Topicalizing psychosocial distress in cancer follow-up consultations: An exploration of the interactional effects of discussion tools

: This paper reports on a case study into the effects on doctor–patient interaction of the use of the Distress Thermometer and Problem List (DT+PL) as a tool to further the discussion of psychosocial distress in the follow-up head-and-neck cancer consultation. The follow-up head-and-neck cancer consultation is a well-defined, goal-oriented, institutional speech event. Its aims include checking patients for recurrence and monitoring patients for after-effects of the malignity and its treatment, and psychosocial distress. The discussion of psychosocial issues, however, is not as well-integrated in the follow-up consultation as is deemed desirable. To remedy this, instruments have been developed to facilitate the discussion of patients’ quality of life. This paper uses the conceptual framework of the Ethnography of Communication and insights from Discourse Analysis and Pragmatics to analyse the impact of one of these instruments on doctor–patient interaction. Fine-grained analysis of the contextual parameters of the follow-up cancer consultation and of interactional data and meta-data suggests that, although the tool is successful in creating affordances for the discussion of psychosocial problems, it may simultaneously—and paradoxically—also put constraints on the range of topics that is discussed during the consultation. Doctors show awareness of this and deploy a variety of strategies to preserve their “normal”


PUBLIC INTEREST STATEMENT
Psychosocial distress is often neglected in followup cancer consultations. Undetected psychosocial distress may negatively impact cancer survivors' quality of life and recovery from treatment and illness. To help doctors and patients discuss psychosocial distress, various discussion tools have been developed, one of which is the Distress Thermometer and Problem List. The introduction of discussion tools in medical consultations creates affordances for the discussion of psychosocial distress, but may also, paradoxically, constrain the topics of discussion. The introduction of this discussion tool also has an impact on the interactional organization of the event, e.g., on the opening of the consultation and the introduction of topics for discussion. Physicians use different strategies to negotiate the context changes caused by the introduction of discussion tools, which have an effect on the duration of the consultation.

Introduction
After treatment for cancer, standard procedure in the Netherlands is for outpatients to have their condition monitored over a number of years, which involves them coming to hospital at regular intervals and meeting with their oncologist for follow-up cancer consultations. The aims of these consultations include (1) monitoring the patient for signs of recurrence, metastasis and second primary tumours, (2) reassuring patients/relieving anxiety and (3) monitoring the (management of) after-effects of the malignity and its treatment, including quality of life (see, e.g., National Cancer Institute, 2010). Since 25-50% of cancer patients experience so much distress during and after treatment and follow-up "that referral to a specialized psychosocial and or (para) medical caregiver is warranted" (Oncoline, 2010, p. 1), over the years the realization has grown-not just in the Netherlands, but internationally-that it is important to monitor cancer patients for psychosocial distress (see, e.g., Arora, 2003;Epstein & Street, 2007;Higginson & Carr, 2001;Oncoline, 2010). However, for various reasons the focus of discussion in the follow-up consultation seems to be on physical issues: both doctors (see, e.g., Beach et al., 2004;Ford et al., 1996;Zhou et al., 2015) and patients (see, e.g., Arora, 2003;Beach et al., 2004;Cameron et al., 2015;Detmar et al., 2000) seem to avoid discussing psychosocial distress, which often results in these problems going undetected and untreated. To remedy this situation, the Integraal Kankercentrum Nederland (IKNL), in collaboration with the Nederlandse Vereniging voor Psychosociale Oncologie (NVPO) and Koningin Wilhelmina Fondskankerbestrijding (KWF) recommend that the Distress Thermometer and Problem List (DT+PL; called "Lastmeter" in Dutch; see Tuinman et al., 2008)-a tool developed by the National Comprehensive Cancer Network (NCCN), USA, for patients to self-report their levels and areas of physical and psychosocial distress (Holland & Bultz, 2007;Tuinman et al., 2008)-be used as a discussion tool in cancer follow-up consultations (see Hewitt et al., 2006;Krebber et al., 2016;National Comprehensive Cancer Network, 1999). The DT+PL has since been implemented in many hospitals in the Netherlands ( Van der Linden & Hoekstra-Weebers, 2016;Van Nuenen et al., 2017) and abroad, with the aim to make sure that all possible areas of distress are included in the consultation, and that psychosocial distress is not overlooked.
Socio-medical studies on the effectiveness of the DT+PL and comparable quality-of-life screening tools have shown mixed results (Girgis et al., 2018;Kotronoulas et al., 2014;Palmer et al., 2011). Some studies report more frequent discussions of psychosocial distress (Detmar et al., 2002;Velikova et al., 2004), better targeted referrals (Van Nuenen et al., 2017), and patients commenting positively on the tool (Van Nuenen et al., 2018), while others conclude that such screening tools have little to no effect on patients' quality of life (Boyes et al., 2006;Hollingworth et al., 2013). Irrespective of the outcomes of these studies, what remains under the radar in quantitative research designs is how the implementation of this type of instrument impacts the lived experiences of the participants in the consultations themselves. To our knowledge, there is only one qualitative study that addresses this issue. Biddle et al. (2016) use interviews to investigate how both doctors and patients experience the introduction of the DT+PL. In this paper we build on their research with a case study that does not only draw on meta-data like interviews but also on interactional data from actual consultations with the DT+PL conducted in one of the main cancer centres in The Netherlands. This will show not just how doctors and patients reflect on the use of the DT+PL, as in Biddle et al. (2016), but it will also highlight how its introduction impacts doctorpatient discourse during the consultation, and how doctors and patients negotiate the affordances and constraints created by the tool. As far as the authors are aware, no such study into the interactional effects of a discussion tool such as the DT+PL has been done before.
Since medical consultations are well-defined institutional communicative events with clear roles, norms and expectations (Ten Have, 1989), the assumption is warranted that the introduction of the DT+PL will have consequences for the organization of the consultation as a speech event (Hymes, 1977). We will, therefore, first zoom in on the social context of the "default" follow-up head and neck cancer consultation (without the DT+PL). This will provide us with a benchmark, which we will use to identify changes in the contextual parameters of the speech event, and in the interactional organization of consultations with the DT+PL as they were conducted by two physicians at the hospital. For our analysis, we will use video data of the consultations, triangulated with excerpts from interviews with the physicians and excerpts from a training session on the use of the DT+PL the doctors took part in (see Green & Wallat, 1981).
Our research question in this investigation is: in what ways does the implementation of the DT +PL impact the contextual parameters of the Dutch follow-up cancer consultation, and how do these effects resonate in the interactional organization of the encounter?

The distress thermometer and problem list: form and procedures
The DT+PL is a holistic assessment tool. It consists of a form comprising the image of a thermometer and a list of items patients can tick (see Figure 1 below).
In the top left-hand corner instructions are given to the patient on how to use the form: "How much distress are you experiencing in terms of problems, complaints, worries? Please fill out the thermometer first. Circle the number in the thermometer below that best summarizes how much distress you have been experiencing in the last week (including today) in terms of physical, emotional, social and practical issues".
The number filled out on the Distress Thermometer, then, serves as an indication of patients' own assessment of their overall level of distress. For the Netherlands, research indicates that patients scoring 5 or higher experience problematic distress (see Oncoline, 2010;Tuinman et al., 2008).
The Distress Thermometer goes together with the Problem List, on which patients can indicate 47 possible areas of distress, 25 of which are physical and 22 psychosocial. For the Problem List (PL) the instructions read: "Please indicate for the areas below whether they have caused you distress or problems in the last week (including today). Please put an X at every question". The PL is followed by a section that gives patients the opportunity to add problems that are not on the list if they find them missing ("other problems") and by an invitation to indicate whether or not they would like or need help in dealing with their problems ("would you like support or help with the problems you have indicated?").
Patients are asked to fill out the DT+PL before they go in for their consultation, and hand it to doctor for discussion. Like all health workers who start working with the tool, the doctors in the study were invited to take part in a training session taught by an oncological nurse experienced in both using the tool and training people in its use. During the session she phrased the general aim of the DT+PL as follows: "[it should be used] to signal and discuss physical, psychosocial, practical and spiritual distress so that quality of life can be improved and serious problems can be averted". This meant, she said, that if patients score less than 5, problems should be discussed briefly, with reference to self-help and the possibility of referral to a specialist. If patients score 5 or more, a more elaborate discussion of the problems is called for, and possibly referral to a psychologist, psychiatrist or rehabilitation physician. In her experience, however, patients rarely want a referral (see also Tuinman et al., 2015); for patients who feel depressed, sombre and/or anxious, mostly it would be enough to be given the opportunity to bring this up; it would let patients who never dared mention these problems before know that it is alright to discuss issues of this kind (but see Biddle et al., 2016, where patients indicate they want a "demonstrable outcome" to filling out the DT+PL).

Background of the data
The data in this study derive from consultations involving head and neck cancer survivors. We have opted for this group of patients, since they are an especially relevant group for research in terms of psychosocial distress. Gil et al. (2012) show that in head and neck cancer survivors levels of anxiety and depression are especially high; after treatment they "were more distressed (HADS-Total) than other groups of patients (head and neck cancer, 14.78; breast cancer, 8.88; colorectal cancer, 10; p = 0.05). In terms of coping scores, head and neck cancer patients had higher levels of anxious pre-occupation than other patients (head and neck cancer, 23.42; breast cancer, 22.49; colorectal cancer, 19.91; p = 0.05)" (p. 364). This may be due to the high impact this form of cancer has on patients' wellbeing, with a range of aftereffects, which vary greatly in severity and duration. They include, for example, disfigurement through the removal of, e.g., an eye, or ear and other facial scarring; lack of mucus which leads to difficulties swallowing and eating; and removal of the voice box, or (part of) the tongue, which leads to communication problems. Apart from this there are long-term effects of radiation therapy, such as thyroid problems which may lead to continuous tiredness.
The data for the study were collected at the outpatient clinic of the head and neck cancer department of a specialized cancer hospital in the Netherlands. 1 One of the rooms at the clinic was equipped to record the consultations. Since the camera was unobtrusively placed, amidst an array of medical machinery (see red circle in Figure 2), it was not immediately noticeable, nor was it in the participants' way. The researcher was not present in the consultation room, to minimize intrusion. Patients had been informed about the study, and given their consent beforehand. To make doubly sure patients were aware of the study and were consistent in giving consent, they were met by a research assistant before they went into the doctor's office on the recording day, to confirm that they had learned about, understood, and agreed in writing to participate in the study. Doctors also usually spontaneously referred to the fact that the consultation was being recorded at the beginning of the consultation. 2 Two physicians (Dr A and Dr D) of the head and neck oncology team and 28 of their outpatients participated in the current study. The patients in the Intervention group filled out the DT+PL before the consultation and handed the form to the doctor, so that it could be discussed during the consultation. Those in the Control group filled it out afterwards. For the collection of the benchmark/control group data, two sessions of standard follow-up consultations per physician were video-recorded, 14 for Dr A and 16 for Dr D. For the intervention study, 20 follow-up consultations were recorded in which the DT+PL was used, 10 for each doctor. A number of recordings could not be used because of technical and language problems, or withdrawal of consent. The remaining recordings were viewed to select the consultations that fit the focus of the current study. Since the study focusses on routine follow-up cancer consultations, the patients had to have finished treatment for a malignancy and they had to have come in for their scheduled follow-up cancer consultation. Patients who did not meet these criteria were excluded from the study. This resulted in a final dataset of 28 recordings, 15 in the Control group and 13 in the Intervention group. A detailed transcription, including relevant prosodic, paralinguistic and nonverbal moves, was made of the videotapes of the consultations for analysis. (For transcription symbols, see Appendix 1.) Apart from the video recordings of the consultations, three sets of meta-data were collected: (1) post-consultation session interviews with both physicians. In these interviews the doctors more generally reflected on the practice of follow-up consultations; (2) video data from a training session on how to work with the DT+PL, which the doctors attended before the Intervention consultations; (3) patients' filled out DT+PL. The patients in the Control group filled out the DT+PL after the consultation and handed it to the research assistant. The patients in the Intervention study filled out the DT+PL before the consultation and handed it to their doctor for discussion during the consultation.

Theoretical framework
In order to identify the main parameters of the cancer follow-up consultation and to bring to light the routines, norms and expectations that the participants in the consultation (doctor, patient and sometimes companion) orient to during the interaction, we adopt a micro-ethnographic approach (ethnography of communication, Hymes, 1977). This approach focuses "simultaneously on language use and matters of context" (Kalou & Sadler-Smith, 2015, p. 630), while endeavouring "to capture what people say and do as a product of how they interpret the complexity of their world" (Sevigny, 1981, p. 68).
In this study we combine etic and emic perspectives. The etic perspective, which involves a more deductive approach to the data, makes use of the SPEAKING model, developed by Hymes (1977) and elaborated by Saville-Troike (2003). It is a comprehensive instrument for the analysis of the contextual dimensions of institutional interactions; it "enables the specification of the content and context of speech … and provides an etic framework from which an emic account can be built" (Kalou & Sadler-Smith, 2015, p. 637). The basic unit of analysis in this framework is the speech event, i.e. a clearly recognizable communicative activity, with a specific aim, "governed by rules or norms for the use of speech" (Hymes, 1977, p. 56), such as, for instance, business meetings, court cases, university lectures, and medical consultations.
The name of the model is an acronym, referring to, respectively, the "Situation", i.e. the spatial and temporal setting and scene of the speech event which determine what speech acts and physical actions are appropriate "here and now"; the set of "Participants" and their roles; the "Ends" or aims of the interaction; the "Act sequence" of the event, i.e. the ordering of the communicative acts and the routines that typify the event (similar to move structure, Swales, 1990); its "Key" or mood (e.g., formal/informal); the "Instrumentalities" (called "Message forms" in Saville-Troike, 2003) or channels of communication; the "Norms" governing the interaction (turntaking patterns, allowable contributions) and the "Genre" or type of interaction.
Since a speech event is an organic whole, each of its elements is reflexively related to the other elements. For instance, the different Ends may influence the way the interaction is organized sequentially (Act sequence), the channels of communication (Instrumentalities), the designated speaker/hearer roles of the participants (Participants), and the ways in which the participants coconstruct the interaction. Careful scrutiny of the contextual parameters of events and their interrelationships then will enable the researcher to come to a deeper and more nuanced understanding of the speech event and will "allow the researcher to reach an informed interpretation of what she has witnessed" (Kalou & Sadler-Smith, 2015, p. 643). Section 5 shows the results of an etic analysis of the "standard" follow-up head and neck cancer consultation, making use of the categories of the SPEAKING model.
The emic perspective of the study, focussing on the lived experience of the participants, involves an inductive approach. It focusses on an analysis of the interviews and the training video, plus a turn-by-turn analysis of the videoed consultations, making use of insights from discourse analysis and pragmatics. Unlike etic approaches, such an analysis does not have as its base an a priori set of categories. It is data driven, in that the analysis focusses on ways that the participants themselves shape and co-coordinate their interactions. The researchers aim to distil from their analysis patterns of behaviour constructed, and concerns expressed by the participants in the speech event. The results of this analysis are presented in section 6.
The combination of these perspectives will show both how the DT+PL affects the consultation as a speech event, and how doctor and patient co-construct the various affordances and constraints generated by it use.

The follow-up head and neck cancer consultation as a speech event
In this section we will use the SPEAKING model to establish the contextual parameters of the "standard" follow-up head and neck cancer consultation in the hospital investigated in this study as a basis for identifying and unravelling the complexities that the DT+PL may introduce into this event.

Situation
Typically, consultations take place in a hospital examination room, with a pre-scheduled, tight timeslot of 10-15 minutes. If consultations overrun that timeslot, this has severe real-world consequences: other patients will have longer waiting times, and the doctors will have longer consultation sessions, which might interfere with their other professional duties. Moreover, contracts with health-insurance companies determine the budget and therefore the duration of a consultation. Exceeding standard consultation time is not funded.

Participants and participant roles
The event has a default minimum set of participants: a physician and a patient. Sometimes the patient is accompanied by a third person (e.g., spouse, family or friend), who we will refer to as "companion", and sometimes a trainee or visiting physician will be present at the consultation as an observer. Doctor-patient participant roles are asymmetrical. The doctor is the expert participant in the consultation; he or she is in charge of the overall agenda for the consultation and manages the interaction, opening the consultation proper, moving into the question-answer sequences of the anamnesis phase, deciding when to conduct the physical examination, initiating wrap-up etc. The patient usually follows the doctor's lead in this (see Ainsworth-Vaughn, 2003). However, it is important to note that the follow-up consultation is a "return visit", and as the follow-up phase continues over a period of 5 years, patients gradually become familiar with the routines of the consultation; they become "expert patients", and may comment on the procedures or make suggestions for what they feel needs to happen (next).
Participant roles are closely linked to the aims of the consultation. The doctor's role is to monitor the patient's condition, check for recurrence or second primary tumours and guide the patient in rehabilitation. The role of the patient is to inform the doctor of how they experience their condition, including any symptoms, worries and questions they have, and to gather information from the doctor on their condition, how to cope with after-effects, how to treat wounds, etc.

Ends and Key
The Ends or Aims are also linked to the Key or mood of the interaction. On the one hand, there is an orientation towards distress, e.g., anxiety about recurrence-which occurs quite frequently in head-andneck cancer patient (Ghazali et al., 2013)-, discomfort through physical after-effects, and psychosocial problems involving the condition of being a cancer survivor. On the other hand, there is an orientation towards wellness, e.g., continued absence of malignancies, diminishing of after-effects, plans for a healthier life-style. The Key in the interaction, co-constructed by the participants, therefore varies between negative and positive, depending on the particular topic under discussion and the level of distress. A positive key may also be related to, e.g., participants' desire to seek and give reassurance, minimize distress and focus on hope in difficult circumstances (see, e.g., Beach, 2013;Sandén et al., 2001; Van der Laaken & Bannink, Manuscript in preparation b).

Act sequence
The Act sequence of the follow-up consultation-i.e. the elements that, routinely, need to be part of the event, and the order in which they normally occur-is very similar to that of surgeon-patient visits documented by White et al. (2013). The physician fetches the patient from the waiting room and guides them to the examination room (see Van der Laaken & Bannink, 2019). When doctor and patient enter the room and settle down, there is usually a brief spell of chitchat in which doctor and patient re-establish their relationship (1). The doctor then opens the consultation proper with a "How are you?" question (HAY; see, e.g., Coupland et al., 1992;Heritage & Robinson, 2006;Van der Laaken & Bannink, 2019) to solicit patient's assessment of their overall condition (2). During anamnesis doctor and patient discuss, among other things, any complaints patient may have, and how they are coping with any after-effects of treatment and malignancy. This episode is followed by a physical examination (3) and discussion of findings, reassurance or rationalization of further tests (4). Advice and/or next steps in terms of scheduling further tests or referrals (5), and a timeline for the next visit (6) signal the closing of the consultation. Steps 1-3 usually occur in this order, but 2 and 3 may be recycled at any point; steps 4-6 may merge completely but logically follow 1-3. Steps 2 and 3 are related to the objective of monitoring for recurrence and after-effects; step 4 to reassurance and relieving anxiety. Chitchat may occur at any point during the consultation. All through the consultation the doctor has access to patients' medical files on the computer, which they update "live" during the consultation.

Norms of interaction and instrumentalities
During the anamnesis and the physical examination, there are clear Norms for interaction: doctor and patient engage in question-answer-discussion sequences "designed to elicit information that is complete and accurate enough for the clinician to arrive at a conclusion" (Frankel, 1995, pp. 247-248). Patients rarely ask questions in these phases so as not to interrupt the doctor's data gathering (see Frankel, 1990;Gill & Maynard, 2006). Once the data-gathering is complete, patients (spontaneously or at the invitation of the doctor) do ask questions, ranging from practical topics like wound treatment to existential anxiety about possible recurrence.
The Instrumentalities in the consultation are mainly the use of voice and gesture, (and touch during the physical examination), but communication also frequently occurs making use of written text on paper or computer, e.g., patients who have brought in lists of questions, and the patients' medical files.

Genre
The follow-up head-and-neck cancer consultation displays the generic features of the broader family of medical consultations (see, e.g., Ten Have, 1989).

Expected effects of the DT+PL
We propose that the implementation of the DT+PL can be expected to affect the contextual parameters of the "standard" consultation on different dimensions. The introduction of a new Instrumentality-the written form carrying extra information about the patient-by default adds a new item to the agenda of the consultation: the discussion of physical and socio-psychological problems the patient has indicated on the PL, which will have to be negotiated without exceeding the 15 minute time limit that has been set for the consultations. The tool may also influence both the routine Act sequence (at what point in the consultation should it be brought up?), and the Participant roles of both patients (who get added agency by the opportunity to nominate particular topics and who may not expect to discuss psychosocial issues with their oncologists) and doctors (who are trained to deal with medical issues and not necessarily with the psychosocial problems that are also featured in the PL).
Below we will see how these projected effects surface in the thoughts and comments of the physicians about the new tool, and how the introduction of the DT+PL plays out in the discursive construction of the consultations.

The data
The analysis of the data is organized in terms of a selection of features from the model of the consultation as a speech event discussed above: Aims and Participant roles, Act Sequence, Time, Key and Instrumentalities. Elements of the model where little or no impact was observed (Situation, Norms and Genre) have been left out. Each section will first focus on what we will refer to as meta-data: how participants in the interviews and training session report on that particular aspect of the speech event, and the (expected) impact the DT+PL has on this. This picture is then refined by means of the interactional data: a turn-by-turn analysis of selected extracts from the consultations.

Aims and participant roles
In the interviews we conducted with the doctors they are very clear about what they see as the most important aims of the follow-up consultations, as illustrated in Data 1 and 2: Dr D: Look, if someone has had an operation and everything is OK, well, then I ask how things are, and whether there are any complaints, and then I check them over very carefully, and then they are on their way home within a few minutes. And then they are really pleased to be outside again so quickly.
[…] And if they have complaints then of course they are afraid that the complaints are related to the disease and that it has returned. Then the aim is to reassure them as fast as possible. So that means doing the physical examination a.s.a.p.

Data 2 (Interview)
Dr A: Check if patient is doing OK, if there is no tumour, so that is the medical, content, business part, the oncological part. Apart from that, it is important whether patient needs to rehabilitate as far as that is possible, so primarily from the patient's perspective. And lastly as a doctor you need to become aware of what the consequences of your treatment may be.
Both doctors indicate that monitoring patients for signs of recurrence, metastasis and second tumours, checking how they are coping with after-effects, and relieving their anxiety on these issues are central to their aims. They do not seem to be focused on psychosocial distress, similar to doctors in other reports on the subject (see, e.g., Beach et al., 2004;Ford et al., 1996;Zhou et al., 2015). This can be seen in Data 3, in which Dr D responds to the question whether he discusses psychosocial issues with his patients: Here, Dr D is quite outspoken about what he considers to be his "job" and makes clear that discussing the psychosocial distress his patients might suffer from is not central to it; other people are better qualified for this (see Biddle et al., 2016). His view appears to be shared by Dr A. During the training session she explains her position by pointing at what she feels as her lack of expertise in the matter: This perceived lack of training in how to recognize indirect signals of psychosocial distress and how to distinguish expected levels of distress from excessive, problematic ones is an important reason for the lack of focus on such issues of the physicians in our study. They say that, as oncological surgeons, they view patients and their complaints through a medical lens: they are trained to treat physical conditions and this is the expertise that they will automatically rely on-and go back towhen they are confronted with a patient's problem, as is illustrated in Data 5: Data 5 (Interview) Dr D: yes we only look at eh, tired? Low Hb level-and that is as far as we get, right, as a matter of speaking.
The data suggest that the implementation of the DT+PL may compel doctors to come out of their comfort zone. The tool will force them into engaging with psychosocial distress, an area that they do not feel expert in.
The DT+PL may also have an impact on the participant roles of doctor and patient, as the oncological nurse puts it during the training session: Data 6 (Training session) Trainer: Of course, with this tool you give the patient the lead in the interaction: these are the problems, this is what I want to discuss.
In Data 6 the trainer spells out one of the interactional effects of the introduction of the DT+PL: the tool will give patients agency, in the sense that they will be given an active role in the selection of "mentionables" (Button & Casey, 1984;Schegloff & Sacks, 1984)-points worthy/suitable to be addressed during the event. The trainer indicates that the information the patients supply in the PL needs to be constructed as an extra item on the "agenda" of the consultation. In this way, the form functions as a topic elicitor and its content as a topic pre-nominator.
The interactional data from the consultations themselves show, however, that doctors still have a crucial role in the on-line discursive construction of the topic initiation during the event, as can be seen in Data 7: Data 7 [Dr A. Patient's DT score is 6 with six problems crossed on the PL, two of which are emotional (fear and tension). Data start after 9 minutes; in Anamnesis doctor and patient extensively discussed patient's headaches and problems with excessive earwax] By relaying what the patient has filled in on the form (P1), i.e. that he suffers from fear and tension, doctor discursively constructs mutual orientation to these problems. Patient reacts to doctor's initiation with the observation that the two symptoms go hand in hand (P2, P4). He only now-in response to doctor's prompt-links them with the physical problems he has with his ear and his head (problems which he and doctor have just discussed at great length), saying (P6): maar dat [de spanning en angst] kan in principe te maken hebben met die met dat oor-met dat hoofd ("but that [the tension and fear] may in principle have to do with that with that ear-with that head"). So, although patient had prenominated anxiety and fear in the PL before the consultation, the doctor was still crucially instrumental in topicalizing patient's psychosocial distress in the actual interactional event.
At the same time, although our dataset shows that the DT+PL indeed succeeds in creating the affordances for patients to co-set the agenda for the event, the information patients indicate on the PL+ DT about their concerns is not always complete, and doctors are sensitive to this, as is illustrated in Data 8:

Data 8
[Dr D. Patient's DT score is 0; only two items crossed on the PL (dry itchy skin and dizziness).

Data begins at opening consultation.]
In P3-P7 doctor and patient co-construct an orientation to patient wellness, with the doctor in P5 using extreme language helemaal goed ja ("great yes") to upgrade his assessment. Then, after a short pause, doctor re-checks with a "no-valenced optimized" question, geen nieuwe klachten verder ("no new complaints otherwise"). The preferred response to this type of question would be a positive confirmation (Heritage, 2010). The doctor's addition of of ("or"), however, cancels the preferred response and opens up the option of relating a negative health outcome. Patient's answer to the question is in line with the positive Key of the interaction before this point (P9), but his nee ("no") overlaps with companion's nou ("well")-a discourse marker that signifies disagreement and a change in footing (a complexity marker, see Mazeland, 2016). Patient immediately latches on to his companion's interruption (P11) and takes over by reporting what she has noticed (and probably was about to bring up herself): mijn vrouw zag hier ((wijst naar zijkant hoofd)) hier aan die kant wat wat ("my wife saw here ((points at side of head)) here on that side some some"). At this point companion takes over completely and tells doctor her point of concern: a small shiny spot that she discovered when she was cutting patient's hair. With patient's medical history (melanoma on the ear) unexplained spots would clearly constitute a warning signal to patient and companion, but the concern was not indicated in the PL. Sensitive to their worry, doctor responds immediately (P17) and shows them that he is taking the complaint seriously by announcing: = uh we gaan even kijken (" = uh we are going to have a look").
Data 8 shows that, although the DT+PL affords patients the agency to nominate topics for discussion, patients sometimes omit to indicate problems on the PL that are relevant for discussion in the consultation-be it through forgetfulness, because they doubt the legitimacy of their concerns, or for other reasons (e.g., that patients do not consider this doctor or this occasion to be the relevant platform to discuss certain issues) (see Biddle et al., 2016). The physicians in our data seem very aware of this; again and again they check and recheck patients' self-assessments. Sometimes prompted by the doctor and sometimes spontaneously, patients on average nominate just over 2 issues outside the PL per consultation.
In the meta-data we saw that the doctors and the trainer discussed the potential agency of patients to put the discussion of psychosocial distress on the agenda of the consultation. In practice, the agency of the patient seems more relative: the interactional data show that doctors still have an important role in actually initiating these topics for discussion, with the patients responding to explicit openings by the doctor.

Act sequence
As shown in section 5, the standard follow-up cancer consultation has a clear Act sequence. An important question, therefore, is at what point in the consultation the content of the DT+PL form should be brought up. In the training session this dilemma is addressed by the oncological nurse: Data 9 (Training session) Trainer: You first just need to consider when the Distress Thermometer gets to be discussed -first physical examination and then pay attention to the Distress Thermometer for a while, or first the Distress Thermometer and then switch to the physical examination.
The trainer here reduces the issue of the timing of the discussion of the DT+PL in the consultation to a simple, binary choice-either before or after the physical examination of the patient-, but Dr D's response shows that he feels that the situation is more complex: Data 10 (Training session) Dr D: […] because usually the people are still nervous when they come, and they are only denerved, if you like, if it is OK. You cannot postpone the physical for too long […] they are on tenterhooks. So I don't think that you can do this first [i.e., discuss the DT + PL]. First the physical examination needs to be completed, I think […] but on the other hand you do want to, because you want to-the first question is "How are things?" That is your first strike always-then you know how things are, right.
Clearly, Dr D is in two minds about the best slot in the consultation for the discussion of the DT +PL: on the one hand, he feels the physical examination should be done as early as possible, since patients need to be reassured that they are-at this point in time-free from cancer; on the other hand, it seems to him that the topic would naturally tie in with the "How are you?" question that routinely signals the opening of the consultation proper (see Van der Laaken & Bannink, 2019). He does not commit himself to any choice, but Dr A formulates a clear plan of action: Data 11 (Training session) Dr A: So I think that I will just look physically, what are the physical complaints, then I'll ask my own questions, and then I'll say, I'll just say, I will be happy to go into it into the questions on the Distress Thermometer after the examination. I just want to check whether that is all OK first.
After careful consideration, Dr A decides that she will only discuss the DT+PL after she has performed her normal routine and she is sure that "all is okay". Note how she defines "okay" as an exclusively medical state.
Dr D, however, detects a problem with this strategy: Data 12 (Training session) Dr D: But then you've got, yes then it is like a bit repetitive, right because if you say, "How are things?", then you will probably broach the first big topics […] and later you would have to come back to that.
It is clear that Dr A and Dr D feel they are facing a dilemma in terms of Act sequence. Should they discuss the DT+PL immediately after the "How are you?" question, after the anamnesis, after the physical examination or at the very end of the consultation when everything they "normally" do has been said and done, with the possibility that repetition creeps in, and consequently, that time limits are exceeded?
In the end both doctors chose to discuss the DT+PL-in varying degrees of detail-during anamnesis, finishing its discussion before the physical examination, in all consultations in the current dataset. Below we will zoom in on the consequences of this choice for the interactional organization of the encounter.
In Data 13 Dr A mentions the DT+PL at the start of the consultation, but then explicitly postpones its discussion to a later moment: Data 13 [Dr A. Patient's DT score is 6, which signifies considerable distress, with 6 items crossed on the PL, two of which are emotional (fear and tension). Data start at the beginning of the consultation, just after participants have entered the consultation room.] After doctor and patient have entered the examination room and settled down, Dr A establishes the procedure she intends to follow for the DT+PL (P3). Being handed the form, she topicalizes it by saying: ga ik zo even naar kijken ("I will look at it in a minute"), deferring the discussion of the DT +PL to an undefined but not far-off point in the future of the consultation. She immediately adds: eerst even het gewone ("first just the standard"), further clarifying her agenda for the consultation: she will stick to her normal routine, i.e. first go through the steps of her "ordinary" consultation before she discusses the DT+PL, implying that it will only come to the table at that moment as an extension of her normal routine.
Just as Dr D predicted in Data 12, when doctors first "do what they always do"-i.e. first take a complete anamnesis-and only then go through the DT+PL, the result might be that points that have already been addressed are revisited. Sometimes, however, this strategy creates affordances for new insights into the patient's condition, as illustrated in the data below.

Data 14
[Dr A. Patient's DT score 7.5 (high distress), with 25 elements crossed on the PL, 7 of which are emotional (emotional control, self-confidence, fears, depression, tension, loneliness, loss of control). Doctor and patient have already discussed a number of after-effects that patient still experiences, and then discussed various points on the PL.] On checking the DT+PL, doctor communicates to patient that, although the form contains mostly known information (P1), she has found something she thinks should be brought up: patient's depression and anxiety (P3). She implies she was not aware of this by means of the Dutch discourse marker toch-roughly equivalent to "after all"-which indicates that speaker's assumptions have been inverted (see Zeevat, 2000). She introduces the topic hesitantly and prefaces it with much delay and a raft of discourse markers hHh is dat ehh () dat u toch ook wat ("hHh is that ehh () that you are after all also a little"). This mitigates the possibly face-threatening act (FTA; Brown & Levinson, 2009) of introducing the sensitive topic of emotional/mental problems such as anxiety and depression and may also be indicative of the doctor's diffidence to discuss these issues (see Biddle et al., 2016;Van der Laaken & Bannink, Manuscript in preparation a). Patient confirms that she suffers from anxiety and depression (P4) but at the same time mitigates their scope: soms wel ja ("sometimes I am yes").
In some consultations in our dataset, the introduction of the DT+PL itself can be seen to present an interactional dilemma for the physician in the opening stage of the consultation: the presence of a form with information on the physical, spiritual and psychosocial well-being of the patient, filled out just minutes before the event begins, infringes on the ritual opening of the consultations, as is shown in Data 15:

Data 15
[Dr D. Patient's DT score is 0; no items crossed on PL. Data from opening of consultation. Patient has taken the DT+PL from her bag and put it in front of the doctor on his desk] By opening the consultation (P1) with "[I] don't have to ask you anymore how you are doing because eh: you have already °filled° this out" ([Ik] hoef niet meer te vragen hoe het met u gaat want eh: u heeft dit al inge°vuld°) doctor makes explicit the interactional dilemma that is the consequence of the procedural arrangements regarding the DT+PL. The fact that patient hands the filled-out form to the physician before the beginning of the consultation proper affects the routine act sequence: it effectively compromises the ritual opening HAY question that all but one of the consultations in our dataset begin with (see Van der Laaken & Bannink, 2019), and may render (parts of) the anamnesis superfluous. The DT+PL is here constructed as supplying doctor with all necessary information about the situation of the patient, and since it is dispreferred to ask for known information (this would be violating Grice's maxim of quantity; Grice, 1975), the doctor is robbed of his ritual opening. The dilemma is recognized by all involved: doctor's remark is met with broad smiles (P3). After reading the DT+PL and some chitchat (not in transcription in Data 15), however, doctor does ask his HAY question after all (P4), creating relevance for the question (see Sperber & Wilson, 1986) and solving his interactional problem by adding the qualification verder ("other than that"), to open the possibility of nominating complaints that may not have ended up on the form.
The meta-data show that doctors were doubtful about when and how to introduce the discussion of the DT+PL in their consultations. They were not sure whether to discuss the document before or after the physical examination; they worried that its discussion might extend patients' anxiety about possible recurrence; and they surmised that it might interfere with the "normal" interactional routine of the consultation. The interactional data show that the doctors in all cases opted for the discussion of the DT +PL before the physical examination, as part of anamnesis. They show doctors explicitly managing the act sequence of the consultation to accommodate the discussion of the DT+PL at an opportune moment (Data 13), and being acutely aware of the impact the DT+PL has on their normal routine of opening the consultation with a "How are you?" question (Data 15). The risk of repetition they discussed in the training session turns out to be warranted, but this may also lead to new patient problems being discovered, which otherwise might have remained undiscussed.

Time
As mentioned in section 5, the doctors in the hospital are restricted in the amount of time they can spend on the individual consultations. If they exceed the 15-minute timeslot, this has real-world consequences for all parties involved. Time is, therefore, a prime concern for the doctors in their discussions of the consequences of the implementation of the DT+PL (see Biddle et al., 2016), as can be seen in the data below: Data 16 (Training session) Dr D: if you have to engage with this […], there is very little time already, and this will cost a lot of time, one way or another.
Dr A: it all costs money and time, and that is simply not available.
However, when comparing the duration of the consultations of the two doctors who participated both in de Control study and the DT+PL study, it turns out that the DT+PL may not be the only factor influencing the duration of the consultation: the average duration of the consultations varies considerably, both between doctors and between conditions (with and without the DT+PL), as can be seen in Table 1. In the Control group, Dr A's consultations took 13:05 minutes on average, versus 8:04 minutes for Dr D's. Dr A's DT+PL consultations took 22:56 minutes on average, whereas Dr D's took 8:30 minutes. So in both conditions, Dr A's consultations took (much) longer than Dr D's. Dr A's DT+PL consultations took almost 10 minutes longer than her Control group consultations, whereas Dr D's DT+PL consultations on average took roughly the same amount of time as his Control group consultations, around 8 minutes. Also, there is considerable variation within each group, and Dr A's consultations were more varied in duration than Dr D's. Dr A's consultations vary between 05:20 and 51:50, and Dr D's between 04:19 and 12:49. 3 In Dr A's Control group the standard deviation from the average is 04:36 minutes, but in her DT+PL group it is much higher still, 15:57 minutes. In Dr D's Control group SD is 03:34, and slightly lower still in his DT+PL group, 02:56. This variation in the duration of the consultations suggests that increased duration cannot simply be explained by the introduction of the DT+PL; there is quite a large difference between the two doctors, both with and without the DT+PL.
One possible reason for this difference may be the problem load of the patients. To get an overview of the problem load of the patients, we counted the number of items crossed on the PL and the number of issues outside the PL that they raised during the consultation. A comparison of  the average problem load of the patients shows considerable variation between doctors and conditions (see Table 2).
In the Control groups, Dr A's patients listed a total of 77 problems on the PL and raised a further 14 problems in the consultations, resulting in an average problem load of 10.11 problems per patient. Dr D's six patients listed 40 problems on the PL, and nominated 7 problems during the consultation which were not on the PL, resulting in an average of 7.83 problems per patient. In the DT+PL groups Dr A's eight patients crossed 106 problems on the PL, and in the consultation nominated 26 problems outside the PL, thus averaging 16.5 problems per patient. Dr D's five DT +PL patients crossed a total of 35 problems on the PL and nominated 4 problems outside the problem list in the consultation, averaging 7.8 problems per patient.
Comparing the problem load per doctor, we see that in both conditions, Dr A's patients on average have a higher problem load than those of Dr D: 20% higher in the Control group, and twice as high in the DT+PL group. Also, the problem load of Dr D's patients is-on average-stable in the two conditions, whereas the average problem load of Dr A's DT+PL group is more than 50% higher than that of her Control group. The higher problem load of Dr A's patients in both conditions, then, might partly explain the longer duration of her consultations: there is simply more to discuss.
There is also a difference in the number of problems that is actually addressed, and the time devoted to their discussion (see Table 3).  In Dr A and Dr D's Control groups, an average of 3 to 3.3 problems per patient is discussed, yet Dr A's consultations are much longer than Dr D's (13:05 versus 8:04 minutes on average). In Dr A's DT+PL consultations, an average of 10.4 problems per patient is discussed, whereas in Dr D's DT+PL consultations, an average of 2.8 problems per patient was discussed. So in the DT+PL group, Dr A discusses 3.5 times as many problems per consultation as Dr D (10.4 versus 2.8), which would contribute to the longer average duration of her consultations.
For Dr D, the average number of problems of the patients and the duration of the consultations remain stable over the two conditions. However, the average number of problems discussed per patient goes down in the DT+PL group (2.8 versus 3.33), the percentage of problems discussed is slightly lower (36% versus 42%), and the time spent per problem goes up by 35 seconds on average. For Dr A, the average duration of the consultations in the DT+PL group is almost double that of the Control group, the number of problems discussed more than triples, a much larger percentage of the patients' problems is discussed (63% versus 30%), but -in spite of the longer duration of the DT+PL consultations-the time per problem discussed is almost halved. 4 Where Dr D discusses more or less the same number of problems per patient in more or less the same amount of time in both the Control group and the DT+PL group, Dr A discusses almost three times as many problems per patient in the DT+PL group than in the Control group, and her DT+PL consultations are almost twice as long. These differences may be due to the way in which the two doctors handle the discussion of the DT+PL: Dr A tended to first go through her normal routine and then discuss the DT+PL, usually starting with items that stood out to her and then discussing the items ticked by patient one by one (see, e.g., Data 7,13,14), whereas Dr D focused the discussion of the DT+PL on "new" complaints (see, e.g., Data 8, P8), thus limiting the number of issues discussed.
In the meta-data doctors expressed their concern that the introduction of the DT+PL would lead to longer consultation times, and the figures on duration seem to bear them out. However, the interactional data reveal a number of variables that affect the duration of the consultations and account for the differences between the doctors: not so much the introduction of the DT+PL itself causes the increase in problems discussed and therefore in the duration of the consultation, but rather matters such as patients' problem load and doctors' strategic approach to the implementation of the discussion of the DT+PL in the consultation.
The DT+PL consultations then seem to be more focussed on discussing problem and distress. This was to be expected as they have been pre-nominated as topics for discussion by the patients when they filled out the PL.
The interactional data, however, show that in the DT+PL consultations there is a similar mix of positive and negative Key in the way these problems are formulated as in the Control group consultations: problems may be emphasized, but they may also sometimes be mitigated or denied; patients may emphasize their wellness in positive glosses after doctor's opening HAY question, but immediately follow them up with a (mitigated) account of remaining problems (see Van der Laaken & Bannink, 2019); and sometimes the DT+PL itself is used to construct a positive Key. 5 Two sets of data will illustrate this.
A closer look at Data 7 shows that patient emphasizes and elaborates on his fears, and establishes them as a natural reaction to cancer, thus focussing the discussion on problems and distress.
Data 7 (repeated) [Dr A. Patient's DT score is 6 with six problems crossed on the PL, two of which are emotional (fear and tension). Data start after 9 minutes; in Anamnesis doctor and patient extensively discussed patient's headaches and problems with excessive earwax] In P1 doctor starts reading the DT+PL and states that patient has indicated that he suffers from fear and tension. Patient confirms this, commenting that the two go hand in hand (P2, P4), which doctor confirms (P5). Patient then links the physical problems he has with his ear and his head to the anxiety he indicated in the DT+PL: Maar dat kan in principe te maken hebben met die, met dat oor-met dat hoofd ("But that may in principle have to do with that, with that ear-with that head") (P6). He states that he experiences these problems as vreemd ("strange"), and then in P10 he naturalizes the fear and tension he suffers from, by means of natuurlijk ("of course"). He does this twice in a row: Dat je je ongerust maakt natuurlijk. Dat is natuurlijk al dat je denkt, zit een tumor in me hoofd ("That you are worried of course. That is of course all that you think, there's a tumour in me head"). With natuurlijk, patient implies that it is self-evident that a patient should be anxious and tense in the circumstances of having had cancer, and that-since he was treated for a carcinoma on his ear-it is "natural" for him to worry that his ear problems might herald a possible recurrence (see Arora, 2003). However, since patient has not raised these worries before, it would seem that patient had not considered the fact that he was worried to be a mentionableif a topic is natural, self-evident, it may not be newsworthy and therefore not topicalizable (see, e.g., Button & Casey, 1984) or mentionable (Schegloff & Sacks, 1984). At the same time, he may simply have been reluctant to discuss emotional problems with his surgeon, considering this outside her brief (see, e.g., Arora, 2003;Beach et al., 2004;Biddle et al., 2016;Heritage & Robinson, 2006). The Key in this data is light from the beginning. In P1-3 Dr D jokes about the impact of the DT+PL on his routine "How are you?" question, and in P4 PT jokes back, smiling broadly, that this will lead them to finish quickly: nou zijn we gauw klaar↑.
Doctor agrees °ja eh uh° ("°yes eh uh°") and then starts reading the DT+PL. In the following interaction the content of the DT+PL is never made explicit, but referred to and responded to almost as if the DT+PL is a separate, independent voice (Instrumentality) in the interaction.
The DT+PL informs doctor that patient is free from complaints (0 on DT, no problems crossed on the PL). This might indicate that there is little to discuss. This inference is reflected in doctor's response to what he has read in the DT+PL: ja we zijn inderdaad gauw klaar ("yes we will really be finished quickly")-there are no problems or complaints to discuss, so the expectation is that the consultation will not take long. Doctor responds to the content of what he has read with a big smile, which would indicate good news, and which links to the jokey mood of the interaction up to this point. In the formulation of his response to the DT+PL doctor refers back to patient's previous utterance that they would be finished quickly with an emphatic inderdaad ("really"). There seems to be no felt need to make explicit what he has read, because both doctor and patient are aware that they now share the information in the DT+PL: patient has filled it out and doctor has read it. Making this information explicit might be considered redundant, i.e. it might constitute a violation of the maxim of quantity, where more information is given than is necessary (Grice, 1975). Doctor then concludes his "interaction" with the DT+PL by putting it down and covering it with his hands.
In P6 patient exhales audibly, which might be interpreted as a sigh of relief or satisfaction, and doctor responds ja ("yes"), and smiles at patient; both are clearly content that patient is doing well. Then patient adds (P8) ja ik kon d'r niks anders van ↑ma:ken ("yes there was nothing else I could ↑ma:ke of it"). With this utterance she confirms doctor's implied inference that she is doing OK by implying that there were no concerns that were causing her distress which she could have included in the DT+PL: there was niks anders ("nothing else") to report than what is in the DT+PL (i.e. absence of distress). Again, there is no explicit reference to the content of the DT+PL. After doctor's continuer ja ("yes"), patient then comments on these positive implications with gelukkig ("fortunately"). The absence of distress on the DT+PL has contributed to an optimistic Key, focusing on wellness rather than distress.
Doctor and patient co-construct a sense of satisfaction with the absence of distress reported in the DT+PL, without explicitly referring to it. The DT+PL-a nonvocal Instrumentality-has been deemed to communicate its message sufficiently without that message needing vocal repetition.
The doctors in the interviews indicated that they feared that the consultations would turn into a cesspool of problems. The figures show that they are partly right in this: more problems are indeed discussed in the DT+PL consultations than in the Control group. 6 The interaction in Data 7 shows that the DT+PL affords doctor and patient a way into discussing psychosocial problems, which might otherwise have remained under the radar. However, as Data 15 shows, the DT+PL can also function as an instrument to generate a positive Key, focused on wellness instead of distress, by showing unequivocally that patient is doing well.

Discussion and conclusions
Where Biddle et al. (2016) were able to show many of the effects of the introduction of the DT+PL on the basis of interviews with doctors and patients, they were faced with problems of recall, due to the fact that the interviews were conducted 13 months after the administration of the DT+PL. They were also-through their reliance on meta-data-unable to show the impact of the introduction of the DT+PL on daily practice, i.e. on the discursive construction of the consultations by doctors and patients.
Our study largely confirms Biddle et al.'s findings with respect to the types of issues involved, but in addition highlights a number of interactional dilemmas and complications for the participants in the communicative event itself. These are related to doctor-patient role relations, the routine enactment of the speech event and to the institutionally required time limits of the consultation. The doctors in this study used different strategies to negotiate these-sometimes conflictingdemands during the actual consultations.
In spite of her concerns about the institutional time constraints, Dr A decided to preserve her normal routine (see Data 13) and to first go through the "default" Act sequence of her routine follow-up consultation and only then to discuss the DT+PL point by point, using the DT+PL as a checklist for the anamnesis. This resulted in the discussion of the DT+PL being "extra", leading to lengthy sessions-the length exacerbated by the large problem load of her patients-and at times to addressing the same problems twice. However, this approach also led to her finding out new patient concerns and problems which otherwise might not have surfaced (Data 7 and 14). In these instances the DT+PL created the affordances for the discussion of these problems.
The risk of repetition and lack of time were problems Dr D had envisaged during the training session (see Data 12). He solved them in the consultations themselves by focusing on new complaints, asking his patients to select one or more topics for discussion from the DT+PL as is illustrated in Data 8 and Data 15. This strategy had an unexpected, extra benefit: it led to the discussion of a complaint that had not been mentioned on the DT+PL. This emphasizes that the instrument, though useful in uncovering hitherto hidden complaints, cannot be trusted to supply all information necessary for a full assessment of a patient's condition (see also Biddle et al., 2016;Higginson & Carr, 2001).
Our data show that the DT+PL was indeed successful in creating the affordances for the discussion of psychosocial and other problems. However, we argue, it might also constrain the topics on the agenda to those issues that are mentioned on the PL-if not used carefully. The doctors in our study show awareness of this by first (or also) going through their "normal", routine anamnesis, and/or by asking follow-up questions to elicit further problems.
Our research shows that it is important to not just trace the ideas and reflections of doctors and patients on the impact of the DT+PL on doctor-patient discourse, but also zoom in on the actual doctor-patient interactions during the consultations. We suggest that further research needs to be conducted with a wider range of cancer patients, in different stages of treatment, to create a more comprehensive picture of how doctor-patient interaction is affected by interventions such as the introduction of the DT+PL as a discussion tool. Important issues to explore include what triggers the longer duration of the consultations we found in our dataset (the simple fact of the introduction of the DT+PL, interactive strategies used by the doctors and/or patients, problem load, etc.), and the broader effects of the tool on topic nomination (who initiates what topic at which point in the consultation).

Limitations to the study
There are a number of limitations to this study. First, the dataset we draw on is too small to warrant general conclusions. We argue, however, that small-scale projects that focus on finegrained, detailed analysis of interactional data will contribute to our understanding of the success or failure of healthcare interventions, such as the introduction of quality-of-life tools in medical consultations, and may therefore be helpful in improving the practices of clinicians (see Barnes, 2019). Second, although doctors' sense that they are ill-equipped to deal with "nonphysical" distress (see Biddle et al., 2016) was brought up as an area of concern in our metadata, we have not addressed this issue in the present paper and not reported on whether, and if