DISCOURSE OF POWER FROM THE ASPECT OF INTERRUPTION IN MEDICAL ENCOUNTERS

The discourse of medical encounters is an excellent example of both institutional talk and the discourse of power and its prominent features can be analysed from various aspects. This paper deals with interruption as an important characteristic of both doctor-patient communication and institutional talk in general. The research is focused on comparing the ways doctors and patients interrupt each other and the amount of power they need for this. First, some previous research in this field has been reviewed – it is discussed how interruptions are different from overlaps, how typical it is for patients to interrupt their doctors, how and why doctors and patients interrupt each other and whether they have equal rights when it comes to interrupting their interlocutors. As we aimed at checking these results and investigating if, how and when patients interrupted their doctors, a corpus of 37 recordings made in a tertiary referral hospital in Belgrade, Serbia, in the department of pulmonology, has been analysed. Examples of interruptions by doctors and patients were analysed according to the principles of conversation analysis and critical discourse analysis. The obtained results confirmed an ever-present asymmetry in doctor-patient communication, although it was not as conspicuous as it had been stated in some previous research. Finally, the difference between the ways in which doctors and patients interrupt each other and the reasons behind these interruptions were emphasized.


Introduction
Doctor-patient communication is one of the most frequently researched sub-types of institutional talk.It is usually referred to as a medical encounter, which can be defined as an interaction between a doctor and a patient taking place in an examination room.Typically, the positions participants in institutional talk take follow the pattern of institutional hierarchy and the authority institutional representatives (i.e.judges, police officers, doctors, journalists, teachers, etc.) have over the clients of the Zb. rad. Filoz. fak. Splitu, 14 (2021), 133-150 Sinadinović, D.: DISCOURSE OF POWER FROM THE ASPECTS OF INTERRUPTION ... institution (i.e.witnesses/defendants, patients, interviewees, students/pupils, etc.), and these positions rely on specific restrictions and procedures (Fairclough 1989: 102;Drew & Heritage 1992: 22;Heritage 2005: 106, Johnston 2005: 119;Heritage & Clayman 2010: 34). 1 Due to these reasons, institutional talk differs from ordinary communication in many ways, from the turn-taking system 2 (Schegloff 1968(Schegloff : 1076;;Sacks 1995 vol.2: 523), the structure of interaction 3 and sequence organization, 4 to the choice of vocabulary 5 and asymmetry. 6 Being very strictly organized and following an established order, a medical encounter is often considered to be almost ritualized (ten Have 1991: 138;Ainsworth-Vaughn 2001: 454).It usually follows a particular model, the so-called biomedical model being the most widely accepted one.This is a six-phase model first introduced by linguists Byrne and Long (1976) and it contains the following steps: (1) establishing a relationship with a patient, (2) finding out the reason for the patient's visit, (3) conducting a physical and verbal examination, (4) considering the patient's condition, (5) discussing further treatment or suggesting further investigation, and (6) ending the conversation.This is an example of a clinician-centred model which has multiple variations that mostly follow the structure of an interview where doctors ask questions and patients are supposed to provide answers.The idea of developing a more patient-centred model resulted in the so-called biopsychosociological model 1 Participants who interact within institutional communication have specific goals that depend on their institution-relevant identities, they are expected to accept specific constraints when it comes to their communicative contributions and they respect frameworks and procedures that are specific for a particular institutional context. 2 For example, in institutional talk, it is not necessary for interlocutors to be equal, but it is quite possible for one of them to ask questions all the time and for the other to provide only answers.
Participants in this type of conversation can even be sanctioned for their contributions when they do not provide an answer to the question they have been asked or when they talk while someone else is talking (Heritage 2004: 165).3 Unlike everyday communication, institutional talk follows some strict rules and a predetermined order (Heritage 2005: 119-123).4 The use of the so-called third turn is rather specific in institutional talk as it is often used by representatives of institutions to influence their clients' behavior.A good example can be found in the situation of breaking bad news when the doctor prepares the patient for the news by making them share their own opinion of the problem and the outcome which changes the patient's perspective of the problem and makes it easier for them to accept even the worst-case scenario (Heritage 2005: 123-128).Another example is the so-called optimization principle where institutional representative chooses questions in such a way that a positive outcome is expected (Boyd & Heritage 2006).5 Compared to ordinary conversation, institutional talk is characterized by more formal lexis which includes descriptive terms, institutional euphemisms, passive and nominal constructions , etc. Another typical characteristic is the frequent use of the first person plural (we) instead of the first person singular (I) with the aim of presenting speakers as institutional representatives rather than individuals.6 Asymmetry is probably the most conspicuous characteristic of institutional talk and it can be noticed on various levels.For example, institutional representatives are believed to have more power than their clients as they keep the initiative throughout the interaction.Institutional knowhow is another example of asymmetry and it means that institutional representatives and their clients usually have completely different attitudes towards the problem becase of which the client has paid a visit to the institution (Heritage 2004: 175-179).Asymmetry of knowledge may be manifested by openly showing superiority in a particular field (institutional representatives) or by misunderstanding instructions or important advice due to a lack of knowledge in that particular field (institutional client) (Heritage 2004: 175-179;Frankel 1990).Finally, institutional clients are not supposed to have the same knowledge that institutional representatives have acquired owing to their education and position and even if they do posses some knowledge they are supposed to be silent about it (Strong 1979;Heritage 2004).
which contains 11 phases and insists on integrated interviewing which involves the patient to a much greater extent (Fortin et al. 2012) 7 .Research in this field covers various aspects of medical encounters and doctor-patient communication and one of its key conclusions is that doctors always have a greater amount of power than patients.
This paper deals with interruption as one of the most interesting aspects of doctorpatient communication.We will first briefly review some previous research in this field putting an emphasis on doctors and patients' habits and rights concerning interruptions.Our aim was to compare these results and conclusions with those that were obtained by analysing the corpus of 37 medical encounters that had been recorded in a tertiary referral hospital in Belgrade, Serbia, in the department of pulmonology. 8We were interested in inspecting how typical it is for patients to interrupt their doctors, how and why doctors and patients interrupt each other and whether they have equal rights when it comes to interrupting their interlocutor.
As expected, the results showed asymmetry and an unequal distribution of power between doctors and patients, although the differences were not as conspicuous as in some previous research.

Interruption in a medical encounter
Interruption is one of the features that unequivocally indicate the unequal position of participants in communication and the demonstration of power.When it comes to ordinary conversation, conversation analysts insist that there should be only one person speaking at a time and that turn-taking should be done with almost no delay and with a minimal overlap (Sacks et. al 1974: 700-701;Schegloff & Sacks 1984: 71-72;Sacks 1985 vol. 2: 222;Schegloff 2000: 2).According to this theory, there are three possible scenarios: the next speaker is chosen by the previous speaker, the next speaker is self-chosen, or the current speaker continues talking, thus becoming the next speaker; however, the position of the speaker cannot possibly be gained by interrupting the current speaker (Sacks et al. 1974: 703).
It should be noted that there is a difference between interruption and overlap.
According to West & Zimmerman (1977: 532), overlap refers to the situation when two speakers are speaking simultaneously, and it is initiated by the 'next speaker' before the 'previous speaker' has completed their statement but is remarkably close to the transition point.On the other hand, being several syllables away from the transition point is defined as interruption and is believed to violate the speaker's rights and prevent them from participating in conversation under equal conditions.Although both interruptions and overlaps can be found in doctor-patient communication, most studies have dealt with interruptions as they are reliable indicators of unequal encounters.Beckman & Frankel (1984) concluded that 7 Although the new model has been partly present in examination rooms in Serbia, it has been concluded that it is difficult for it to be fully integrated due to the lack of time and rather short medical encounters when compared to the medical encounters in English speaking countries (Sinadinović 2017).8 Recordings from this corpus belong to a much larger corpus in Serbian which was collected for the purpose of writing the doctoral thesis Linguistic Aspects of Doctor-patient Interaction in English and Serbian (Sinadinović 2017).The name of the institution where the medical encounters were recorded is not given to fully protect the privacy of all the participants in this research.
approximately 18 seconds passed between the doctor's question and the moment they decided to interrupt the patient's answer to ask another question or control the answer.According to Fairclough (2001: 38), there are several reasons for doctors interrupting their patients and they usually do not include the wish to do all the speaking: they do this to control their patient's contributions, to direct them, to prevent them from repeating themselves or sharing irrelevant information and to make sure they would elicit only relevant information.Mishler (1984: 100-103) explored the ways doctors interrupted their patients.For example, they may interrupt the patient simply by not paying attention to what they are saying (by deciding not to comment or not showing interest in the topic) and this is called interruption by inattention.This is a passive interruption that is often combined with introducing a new topic, in which case it is a more active form of interruption.Finally, active interruption is defined as giving a clear signal to the patient to stop talking about a particular subject, which is usually done by using the particle ok.Mishler noticed another, rather subtle, way of interruption -by not asking for clarification of a confusing statement or story, the doctor indirectly shows to the patient that they are not interested in what they were saying and that they are ready to change the subject.Irish & Hall (1995: 876) concluded that doctors tended to interrupt their patients by asking questions, whereas patients interrupted doctors by using statements.West (1984) explored interruption in doctor-patient communication from the aspect of gender.Having analysed 21 medical encounters recorded in a medical centre in the south of the USA, she found that doctors interrupted their patients much more often than the opposite, as well as that male doctors interrupted their patients more frequently than female doctors.Apart from interrupting their patients less frequently than their male colleagues, female doctors in this study were proven to have been more frequently interrupted by their patients (they were interrupted in 68% of cases whereas their male colleagues were interrupted in only 33% of cases).Moreover, female doctors were more frequently interrupted by their male patients than by the female ones.West explained these results by gender differences and the position of women in the society of the time (the beginning of 1980s).This author also noticed that male doctors most often interrupted their patients by asking them another question without waiting for them to complete their previous answer and that they tended to interrupt the patient's layman opinion by offering their expert opinion which showed their domination and control (West 1984: 95-96).Analysing a single, but rather long and very representative, medical encounter, Klikovac (2008: 47-48) also found several reasons for interrupting the interlocutor.
The doctor interrupted the patient to make him say something he wanted to hear; he sometimes interrupted because he simply followed his agenda and did not want to allow the patient to introduce a new topic; finally, he interrupted the patient when he wanted to close a particular subject or the entire encounter, which is similar to Mishler's active interruption.The patient also interrupted his doctor, but for completely different reasons -he did it to provide a negative answer to the doctor's question, to introduce a new topic or sub-topic (as he believed he had something important to say), to confront the doctor by claiming the opposite of what he said or to refuse the doctor's suggestion (Klikovac 2008: 47).It is important to mention that Klikovac (2008: 48) concluded that the patient did interrupt his doctor, but the doctor had the right not to allow to be interrupted, unlike the patient.She also noticed that the doctor interrupted his patient more frequently towards the end of the encounter, which relates to the phases of medical encounter 9 and their rules and tasks.On the other hand, ten Have (1991: 149) noticed that doctors most often refused to answer patients' questions and interrupted them more often in the first part of the encounter because they tried to collect all the relevant pieces of information, whereas they were more open to questions and interrupted their patients less frequently towards the end of the encounter.

Research
For this research, we analysed 37 recordings made in the pulmonology department of a tertiary referral hospital in Belgrade, Serbia. 10The recordings, with a total duration of 160'75'' (and an average length of 4'34'' per recording), were made with a digital voice recorder and then transcribed using a slightly adapted Jefferson Transcription System (Jefferson 1984(Jefferson , 2004). 11The transcribed material was analysed according to the rules and principles of conversation analysis and critical discourse analysis.Out of 37 patients who took part in this research, there were 24 male patients and 13 female patients, and they were all examined by one male doctor.There were seven first encounters (when the doctor and his patients met for the first time) while all the other encounters were follow-ups.
It is important to say that all the participants were aware of the researcher's presence in the examination room, and they knew the encounters were being recorded.However, they did not know what was researched and what they were expected to do, so we believe that they mostly behaved naturally.The participants had been informed there was no right or wrong behaviour and that the recordings would be used for scientific purposes in the field of linguistics.In all individual medical encounters, participants could decide whether they would allow the researcher to be present during the medical encounter and we would like to mention that none of them expressed dissatisfaction, refused to take part in the research or were reluctant about the participation.We analysed how and why doctors interrupted their patients and how and to what purpose patients interrupted their doctor.Relying on the principles of critical discourse analysis, we also investigated the differences between the doctor and his patients concerning the reasons for interrupting their interlocutor and the amount of power they showed in the process.Finally, we were interested if there were any significant gender-related differences, both in the total number of interruptions and in the reasons behind those interruptions.Out of the total of 303 interruptions, the doctor made 179 (59%), whereas his 9 At the very beginning of the encounter, doctors should listen to their patients carefully in order to find out why they came.However, when they start asking questions or giving explanations, they tend to interrupt their patients more often and they refuse to be interrupted themselves (Klikovac 2008: 48). 10 All medical encounters from the corpus were recorded in May 2013.11 The transcription system used in this research is given at the end of the paper.patients interrupted him 124 times, including the 16 cases where an accompanying person, usually a family member, interrupted the doctor instead of the patients themselves (41%).Approximately, the doctor interrupted his patients 4.8 times per encounter, whether patients interrupted the doctor 3.4 times per conversation.Both the doctor and his patients interrupted each other in 16 different ways.Some of these ways were typical for both the doctor and his patients, but there were also some specific ways for interrupting an interlocutor that were characteristic either for the doctor or for his patients.

How and why the doctor interrupted his patients
In this research, the doctor most frequently (36 times) interrupted his patients by asking a new question (Example 1) or by encouraging them to say more about their problem (20 times) (Example 2).He often interrupted his patients by agreeing with them or showing understanding (17 times) (Example 3), by explaining/giving an expert opinion (15 times) as well as by directing the patient's response (15 times) (Example 4).In 12 cases the doctor actively interrupted the patient (Example 5).The doctor also interrupted his patients by reassuring or calming them down (Example 6), which happened in eight cases.Patients were interrupted seven times by having their statements finished (Example 7).In five cases the doctor interrupted his patient by giving a response before hearing the entire question (Example 8).Apart from this, the doctor interrupted his patients by seeking an explanation (12 cases), by picking up where he left off, without paying attention to patients' efforts to ask a question or make a comment (eight cases), by giving instructions to the patient (eight cases) and by contradicting the patients (seven cases).
Example 1 D: [...] Mm-hmm.Have you been elsewhere?P: Well: there were some bacteriology and immunological tests// D: //Have you done that?P: Yes [...] 12 In this example the doctor interrupts his patient by asking her a new question.It is obviously a part of the encounter which is dominated by the doctor as he is trying to find out as many details as possible about the problem the patient is complaining about.The doctor starts the sequence by asking a yes/no question, but the patient does not reply as predicted.Slightly hesitating, she tries to explain that she has done something about her problem, but there is an impression that she is not quite sure what tests she is talking about and why they were performed.The doctor interrupts her by asking a new question (Have you done that?), once again expecting an affirmative or a negative answer and this time the patient responds properly, by providing an affirmative answer.Interrupting the patient in this way, the doctor shows her that he fully controls her contribution and that it is he who determines what and how much is said in each sequence, as well as that there should be no superfluous or vague expressions as every single detail can be important for making a diagnosis.Once again, the doctor does not interrupt the patient to emphasize his dominant position in the conversation, but to show that he agrees with what the patient is saying and give him support.The doctor's yes, better is related to something that the doctor supposes the patient wanted to say -that it has become much better/ easier to regulate blood clotting factors since the patient reduced the amount of green vegetables in his diet.Using such a formulation, he lets the patient know that he agrees with his opinion even though he was not able to finish his statement.The doctor continues his turn trying to encourage the patient by comparing the evidently good current results with some previous results that were unsatisfactory.
Example 4 P: [...]Especially whenever I have a cold, some mucus and so on, I have a sort of pressure in the lungs, difficulties breathing, I even have some, how should I put it, manifestations on the right: I don't know how to describe it, if it's pain, as if (.) something is going on// D: //Mm-hmm.Some indeterminate problem, isn't it?P: Well, I don't know how to define them, but discomfort and: like there is some pain:// D: //And this all happens when you have a cold and when: P: It is especially pronounced when I have a cold [...] 15   This example illustrates the opening part of the encounter in which the patient is describing her problem.The doctor lets her speak and he reacts only when he notices she has difficulties in formulating what she wants to say -she hesitates (using the fillers how should I put it, I don't know how to describe it), she supposes/ guesses (if it's pain, as if... something's going on), she uses a pause.This is when the doctor interrupts her, first using the particle mm-hmm with which he shows that he is listening and he understands what she wants to say, he supports her efforts to say that and then he directs her by offering her the right formulation in the form of a tag question (Some indeterminate problem, isn't it?).The patient still cannot define what she is complaining of but continues to describe the symptoms.The doctor interrupts her once again, in the same way, but this time the patient accepts his directions and provides the answer the doctor wanted to get.
Example 5 P: Now I have three inhalers (.) Berodual /.../ I was even prescribed infusion, about seven, and I could stand no more// D: //Yes, yes (..) Ok, here it is.You will continue taking your medications just like you did 16   According to Mishler's (1984: 100-103) classification, this is an example of active interruption.This is obviously the closing part of the encounter where the doctor gives directions and prescribes the therapy.The patient is trying to introduce a new topic or even tell a story, but the doctor interrupts him immediately.He does that by using the affirmative particle yes, twice.Its role is to show the patient that the doctor has heard him, but that it is time to close the topic and move on.Then there is an ok, just like in Mishler's examples, which announces the end of the phase and eventually the end of the encounter.This ok sounds final and the proof for this is the fact that the doctor moves on to giving directions to the patient.This is another example when the doctor does not interrupt the patient to take control over the conversation, but to reassure the patient.In his first turn, the doctor informs the patient that technically his performance on the (spirometry) test was not good, so the patient suggests repeating it.The doctor interrupts the patient's statement by claiming it is not necessary to do so.He emphasizes this by using the particle no twice and by denying it is necessary to repeat the test, also twice.By doing this the doctor calms his patient and tells him the result is not that bad and that these are probably some smaller irregularities which will not have a huge effect on the diagnosis.//It's just that your colds last longer P: ((coughing)) Here it is (...) 18   In this example the doctor interrupts his patient by finishing the statement she has started.By doing this he probably wants to show that he knows what it is all about, but he could also send a totally different message -that he is impatient and that he thinks she is taking too much time.However, it seems the patient took no offence, and even demonstrated the type of cough she had a problem with.
Example 8 P: [...]Good.Because my GP asked me if I could use those other inhalers (.) Berodual and // D: / / O f course.You take Berodual when needed and this round tablet and the new one you should take REGULARLY 19  The patient opens the sequence by trying to ask a question.Instead of asking the doctor directly if he could use other inhalers he has, he decides to use an indirect question and he emphasizes the fact that it is not he who is asking, but his GP.Due to such a formulation, we could conclude the patient is reluctant or he even feels sorry for asking a question, that he feels embarrassed.As it is a clumsily asked question, the doctor interrupts it as soon as he's heard enough and he answers the question which has not even been asked completely.Such an interruption can be interpreted in two ways -the doctor may want to help the patient by letting him know he does not have to finish the question and he will still get an answer, or he might want to send the message that it takes the patient too long to ask the question and that he should hurry up.After providing the answer to the question, the doctor adds one more piece of information concerning the way the prescribed medications should be used.
All the ways the doctor used to interrupt his patients are presented in Table 1.

How patients interrupted their doctor
The patients from this corpus most often interrupted their doctor to show they agree with what he was saying (27 times) and they did it with the help of the minimal responses of all right, okay, yes, that's true, that's right (Example 9).In 11 cases they interrupted the doctor by expressing their own opinion (Example 10), whereas in 10 cases they did it by introducing (or trying to introduce) a new topic (Example 11).
Giving a negative answer to the doctor's question was another reason to interrupt the doctor and this was the case nine times (Example 12).Patients also interrupted by communicating a piece of information (Example 13) which was the case on eight occasions.The doctor was interrupted six times by asking for clarification (Example 14).
Patients sometimes interrupted their doctor by making small talk (Example 15) which happened in three cases.Apart from this, in 19 cases patients interrupted their doctor to show they understood what he was saying/explaining and they mostly used minimal responses (e.g.yes, all right, mh-mmh, okay) to that purpose.They also interrupted their doctor by telling a story (eight cases), by giving an explanation about what they were doing concerning the problem they had (seven cases).by picking up where they left off (six cases), by giving a response before hearing the question (four cases), by asking a question (three cases), by finishing the doctor's statement (three cases), by contradicting the doctor (two cases) or by demanding something from him (two cases).
Example 9: D: Well, it makes sense then, if they think it goes together with those respiratory symptoms and: if they think it is allergy, how shall I put it, with some pulmonary consequences that produce that allergy cough↑ then you can get back to us to do some more tests, after the allergologist// P: //Mm-mhm.Okay. 20 The doctor starts this turn by commenting on some previous tests and some conclusions about the patient's diagnosis.The patient interrupts him using two minimal responses -mm-mhm and okay.In this way, she wants to show that she agrees with what the doctor is saying, maybe even to show that she has had the same or similar idea.
Example 10: D: All right, and do you have problems because of your primary disease, do you have difficulties breathing:// P: //I think that: my condition is much better than in previous: [...] 21 The doctor opens this sequence with a semi-open question and then he starts directing the patient towards the desired response.The patient does not interrupt the doctor to provide an answer to the question, but to express his own opinion.He slightly hesitates (which can be seen through the long vowels) and makes an introduction (I think) to show it is his own subjective opinion.With this interruption, he answers the doctor's question in a way, by stating he has no problems, no difficulties breathing and other complaints the doctor probably wanted to mention before he was interrupted.
Example 11: P: And can it be because of the thyroid gland// 20 L: Pa onda ima smisla, ako oni smatraju da to ide sa tom simptomatologijom plućnom a: ako misle da je to/ta alergija, da kažem, sa nekim posledicama na pluća u smislu da je taj kašalj na alergijskoj osnovi↑ onda može da se vratite kod nas da mi odradimo još neka naša ispitivanja, nakon alergologa// P: //Aha.Dobro 21 L: Dobro, a da li vi imate neke smetnje od strane te osnovne bolesti, da li imate otežano disanje:// P: //ja mislim da: mi je stanje mnogo bolje u odnosu na prethodne: D: //Yes, yes, it can be psychogenic↑ So (.) you were saying// P: //For example, I also have a problem with my stomach.I had that bacterium:// D: //Well that could also cause problems, because of digestion difficulties 22   In this example, the patient interrupts the doctor to introduce a new topic.The patient opens the sequence by asking a precise yes/no question (And can it be because of the thyroid gland).The doctor interrupts him by providing an affirmative answer, before even hearing the question and adding another possibility that did not cross the patient's mind (it can be psychogenic).Then he reminds the patient of something he has said before, announcing a new topic.However, the patient immediately interrupts his doctor by saying that he has stomach problems (which obviously is not his primary complaint) and in this way he also tries to introduce a new topic.Besides, he offers a story about the origin of those problems (I had that bacterium).The doctor does not ignore this attempt, but he accepts the patient's initiative, agrees with him and offers him a medical explanation for his problems.
Example 12: D: [...]And you haven't had any other pulmonary problem?That involved diagnosing, using medications// P: //I haven't (.)No 23   In this example the patient interrupts the doctor as he hurries to provide a negative answer without letting the doctor finish his statement/question.He even gives a double negative answer (I haven't and No), divided with a short pause, probably wishing to emphasize that what doctor is interested in has not happened to him and that he can move on to the next question.
Example 13: PD=patient's daughter, accompanying her for the check-up D: […] Yes.Then you will do the following.One to two days you will take a quarter and then// PD: / / S h e shouldn't take any today↑ That's what he told us D: No, no […] 24 In this example, the patient's companion (her daughter) interrupts the doctor by providing him with a piece of information.At the beginning of the sequence, the doctor gives some precise instructions to the patient, which is emphasized by the phrase Then you will do the following and this choice does not leave too much space for changes.However, the patient's daughter interrupts the doctor by telling him what the previous doctor has suggested; she provides the doctor with an important piece of information which could somewhat change his treatment plan.Nevertheless, the doctor resolutely refuses that option, and he continues explaining why his idea is better (it is not completely contained in the example).
Example 14: D: Here you are.See you in two months, take your therapy regularly// P: //So, like I did before?D: Exactly, and there won't be any problems. 25  Here the patient interrupts the doctor to ask for clarification about the instructions he got from him.Although the doctor obviously has not completed his statement and there is a possibility that he would give some more instructions, the patient interrupts him as soon as he has mentioned therapy because he wants to be sure he understood everything correctly.The patient does this by using a statement with an interrogative intonation and it seems he already knows what he is supposed to do but just wants to check by asking for clarification.The doctor gives an affirmative answer and encourages the patient by saying there won't be any problems.//Well, ok, you have an advantage at the very beginning ((laughing)) 26   In this example the patient (an elderly woman) interrupts the doctor with the aim of making a joke and continuing small talk that the doctor himself started.In his first turn the doctor addresses his patient in a joking way, obviously replying one of her previous remarks.She then interrupts the doctor to make a joke herself (I've decided to live to be 105).The doctor interrupts her again with another humorous remark followed with a loud pause (laughter).It is clear that the doctor and the patient have known each other for some time, as well as that the patient is an elderly lady, but the doctor is sure she would not be offended if he makes a joke about her age.
All the ways patients from this corpus used to interrupt their doctor are presented in Table 2.

Gender-related differences
When it comes to gender-related differences, we have got interesting results.
Although male patients made a larger number of interruptions (56 interruptions or 45% of all cases) than women patients (who interrupted the doctor 52 times or in 42% of cases), female patients interrupted the doctor much more frequently than male patients -while female patients made four interruptions per encounter, male patients made only 2.3 interruptions per conversation.We did not count companions (a daughter and a son) who made interruptions in 13% of cases.
It is also interesting to mention that the doctor interrupted his male patients and his female patients an equal number of times -99.However, there are 24 male patients and only 13 female patients, which means that male patients were interrupted 4.1 times per conversation, whereas female patients were interrupted as many as 7.6 times per conversation.

Conclusion
As a widely researched sub-type of institutional talk, doctor-patient communication is explored from various aspects, and interruption is one of the most frequently studied characteristics of this type of communication.Researchers take it as a good example of the discourse of power and different roles that institutional representatives and institutional clients inevitably take.However, research results vary, from those that give only minimal or no rights at all to institutional clients to those that show clients can express their opinion, ask questions, or interrupt institutional representatives.
For the purpose of this research, 37 recordings made in pulmonology department of a tertiary referral hospital in Belgrade, Serbia were analysed.There were 303 interruptions in total, out of which 179 (59%) were made by the doctor and 124 (41%) by his patients.In other words, the doctor made 4.8 interruptions per encounter, while his patients interrupted him 3.4 times per conversation.So, there is an expected asymmetry between the doctor and his patients, but it is not as conspicuous as some early research (Mishler 1984) claimed -according to our results, patients do interrupt their doctors and they even do so rather often, but still not as frequently as their doctor.
The doctor used 16 different ways to interrupt his patients, asking a new question, encouraging the patient, agreeing with the patient and directing the patient's response being the most frequently used ones.By looking into these ways and the frequency of their use, we concluded that the doctor demonstrated power and showed his dominant position in 54% of cases (he directed his patients' responses, actively interrupted them, asked new questions without waiting for the patient to finish, completed his patients' statement, etc.).On the other hand, he showed support in 25% of cases, when he agreed with his patients, encouraged them to say more about the problem or reassured them.Most of the ways the doctor used to interrupt his patients and the reasons behind the interruptions were also noted in previous research (Mishler 1984: 100-103;Irish & Hall 1995: 876;Fairclough 2001: 38;Klikovac 2008: 47-48).However, the doctor from our research also interrupted in order to make a joke or make small talk, as well as to give instructions to his patients.
Interestingly enough, patients in this research also used 16 different ways to interrupt their doctor.By analysing these ways and how often each of them was used, we can come to a conclusion that patients more often interrupted their doctor to show they followed the doctor's agenda, understood what he was saying and agreed with what he said -they did this in 44% of cases.However, contrary to some early research in this field (Mishler 1984), patients also interrupted their doctor in the way that clearly included showing initiative (e.g.introducing a new topic, asking a question, picking up where they left off without paying attention to the doctor's question, asking for clarification, expressing their own opinion or demanding something) and this happened in 32% of cases.They even made small talk and told a story in almost 9% of cases.So, patients do not have a completely passive role in a medical encounter, a medical encounter is not a typical interview with predetermined roles, but interruptions occur on both sides.Contrary to what Irish & Hall (1995: 876) concluded, patients in our research did not only use statements when interrupting their doctors, but they opted for questions as well.Finally, by inspecting the interruptions made by male and female patients, we noticed that female patients interrupted the doctor significantly more often than male patients -while female patients made 4 interruptions per encounter, male patients Sinadinović, D.: DISCOURSE OF POWER FROM THE ASPECTS OF INTERRUPTION ... Zb. rad. Filoz. fak. Splitu, 14 (2021), 133-150 did so only 2.3 times per encounter.So, the results we obtained in this domain differ from the conclusions of some previous research, including West's (1984) research in which a difference was made between male and female doctors and where female doctors were marked as those who interrupted less often but were interrupted more than their male colleagues.In our research there was only one (male) doctor, so we could not make that sort of comparison, but we believe that the results we obtained with female patients clearly show things have changed when compared to the early 1980s.

/ repair
Ključne riječi: analiza konverzacije, diskurs moći, institucionalna komunikacija, komunikacija liječnika i pacijenta, kritička analiza dikursa, medicinski susret, prekidanje  Zb. rad. Filoz. fak. Splitu, 14 (2021), 133-150 afternoon (.) Please, have a seat (...)How are you, /…/?P: To tell you the truth, I feel much better than before// D: //Okay: 13In comparison with Example 1, the doctor chooses a completely different way of interrupting his patient.While in Example 1 he interrupted his patient with the intention to keep her contribution and an entire sequence under control, in this example the doctor interrupts the patient to encourage him to continue talking about his problem and provide him with as many details as possible.This excerpt belongs to the opening part of the encounter when patients are supposed to state the reason for visiting their doctor, in their own words.They exchange greetings and the doctor tries to relax his patient by asking an open-ended question.When the patient comes close to the transition point, the doctor interrupts him by using the particle okay which in this case means I hear you, I follow what you're saying, I want to hear more.Example 3 P: [...]Since/Since I stopped taking these (.)this greenstuff and this// D: //Yes, better (.) Actually, it's regulated better.Well, great, today it is/it is /.../ and you know it sometimes was [...] 14 ...]And do you have any problems now?P: No (.) Nothing.I neither have a pain: nor do I have difficulty breathing:// D: …] You know what, it's a job half-done when you say you feel FINE.That is// P: //I've decided to live to be 120.I've said I'd do it out of spite// D: Sinadinović, D.: DISCOURSE OF POWER FROM THE ASPECTS OF INTERRUPTION ...

Table 1 .
The ways the doctor interrupted his patients