Views of Israeli healthcare professionals regarding communication of genetic variants of uncertain significance to patients

Abstract While genomic medicine is becoming an important part of patient care with an ever‐increasing diagnostic yield, communicating variants of uncertain clinical significance (VUSs) remains a major challenge. We draw on qualitative analysis of semi‐structured interviews conducted in 2020 with 20 Israeli healthcare professionals and stakeholders involved in communicating the results of genome‐wide sequencing to patients. Respondents described four main strategies of communicating VUSs to patients: preparing the patient pre‐test for uncertainty; adapting the level of detail to the patient's needs; upgrading versus downgrading the VUS; and following up on the possible reclassification of VUSs. These strategies were expressed differently by physicians and genetic counselors, varying according to their specialty and perception of the patient's situation. We discuss the strategic management and communication of uncertain genomic test results with patients in the context of meeting patients' expectations and working toward genetic causality through genomic narration and designation.

challenges across medical settings and indications. In the context of prenatal genetic diagnosis, some prospective parents have reported feeling anxious after receiving a VUS result and considered the uncertain and unquantifiable risks 'toxic knowledge' that caused lingering worries about their child's development (Bernhardt et al., 2013).
Another study found that following the initial shock, parents had no enduring concerns about VUSs (van der Steen et al., 2016). Cancer patients who were more worried about the course of their disease disregarded VUSs (Clift et al., 2020). Finally, patients with undiagnosed symptoms reacted positively to VUSs which were considered as a potential hint that may lead to future diagnosis (Kiedrowski et al., 2016;Skinner et al., 2017). HCPs' and patients' responses to uncertainty, therefore, are not simply a function of the abstract uncertainty itself; they are subjectively shaped by the contingent meanings given to that uncertainty within the specific contexts of a diagnostic odyssey (Werner-Lin et al., 2016. Genetic counselors (GCs) are often the main professional authority for patients to make sense of genetic uncertainty, and as a result, GCs often feel a need to manage uncertainty in a way that allows them to facilitate collaborative decision-making regarding treatment, follow-up, prevention, and testing other family members (Rauscher, 2017).
Communicating with patients about VUSs is paradoxical: Each VUS, by nature of its liminal, temporal qualities, invites communication about what is still uncommunicable due to inherent gaps of data and understanding. Therefore, the VUS requires narration (Timmermans et al., 2017) and designation (Navon, 2019). These terms, which do not have a strict definition, refer to communication efforts that attempt to create meaning from uncertain results. This includes an effort to fit the VUS into relevant genetic models, for example of expressed disease characteristics in biological family members, even though the associated phenotype is not distinct enough for diagnosis. The VUS may lead HCPs and their patients down any number of paths, including some dead ends. Communication and uncertainty are known to go hand in hand in health care (Mishel, 1990).
Yet, in the case of VUSs, uncertainty is at the core of communication itself. HCPs must negotiate the expectations of patients, including negotiating how communication of uncertainty affects patients' perceptions of the HCPs' credibility (Zhong et al., 2020). Furthermore, studying communication strategies of VUS disclosure has important implications for how genetic counselors can help facilitate meaningmaking for patients (Scherr et al., 2020). This study seeks to examine HCPs' communication methods and challenges regarding the disclosure of VUSs in clinical care, a topic which we still know relatively little of and, as a result of its recent introduction into practice, does not have clear professional guidelines. In a context of significant clinical uncertainty, for both experts and patients receiving VUS from panel testing, identifying communication processes is an essential step toward developing guidelines and supporting interventions for decisions that will become increasingly common as panel and exome testing spreads.
Uncertainty is usually constructed as negative in medical training (Atkinson, 1984;Han et al., 2017;Lupton, 2013;Newson et al., 2016). Nevertheless, team discussions of uncertainty, including that related to VUS, can be used productively as a catalyst for social action (Timmermans et al. (2017). VUS communication can keep patients under medical purview in an attempt to establish genetic causality, especially when facing patients who are searching for a cause of their symptoms, specifically tumor profiling oncology, cardiomyopathy, and undiagnosed-disease genetic testing patients (Clift et al., 2020).
To provide much needed empirical evidence about communication of VUS in the fields of cancer and cardio genetics (where panel and exome testing is being integrated into care thus entailing routine discovery of VUSs), this paper draws on interviews with Israeli healthcare professionals involved in genomic medicine. These were chosen as they are involved in precision genomic medicine either as clinicians referring to testing, interpreting genomic results, and following up patients. Identifying the strategies that HCPs and particularly GCs use to communicate uncertainty should help advance research on communication and counseling in precision genomic medicine.

| Study design
Semi-structured interviews were conducted as part of a broader study to investigate ethical and social issues of negotiating genomewide sequencing (GWS) in clinical care.

| Recruitment and sample
Following IRB approval, we contacted relevant (mostly oncogenetic) physicians and genetic counselors from 6 different medical centers across Israel. Participants were identified and invited to participate by the clinical author of this paper (SSR) based on her professional experience and networks. The first author (a female MA student in medical sociology with training in qualitative methodology and an interest in the research topic) conducted all the interviews. Interviews lasted between 30 and 60 min, and given the COVID-19 pandemic circumstances, were mainly conducted via telephone or online. No one else was present in the interviews besides the participants and the researcher. All those approached agreed to take part in the research. None of the researchers or the participants expressed a need to carry out repeat interviews. Where needed, the interviewer made field notes after the interview, elaborating for example on the interview data, setting, and circumstances.

| Procedures
Based on relevant literature and the clinical experience of the fourth co-author, the research team (composed of two male researchers and two female researchers) prepared the guide for the semi-structured interviews, including questions addressing the following topics: (a) how medical geneticists, genetic counselors, and clinicians cope with the challenge of uncertainty in genetic testing results? and (b) What strategies do HCPs use to communicate uncertainty regarding VUSs results to patients? Specific questions addressed experiences and challenges concerning communicating VUSs with patients (see Table 1). Additional questions concerning raw data and incidental findings which were part of the guide are not discussed here.

| Data analysis
The anonymized transcripts were analyzed thematically to uncover discursive themes and categories of themes recurring within and across groups of respondents, for example, groups of different HCPs (Charmaz & Belgrave, 2012). We followed an exploratory, qualitative methodology, which is aimed at advancing existing theory using unstructured, open-ended data, and is thus particularly appropriate for the current investigation. Following qualitative data analysis methods, our approach was to develop themes by constantly making comparisons and noting relationships among initially identified codes, inductively specifying and refining these codes, and then putting them in thematic categories or families. This iterative, reflective practice enables to build new theories or to extend existing theory on issues that have been largely understudied (Charmaz & Belgrave, 2012).
Following a review of the relevant literature, preliminary codes included communication strategies, consent, and responsibility to recontact ('following up'). The research team discussed the initial interviews and agreed that they pertained to these codes, which were then analyzed further inductively. The texts pertaining to each of these codes were extracted, and then in the next rounds of the analysis and discussions, more specific themes within the text pertaining to these preliminary codes were identified, as well as new themes that emerged from the analysis. For example, the preliminary, general code of 'communication strategies' was refined into several specific strategies, including preparing the patient for uncertainty, adapting the level of detail, down-or upgrading the VUS, and following up on the possible reclassification of VUSs. These more specific strategies of communicating VUSs were thereafter treated as themes, each with its relevant set of codes. Through such inductive analysis, 'communication strategies' which were treated to begin with as a preliminary code were replaced by more refined coding and came to be regarded as a broad thematic category. The research team did the coding together on the first few interview transcripts, discussing the relevance of the themes and agreeing on needed modifications and reclassifications. The first author completed the coding of the remaining transcripts. In this manner, we collected data and analyzed them simultaneously, starting with the initial interviews and their analysis and following with an additional round of interviews and their analysis. Following the analysis of the first round of the interviews, we made some mild changes to some of the questions, adjusting the phrasing for clarity and simplicity. We did not use a specific software for qualitative analysis. Participants were able to receive their interview transcripts for comment and/or correction, yet none of the participants requested this.
We discussed new findings as they appeared and their relationships to the codes in team meetings, where agreements were reached to prevent the potential bias of a single rater. The iterations stopped after the first ten interviews, when the authors agreed on all the themes and no new themes were identified, suggesting that theoretical saturation of the sample was achieved (Corbin & Strauss, 2008). Each of the themes is described below and illustrated with quotes from respondents, who are given pseudonyms. These quotes were translated by the authors from Hebrew to English.
Quotes were selected because they were noted by at least two of the authors as examples that best captured the identified themes.
We did not conduct participant checking since in studies that are not participatory or collaborative there is little evidence that member checks improve research findings (Thomas, 2017). We focus here on views presented concerning communication pathways regarding the initial presentation of VUSs. Due to space restrictions, we cannot elaborate here on communication regarding the follow-up of VUSs reclassification and recontact, a topic we focus on elsewhere.

| RE SULTS
The sample was comprised of healthcare professionals who communicate GWS results to patients: 8 genetic counselors (two of them with cardio-genetic expertise), 5 medical geneticists (3 specialized in cardio-genetic), 5 oncologists, and 2 legal experts with bioethical expertise concerning GWS in Israel (total n = 20). All participants had more than 5 years of experience in their specialty. All participants were white non-Hispanic and worked in a public medical center (except for the 2 bioethicists) (see Table 2 for experts' demographics).

TA B L E 1 Semi-structured interview guide
Five core themes were defined: The impact of VUSs on patient care; preparing the patient for uncertainty; adapting the level of detail; down-or upgrading the VUS; and following up. We mention the first theme in brief as it provides the broader context for communicating VUSs to patients and focus in more detail on the other four themes to highlight the variety of communication strategies used by HCPs.

| The impact of VUSs on patient care
While all our respondents agreed that VUSs presented a challenge

| Preparing the patient for uncertainty
Most HCPs explained that they prepare their patients in advance, before the genetic test is conducted, regarding the possibility of receiving uncertain results. All the GCs, as well as most of the clini-  saying. This is paternalistic, but I think it's not so bad to be paternalistic in this situation. Some people can understand VUS and for others it is difficult.

| Adapting the level of detail
Our respondents furthermore described two main strategies for communicating VUSs. One strategy stressed the neutrality of presenting the VUS. This approach was more common among clinicians outside of genetics (oncologists and cardiologists). However, it was also endorsed by a few of the GCs. The second pathway, more common among GCs as well as a few of the physicians, was 'to paint the VUS' (as they called it) by giving the unknown variant a known meaning, described in the following section.  Ali, an oncologist, described the influence of the severity of the patient's medical condition on the communicative pathway chosen regarding VUS in the following words:

| Down-or upgrading the VUS
Some patients see us while they're dealing with active cancer. They are less interested in uncertain genetic changes. If we tell them they are carriers of some VUS, they might want to tell their family members.
But at the end of the day, these are patients undergoing chemotherapy, in an acute condition, and they rightfully leave the VUS for later. How much can a person handle? So, often they would be indifferent, and we say 'okay this is something we will handle later on'.

| Following up
In While following up on reclassification that may lead to recontact was a thorny issue for all the respondents, they usually did not communicate their concerns to the patient. Nevertheless, it remained in the background as a source for potential errors and concerns of personal and systemic constraints among the HCPs.

| DISCUSS ION
In this study, we identified how HCPs in genomic medicine handle the uncertainty of genetic variants and examined their strategies for managing and communicating such uncertainty. Whereas variant classification in Israel follows the ACMG guidelines (Li et al., 2017;Richards et al., 2015), no guidelines are in place regarding the way VUS are communicated to patients and the long-term management of reclassification and recontact past patients with new evidence.
Focusing on HCPs' views regarding the daily practice of VUS communication and the perceived gaps between what should and can be done, the current investigation extends research and theory on uncertainty management beyond the perspectives of patients who undergo genetic testing as well as beyond the formal guidelines of doctor-patient communication in personalized genomic medicine.
Our findings extend and provide new insights into previous studies of VUSs communication. Zhong et al. (2020)  but rather combined them in various ways.
The communication pathways we termed as de-neutralizing the VUS through interpreting it as likely benign or likely pathogenic, that is, downgrading or upgrading the VUS, can be linked to the institutionalized expectation that it is the role of clinicians to solve diagnostic uncertainties (Brashers et al., 2006;Jutel, 2009). Such expectation can account for the leeway taken by HCPs in interpreting VUSs in a manner that extends beyond the lab report. Furthermore, VUSs reinterpretation by HCPs highlights how dynamic knowledge about genetic variants is used to carve out new medical conditions that are shaping people's lives. Navon (2011Navon ( , 2019 termed this way of moving from genotype via phenotype to people's lives as 'genomic designation'. This phenomenon had existed together with the development of the profession of genetic counseling, but has recently increased considerably, not just in volume but also in more fundamental ways. It is no longer the ideal type of the 'gene-for' or 'chromosomal abnormality for' that solely dominates genomic health communication. Instead, HCPs and patients doing precision genomic medicine are faced by a host of genetic variations whose clinical manifestations and disease classification require selective elucidation (Millo et al., 2021;Shkedi-Rafid et al., 2016). Each VUS is a potential starting point for a process whereby genetic variants are leveraged (Navon, 2011(Navon, , 2019. Previous influential sociological accounts of the genomic view of the self, such as Paul Rabinow's (1996) 'biosociality' and Nikolas Rose's (2007) 'molecular gaze', were based on and explored already-existing phenotypically based genetic conditions.
Our findings show how in specific cases of higher clinical suspicion VUSs can be constructed as embarking on a potential career of genetic causality.
In that manner, when HCPs tended to look at the VUS as a suspect when they initiated the genetic testing based on severe symptomatology and followed-up on it with segregation analyses, they were doing uncertainty narration (Timmermans et al., 2017). Such communication efforts that attempt to fit genetic models to data on expressed disease characteristics in biological family members are also illustrative of 'genomic designation' (Navon, 2019). In some cases, such narration or designation backfires, as in the story told by the GC about the VUS that was found in a patient who died of a heart attack at a young age, and for years served to guide surveillance in unaffected family members, until the clinical team realized that it might have been a mistake. Such a narrative demonstrates the customization of specific variants through a plot that begins with a mystery, develops through clinical discussions and research analyses, and hopefully ends with a resolution of the causal ambiguity (or not, in this case). It should remind us that providing the patient with an action they could take based on the VUS, which may be recommended for reducing ambiguity (Scherr et al., 2020), does not always correspond with a beneficial coping strategy.
Existing theoretical frameworks on uncertainty communication stress the goal of normalizing uncertainty through mechanisms such as cognitive reappraisal, information seeking, and acceptance (Brashers, 2001;Mishel, 1990). Rather than appraising uncertainty as a threat to their expertise, HCPs engaged in information seeking to make sense of it. In addition to substantiating work on uncertainty management, the current study's findings shed light on the HCPs' concrete strategies for communicating uncertainty. The HCP's strategy of adapting the level of detail to the patient corroborated other adaptation strategies for assessing the individual's informational preferences and capacity for understanding uncertainty (Bansback et al., 2016), including the use of subjective and qualitative descriptions that help to place medical risks in persons' life (Simpkin & Armstrong, 2019). The HCP's strategy of de-neutralizing the VUS illustrates seeing uncertainty as an opportunity instead of merely a lack of knowledge (Hillen et al., 2017

| Limitations
There are some limitations to this study. Although the participants' specialties contributed diversity to the sample, the sample size was

ACK N OWLED G M ENTS
We thank all the healthcare professionals who participated in the study. Jehannine Austin served as action editor for this manuscript.

Alma Levin Fridman, Aviad Raz, Stefan Timmermans, and Shiri
Shkedi-Rafid do not have any conflicts to report.

H U M A N S TU D I E S A N D I N FO R M E D CO N S E NT
This study has been performed in accordance with the Declaration of Helsinki and has been approved by the Ethics Committee of Hadassah Medical Center #0447-19-HMO. Informed consent to participate in the study has been obtained from participants.

A N I M A L S TU D I E S
No non-human animal studies were carried out by the authors for this article.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.