and empathy in relation to these issues, especially after May 2021, the interactional dynamics of contact tracing practice itself, including the conversational scope and dynamics of the phone calls, remained fairly absent from the newspapers' pages.

6. Conclusion

This paper started by introducing strategies and contact tracing endeavors with the purpose of containing the COVID-19 pandemic in Flanders, Belgium. With its specific focus on telephone contact tracing, this paper compared its prevailing media image with the characteristics and functional orientations of on-the-ground interactional practice.

Our analysis of the contact tracing interactions uncovers the complexities and interdependency of different but interrelated functions. The purpose of the contact tracing telephone conversations clearly not only includes the gathering of information and provision of instructions. Contact tracers are simultaneously transversally tasked with maintaining an empathetic and caring stance, while managing their role as a representative of the government's public health policy. In addition to the difficult endeavor of balancing these four tasks, contact tracers are also expected to communicate efficiently. Our outline of this broader analysis in this paper specifically concentrated on the transversal care-imperative. In Section 4, we presented a particularly apt example of this transversal care-imperative which is representative of many interactional sequences in the corpus. For a more complete overview of the complex interrelations and interactions of these different functions of the contact tracing call, see De Timmerman et al. (2022) and Slembrouck et al. (2021).

On the basis of our comparison of this interactional reality to the media-generated image of COVID-19 telephone contact tracing in Flanders, we conclude that the media, on the other hand, have mostly tended to portray contact tracing quite negatively, or with the purpose of informing the public of its existence, its flaws, and – in a limited number of cases – its interactional affordances. This is manifested in an abundance of (sometimes quite sensationally presented) reports on the faults or malfunctions of the contact tracing system as well as the mostly narrow portrayal of the purpose of contact tracing as limited simply to information exchange: that is, gathering personal information on people's contacts and whereabouts. Here, we see Ogbogu and Hardcastle's (2021) findings reflected, as one can argue that the media representations found in our dataset of newspaper articles can be considered descriptive and fairly uncritical in the sense that representations do not accurately portray the realities of contact tracing talk. Moreover, we also see reflected in our data O'Connor et al.'s (2021, p. 19) observation about “feeding public alienation by purveying deficit model assumptions,” as many of the articles in our dataset focused on practical, economic, and privacy-related “deficits” of Flemish contact tracing endeavors, without (sufficiently) addressing the interactional affordances of actual contact tracing practice, including its dimensions of patient and citizen support. The latter could have greatly benefited contact tracing's efficacy by nurturing public trust in organized contact tracing practice. The only exceptions in our media corpus include the explicit promotion of the caring stance by stakeholders, which occurred quasi exclusively in the early stages of telephone contact tracing practice.

In an interesting way, a lot of the prescriptive “how to”-literature related to contact tracing in contexts of HIV and TB stresses the pitfalls of stigmatization and loss of privacy. Respect for privacy is vital to secure reliable information that can be used efficiently to contain virus spread. However, as it turns out, telephone contact tracing is equally about addressing uncertainty, attending to questions which an IP may have in terms of where things are going – e.g., vaccination, providing affirmation for emotional concerns, or even giving medical advice (despite the fact that CT's are not professionally qualified to do so), etc. The media coverage arguably took communication for granted and was principally interested in the institutional task of “contact tracing” as narrowly understood, not really its accomplishment through conversational practice and interaction.

In sum, we can conclude that, above all, the media communicated a rather distorted image of telephone contact tracing to the public. This can be considered problematic, as most citizens had no knowledge of or experience with this complex type of conversation prior to the COVID-19 pandemic and had only media coverage to rely on. Even though the role of the media in a functioning democracy is arguably to hold governments accountable for their policies and actions, we do not see this manifested sufficiently in our data for the dimension that the contact tracing telephone conversation as an institutional act was expected to be a care-centered conversation as much as an information-exchange type of interaction. More research is needed into the specific consequences and effects of media construal on the relative success of the contact tracing strategy, but from our analysis it is possible to conclude that the prevailing media image may have influenced citizens' responses to contact tracing and thus affected the efficacy of contact tracing via telephone in Flanders during the COVID-19 pandemic.

Data availability statement

The datasets presented in this article are not readily available because requests to access the datasets should be directed to Stef.Slembrouck@UGent.be.

Ethics statement

The studies involving human participants were reviewed and approved by the Ethics Committee of the Faculty of Arts and Philosophy of Ghent University and the Ethics Committee for the Social Sciences and Humanities of the University of Antwerp. All data was gathered with informed consent for participation.

Author contributions

All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

Funding

The research reported on in this paper was made possible with FWO-funding. Grant number G0G6120N with SS, MV, HB, RC, TG, and DW. The researchers hired by the project were A-SB, EC, RDT, AD, and SVDG.

Acknowledgments

The authors of this paper wish to thank their research partners and the funding agency FWO Vlaanderen, as well as the organizers of the Language of COVID-19 conference (2021). The authors also wish to thank our co-PI's and researchers: Hilde Bastiaens, Robert Colebunders, Tina Goethals, Elien Colman, Alfred Dusabimana, and Dirk Wildemeersch..

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

1. ^For a detailed analysis and discussion of each function, see De Timmerman et al. (2022).

References

Antaki, C. (2011). Applied Conversation Analysis: Intervention and Change in Institutional Talk. London: Springer.

Google Scholar

Begun, M., Newall, A. T., Marks, G. B., and Wood, J. G. (2013). Contact tracing of tuberculosis: a systematic review of transmission modelling studies. PLoS ONE 8, e72470. doi: 10.1371/journal.pone.0072470

PubMed Abstract | CrossRef Full Text | Google Scholar

Bowen, G, A. (2009). Document analysis as a qualitative research method. Qual. Res. J. 9, 27–40. doi: 10.3316/QRJ0902027

PubMed Abstract | CrossRef Full Text | Google Scholar

Candlin, C. N., and Candlin, S. (2003). Healthcare communication: a problematic site for applied linguistics research. Annu. Rev. Appl. Linguist. 23, 134–154. doi: 10.1017/S0267190503000230

CrossRef Full Text | Google Scholar

Corbin, J., and Strauss, A. (2008). Basics of Qualitative Research: Techniques and procedures for Devel-Oping Grounded Theory, 3rd Edn. Thousand Oaks, CA: Sage.

PubMed Abstract | Google Scholar

De Timmerman, R., Bafort, A. S., Van de Geuchte, S., Vandenbroucke, M., and Slembrouck, S. (2022). “Formulations of risk and responsibility in COVID-19 contact tracing in Flanders”, in Discourse Perspectives on Risk and Responsibility Belgium, eds A. Ädel, C. Nyström Höög, J. Östman (Amsterdam: John Benjamins).

Google Scholar

Fereday, J., and Muir-Cochrane, E. (2006). Demonstrating rigor using thematic analysis: A hybrid approach of inductive and deductive coding and theme development. Int. J. Qual. Method. 5, 80–92. doi: 10.1177/160940690600500107

CrossRef Full Text | Google Scholar

Filardo-Llamas, L., and Perales-García, C. (2022). Widening the North/South divide? representations of the role of the EU during the COVID-19 crisis in Spanish media: a case study. J. Lang. Polit. 21, 233–254. doi: 10.1075/jlp.21065.fil

CrossRef Full Text | Google Scholar

Fortuin, M., Uydebrouck, M., Wanlin, M., Vermeire, P., and Schandevyl, W. (1998). Tuberculosis incidence and surveillance in Belgium. Arch. Public Health. 56, 199–208.

Google Scholar

Green, T., Talbot, M. D., and Morton, R. S. (2001). The control of syphilis, a contemporary problem: a historical perspective. Sex. Transmit. Infect. 77, 214–217. doi: 10.1136/sti.77.3.214

PubMed Abstract | CrossRef Full Text | Google Scholar

Hutchby, I., and Wooffitt, R. (2008). Conversation Analysis. Cambridge: Polity Press.

Google Scholar

Hyman, J. M., Li, J., and Stanley, E. A. (2003). Modeling the impact of random screening and contact tracing in reducing the spread of HIV. Math. Bio. 181, 17–54. doi: 10.1016/S0025-5564(02)00128-1

PubMed Abstract | CrossRef Full Text | Google Scholar

Jacob, S., and Lawarée, J. (2021). The adoption of contact tracing applications of COVID-19 by European governments. Policy Design Pract. 4, 44–58. doi: 10.1080/25741292.2020.1850404

CrossRef Full Text | Google Scholar

Jones, R. H. (2015). “Surveillance,” in The Routledge Handbook of Language and Digital Communication, eds A. Georgakopoulou and T. Spilioti (Oxfordshire: Routledge), 422–425.

Google Scholar

Lakoff, G. (2006). Simple framing. Rockridge Institute 14.

Google Scholar

Mroz, G., Papoutsi, C., Rushforth, A., and Greenhalgh, T. (2021). Changing media depictions of remote consulting in COVID-19: analysis of UK newspapers. Brit. J. Gene. Pract. 71, e1–9. doi: 10.3399/BJGP.2020.0967

PubMed Abstract | CrossRef Full Text | Google Scholar

O'Connor, C., O'Connell, N., Burke, E., Nolan, A., Dempster, M., Graham, C. D., et al. (2021). Media representations of science during the first wave of the COVID-19 pandemic: a qualitative analysis of news and social media on the island of Ireland. Int. J. Environ. Res. Public Health 18, 9542. doi: 10.3390/ijerph18189542

PubMed Abstract | CrossRef Full Text | Google Scholar

Ogbogu, U., and Hardcastle, L. (2021). Media representations of COVID-19 public health policies: assessing the portrayal of essential health services in Canadian print media. BMC Public Health 21, 1–6. doi: 10.1186/s12889-021-10300-2

PubMed Abstract | CrossRef Full Text | Google Scholar

Rampton, B. (2019). “Interactional sociolinguistics,” in The Routledge Handbook of Linguistic Ethnography, ed B. Rampton (Routledge), 13–27.

Google Scholar

Schegloff, E. (2007). Sequence Organization in Interaction. A Primer in Conversation Analysis. Cambridge: CUP.

Google Scholar

Slembrouck, S., and Hall, C. (2011). “Family support and home visiting: understanding communication, “good practice” and interactional skills,” in Handbook of Communication in Organisations and Professions (Berlin: Mouton de Gruyter), 481–498.

Google Scholar

Slembrouck, S., and Hall, C. (2019). Advice giving, managing interruptions and the construction of ‘teachable moments'. Appl. Linguist. 40, 1–21. doi: 10.1093/applin/amx004

CrossRef Full Text | Google Scholar

Slembrouck, S., Vandenbroucke, M., De Timmernam, R., Bafort, A. -S., and Van de Geuchte, S. (2021). “Support and surveillance during contact tracing phone calls in the COVID-19 pandemic: a “new” genre characterized by multiple pressures,” in Paper Presented at the Annual DANASWAC Seminar, 17 August 2021.

Google Scholar

Tomnay, J. E., Pitts, M. K., and Fairley, C. K. (2005). New technology and partner notification-why aren't we using them? Int. J. STD AIDS 16, 19–22. doi: 10.1258/0956462052932700

PubMed Abstract | CrossRef Full Text | Google Scholar