Diagnostic uncertainties, ethical tensions, and accounts of role responsibilities in genetic counseling communication

3.1 Participants

Ten pre-test consecutive genetic counseling sessions concerning one extended family were audio-recorded within a hospital department of medical genetics in Norway.³ The clients were from a family with a high burden of colorectal cancer; they had previously been tested for known cancer genes, with negative results. Ten different family members attended the pre-test clinic sessions. Two counselors were involved across the clinics, and both had undergone a 2-year university Masters level program for genetic counselors, which incorporated essential aspects of genetics as well as communication training. Although both counselors were present in a given clinic, one of them took on a passive observer role.

3.2 Recruitment procedure and data collection

Exome sequencing was offered to explore and explain the cancer predisposition in this extended family. The study design was approved by the Regional Committee for Medical Research Ethics and the participants were recruited by means of a personal letter explaining the aim and scope of the study. Participation was based on written informed consent, emphasizing its voluntary nature, that is, their right to leave the study whenever they chose to, without having to explain their decision.

The data were collected over a period of fourteen months, between 09/13 and 11/14. The audio recordings in Norwegian were first transcribed verbatim and then translated in to English by the first author and later verified by a competent bilingual speaker.⁴ The names and places have been anonymized to protect confidentiality. The following simplified transcription conventions were used.

3.3 Analytical framework

We broadly adopt theme-oriented discourse analysis (ToDA, Roberts & Sarangi, 2005), which attests the jointly accomplished nature of interaction through participation structure (Goffman, 1981). The joint nature of the interaction is explored through: Mapping the structural, interactional and thematic trajectories; identifying key interactional events within the context of interest; analyzing interaction within and across phases in terms of focal themes (e.g., decision-making) and analytical themes (linguistic/rhetorical devices); and connecting the findings to outcomes.

A first step in our analytical procedure is to identify ‘focal themes’, similar to the grounded theory approach (Glaser & Strauss, 1967; see also Braun & Clarke, 2006). The next step involves a selection of ‘analytical themes’ or rhetorical devices that are utilized in the articulation of the focal themes. In ToDA, it is important to map the analytical themes onto the focal themes and assess their rhetorical import. Through iterative readings of the transcripts and thematic mapping (Sarangi, 2010b) involving the first two authors, the following three inter-related focal themes emerged: diagnostic uncertainties, risks, and benefits of knowing and role relationships. This pre-analytic process constitutes a form of coding, not with an aim to quantify the occurrences of the focal themes but to reach consistency in terms of theoretical salience, with the possibility of identifying their nuanced discursive manifestations in the data. As will be evident in our analysis, a suite of rhetorical devices is deployed by the genetic counselors and the clients in their accounts (Sarangi & Clarke, 2002; Scott & Lyman, 1968), for example, contrast, extreme case formulation, reported speech, hypothetical construction and pronominal reference.

Table 1 below provides illustrative examples of the three focal themes and their interrelationship, supplemented by corresponding analytical themes.

The examples for in-depth discourse analysis are taken from the 10 pre-test sessions, where recruitment of clients to ongoing research studies features in the discussion. As we will see, clients’ desire to know includes their right to access available genetic information. The topic of uncertainty comes to the fore as the research studies are aimed at reducing prevailing diagnostic uncertainties.

4.1 Tensions surrounding diagnostic uncertainties

Our first case in Extracts 1a and 1b concerns a female client (CF), in her late 50s, who is already affected with breast cancer and has had several polyps removed. CF has expressed her intention to remain informed about her progressive condition. She is also keen for her son (aged 29)—who has a serious illness that is not cancer related—to be considered for surveillance for potential colon cancer. CF’s brother has had prostate cancer. The genetic counselor (GC) describes the nuances surrounding secondary findings as a by-product of exome sequencing in the ongoing research study in which CF participates. She raises the issue of the client's right to have access to the potentially beneficial health information from the test result, while alerting her to the risk of knowing ‘that kind of information’.

At the outset in Extract 1a , ‘incidental findings’—glossed as ‘things you were not looking for’—are implicitly contrasted with ‘expected findings’ through a test procedure (turn 01). The assumption is that while ‘incidental findings’ can generate uncertainty, ‘expected findings’ would serve as a step toward diagnostic certainty. GC then offers an expanded explanation of what actions might follow the secondary findings. The potential problem of knowing is underlined and a contrast is made between test results being uncertain and how uncertain test results may be received by different clients (cf. McIntosh, 1978). The current circumstances of CF are contrasted with those who ‘may be a bit prepared’, thus implying that she might get information she is not prepared for. This contrast serves as a background for GC’s question (turn 01) where she seeks CF’s opinion of whether she wants access to the information surrounding secondary findings. The question serves a rhetorical function where the topic of wanting to know ‘that kind of information’ is linked to a risk of knowing scenario.

CF responds by saying that she would exercise her right to access all test-related information including diagnostic uncertainties (‘what is there to be known’). She upgrades her preparedness for uncertainty cumulatively: ‘everything that you know is [absolutely] fine’; ‘I’ve reached this age where every year is a plus’; and ‘nothing can be seen as negative’; ‘that's absolutely not a problem with me’. This justification implies that CF can cope with all information, even if nothing is conclusive. In the latter part of turn 08, she signals the ‘benefits of knowing’ (‘you can take precautions’) and reasserts her stance to access all available information through an extreme case formulation (‘that's absolutely not a problem with me’). In essence, she expresses her resilience and readiness to deal with uncertainty. CF also alludes to consensual familial communication (‘we have thought of that’), signaling that she is not simply exercising her individual desire to know but also speaks for the family.

The interaction continues as follows in Extract 1b, with GC questioning CF’s desire for accessing information in the context of diagnostic uncertainties arising from genomic testing.

In turn 13, GC latches on to the ‘benefits of knowing’ as she lists hypothetical scenarios where preventive measures can be taken (e.g., avoidance of certain medication; implantation of pacemaker and ICD; modification to alarm sounds; refraining from competitive sports). Following this long explanation/information sequence, in turn 15, GC returns to the question (as posed in turn 01) of whether CF wants to receive health information, but this time she draws attention to a possible tension between CF’s desire for knowledge and the future scenarios to cope with such knowledge. By pursuing the ‘risks of knowing’, GC underlines the need to clarify secondary findings in the context of family testing (Hallowell et al., 2015). As earlier, CF reasserts her wanting to know what can be known. Note that she uses the inclusive ‘we’ to characterize the family as uncertainty averse and formulates an extreme case scenario (‘the worst thing for us’, turn 16). In turn 18, CF offers another characterization via extreme case formulation—‘no one wants negative messages’. Turn 20 echoes CF’s wanting to know stance as in turn 16, while acknowledging the delicacy of the question at hand.

Let us consider the last part of this extended extract (Extract 1c) where GC returns to CF’s stance and seeks additional risk information.

In turn 21, GC continues with her query to ascertain what makes CF so resilient with regard to accessing (potentially) uncertain health information and if this is due to her experience of living with cancer, which is implicitly contrasted with the healthy population's lack of illness experience. The underlying question is: Would healthy individuals and affected individuals respond to diagnostic uncertainties differently? CF’s response in turn 22 (‘hard to say’) is somewhat hesitant, but when pursued further by GC in turn 23, CF characterizes herself as ‘a bit lucky’. She underscores how wanting to know or not know may be premised on different motives and then makes a contrast between her positive attitude to life (turn 24) and the healthy population's potentially ambivalent attitude to life. In turn 30, CF again returns to generalization with the use of ‘we’, this time as a marker of differentiation between herself and her brother—‘we tackle these things in different ways’. The fact that the brother is not interested in knowing his genetic risk could be construed as implicit blame. This works to frame CF as the kind of person who wants to take responsibility for knowing additional risk information. For exemplification, CF uses contrast to juxtapose her way of managing uncertainty (‘the way I am as a person’, turn 28) to that of her brother's (‘who sweeps everything under the rug’, turn 38), thus alluding to possible familial tensions surrounding genomic testing. Some of the tensions encompassing diagnostic uncertainties can be ethical in nature, to which we turn next.

4.2 Ethical tensions surrounding diagnostic uncertainties

In Extract 2a, the male client (CM), in his mid-forties, has been part of a surveillance system where he is going through colonoscopy every second year. He has had several polyps removed in the process. The extract opens with GC explaining that the exome sequencing test can generate secondary and uncertain findings, raising ethical issues about what information clients can be given access to. Prior to the opening of the extract, GC and CM have already touched upon the topic of recruitment for the ongoing research study.

In turn 01, GC provides institutional guidelines surrounding what information can be made known, and through a contrast she demonstrates what may count as ‘benefit of knowing’ versus. ‘risk of knowing’ (Sarangi et al., 2003) with regard to secondary findings. GC justifies a collectively shared ‘ethical perspective’ (‘if we find anything’), although it is somewhat unclear what the referent of ‘we’ is here. Does it collectively represent the counselors’ and the researchers’ points of view or only of the former? GC’s account comes across as a tendency to safeguard the institutional guidelines and their gatekeeping function with regard to available information (‘we are not going to give you an answer’). GC enhances her ethical stance through contrast: knowing in order to take ‘preventable measures available’ and avoid ‘serious risk to life’ versus. knowing ‘anything that is serious that has no treatment’. The issue of clients’ wanting to know becomes a part of her role responsibility, where GC orients to both scientific others (‘we find the cause’) and familial others (‘your family’) before shifting to clinical others vis-à-vis the ethical tensions outside of the research project (‘disease without treatment’).

The ethical tensions regarding how the intervention regime would unfold and who would have the right to what can be known if there are any secondary findings are placed in the family context. Due to the hybrid nature of the counseling session, GC is ethically torn between being client-centered in order to reduce diagnostic uncertainty (i.e., arranging ‘separate counseling session’) while, simultaneously, attempting to recruit clients for research to accomplish scientific (epistemic) certainty (i.e., encouraging ‘participation in research project’). This tension between the clinical setting and the research setting confirms the relevance of how differential ethical principles should be operative in these two contexts (Hallowell et al., 2015).

In turn 05, GC reaffirms her ethical position concerning non-disclosure of diagnostic findings for which ‘there is no cure’. ‘Risk of knowing’ serves as a justification for non-disclosure of ‘irrelevant’ test results. The tension between CM’s desire to know and GC’s ethical concerns is articulated through a hypothetically framed question pointing to the contrasting stance: ‘if I say that I do want to have it can't I get it then?’ (turns 8, 10). GC justifies her stance on non-accessibility through the use of reported speech (‘the ethics committee's perspective’), which indexes an ethical dilemma in that certain test results can be irrelevant and cause anxiety, which may border on ‘risk of knowing’. However, CM manifestly disagrees with the ethics committee's non-disclosure policy through extreme case formulation (‘they are [absolutely] wrong’, turns 14, 16) and affirms his preference regarding disclosure of any test result. In turn 15, GC aligns with CM and his right to access genetic knowledge while she also feels compelled to uphold the ethics committee's injunctions about non-disclosure, which counts as an ethical dilemma.

The interaction continues as follows in Extract 2b.

In turn 19, GC alludes to other clinicians’ and the ethics committee's collective stance on this delicate matter. Through a reasoned demand (‘I want to know it anyway’), CM justifies his stance again through contrast: ‘A genetically predisposed […] illness’ is implicitly contrasted with an accidental ‘bad cancer gene’. CM’s account signals that disclosure about inheritance is in the welfare of him as a client and this serves as a justification for demanding access to health information (turn 22). CM upgrades his desire for diagnostic certainty while underlining his ability to cope with epistemic uncertainty linked with hereditary conditions that have no treatment (turns 24, 26). As in extract 2a, in turn 28, CM disagrees with the official protocol about non-disclosure and voices his strong critique, with a rhetorical question—‘why shouldn't I want to know?’—followed by a temporal contrast: ‘I can live my life now [before I get it]’. CM justifies his right to access information which he can act on and thus demonstrates that he has the capability to assess and cope with uncertain knowledge. As a final gesture, in turn 30, CM asserts his autonomous stance toward both knowing and what to do with his knowing.

In what follows, we consider how clients and counselors orient to their role responsibilities as a way of foregrounding the ethical tensions surrounding diagnostic uncertainties, while justifying their respective role responsibilities regarding what can be known.

4.3 Clients’ orientation to familial others vis-à-vis diagnostic uncertainties

Extract 3 is from the same encounter as extracts 2a & 2b, and here we focus on the client's (CM’s) orientation to self-other relations when exercising his desire to know in the event of diagnostic uncertainties. It is worth reiterating that the client and his extended family had previously been tested for known cancer genes, with negative results. Exome sequencing was offered to further explore and explain the cancer predisposition in this family. Previously, GC and CM have talked about the rationale behind exome sequencing, as illustrated in prior extracts, and about researching for certainty to reduce uncertainty in this specific family. GC initiates the topic of disclosure of test results, and she articulates diagnostic uncertainty as part of the family history (‘a family where we don't know the details’). In framing such disclosure as a sensitive issue, GC simultaneously dwells on the specific value of targeting the research study on one particular family.

The extract opens with GC’s account of ‘risk of knowing’ (‘might become a burden to you’), followed by a contrast: ‘you do not want to know’ versus. ‘she [sister] wants to know’. Wanting to know available test results is foregrounded as a sensitive—and even divisive—issue in the family sphere. GC formulates a potential ethical dilemma regarding familial differences by sketching a hypothetical future scenario. In other words, GC raises the ethical concerns about CM’s coping strategies when dealing with certainty/uncertainty (turns 05–07) through use of reported speech (‘ethics committee thought it might be a burden to know’, turn 05). Our analytical interest here is how CM responds to such an ethical framing. As before (see extract 1c in particular), the rhetorical device of contrast becomes a resource to orient to familial others’ attitude to uncertainty in test results. In turn 08, CM disregards the ethical concerns and justifies his stance toward wanting to know (‘I am strong enough’; ‘I am curious enough by nature too’). The sister is implicated as he exercises his right to access health information (‘I know my sister so well that for us I don't think there's a problem’). Unlike in extract 1c, the brother and the sister are presented as a unity, rather than being differentiated. CM upgrades his certainty about wanting to know as opposed to the diagnostic uncertainty announced by GC through extreme case formulations (‘I will handle it anyway’; ‘I want to know everything […] ‘absolutely sure’), thus characterizing himself as someone who can cope with the ‘risk of knowing’.

As can be seen in turn 12, CM orients to both his own and his family's future wellbeing in an altruistic manner (‘looking at my family […] strengthens even more […] it is cool to contribute to research that other may benefit from’). The rationale underpinning his decision to want to know does not fully acknowledge the embedded dilemma as indicated by GC (turn 12). In turn 14, CM simultaneously characterizes himself (‘I am egoistic’) and points to a hypothetical future event which is preventable through knowing (‘maybe I don't die of cancer early’). This constitutes an essential part of CM’s justification for wanting to know because of the benefits of knowing as well as the timing of knowing (‘if anything could be picked up at an earlier stage that could maybe get something started’), while orienting to both self and familial others.

4.4 Counselors’ orientation to scientific and clinical others vis-à-vis diagnostic uncertainties

Our final case in Extract 4 concerns a male client (CM) in his early 30s, who is the youngest in this extended family. He has just started with colonoscopy and has had some polyps removed. His father is seriously ill with colon cancer. Previously, the genetic counselor (GC) has mentioned the amount of uncertain information the exome sequencing test may generate. However, in this case, as opposed to extract 3 where we drew attention to the client's self-other-orientations, we look into GC’s other-orientations to manage the tensions surrounding diagnostic uncertainties. In the encounter, GC repeatedly mentions CM’s right to access genetic information in light of this family dealing with diagnostic uncertainty and their right to get answers and explanations. It is worth noting that CM is a participant in the research study, and that GC has an obligation to disclose to him any findings resulting from the research study. The question of access to information is foregrounded in the interaction.

GC begins by issuing a statement of what is there to be known in the context of exome sequencing based on an implicit contrast between what is ‘of significance healthwise’ and the findings that are not considered significant (turn 01, 03, 05). GC has a twofold other-orientation: first she affiliates with the scientific others who are alluded to through ‘this study’; and then with both clinical others and scientific others (‘feedback[…] of significance healthwise’). The label ‘of significance healthwise’ indexes the client's right to be given certain information about secondary findings that would possibly affect his health and thus serves as a justification for CM’s future right to access information that falls within this label. She orients to scientific others (‘we have found’) in the generation of new genetic knowledge (turn 08). Rather paradoxically, CM is armed with the right of access to all available information but is simultaneously disarmed when urged not to exercise this right. That is, CM is framed as being autonomous, with the disclaimer that he must act responsibly.

In turn 10, GC orients to clinical others when implying an inclusive preferred action (‘we rather see you don't’) in light of lots of uncertain—even irrelevant—information. In turn 18, she inquires about CM’s stance on this matter. As can be seen, in turn 19, CM claims, contrastively, that he is not after information per se but in search of answers. CM further elaborates his position (‘as I said’) and offers a justification via orientation to general others through the pronominal choice ‘one’ (‘that one get answers’). Through this orientation his role responsibility aligns with clinical others and other-others, all of whom are implicated in the management of uncertainty. In adding ‘it is no more than that’, he acknowledges the complexity of the issue (turn 21).

In turn 22, GC once again reaffirms CM’s moral and legal ‘indisputable right’ but uses self-reported speech to foreground the role relationships between them as clinicians and the client (‘we don't sit here saying you should not ask for it’). CM responds by acknowledging the clinicians’ dilemma through his limited understandings as a client (turn 29). GC then responds with a hypothetical construction: whether CM as a research subject would act on his right to access information (‘if I should ask for access’). This hypothetical scenario is, however, embedded in the specific context of exome sequencing producing a set of genetic findings (‘it is going to be a lot’) that needs to be sorted out (‘will take a long time to process’). CM aligns with GC’s stance on exome sequencing as producing a lot of information framed as ‘complicated arrangement’, which alludes to benefits of not knowing. By upholding this complex future event, GC justifies against disclosure or ‘risks of knowing’, as she explicitly comments on the hybrid nature of the encounter as a research study and a clinical encounter (see also extract 2b). This is in line with Hallowell et al.’s (2015) injunction about the need for different ethical guidelines regulating study findings vis-à-vis disclosure of such findings in research versus clinic settings.

6.1 Study limitations and practice implications

Our study has several limitations, typical of any qualitative study, in that the findings are not generalizable across genetic conditions and across clinical settings and families. While not generalizable, the findings are potentially transferrable to other research settings addressing the interface of new technologies of genomic testing and the communicative challenges facing genetic counselors in the clinic setting. In genetic counseling, dealing with uncertainty is not merely a matter of providing warrants for what is certain and known, but also a matter of managing diagnostic and epistemic uncertainties and the attendant ethical tensions through accounts of role responsibilities.

Conflict of interest

Gøril Thomassen Hammerstad, Srikant Sarangi, and Inga Bjørnevoll declare that they have no conflict of interest.

Human studies and informed consent

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000 (5). Informed consent was obtained from all patients for being included in the study.

Animal studies

No animal studies were carried out by the authors for this article.