To identify the communicative practices of doctors and patients, conversation analysis was used, a methodology that reveals the organising principles underlying interaction and, when applied to collections of texts, gives insight into regularities and patterns that are present during the conduct of different social activities.13 Conversation analysis is carried out through repeated audit of recordings, supported by detailed transcriptions of the interaction that is captured in them. Consequently, the researchers began by making careful transcriptions of all the consultations selected for in-depth analysis, using a simplified version of the Jefferson system which is widely adopted as an adjunct to conversation analysis,14 and these transcriptions were subsequently used to support analysis based on the recordings. Transcription conventions are shown in Box 1.
Box 1. Transcription conventions.
(The names of all participants have been changed)
D: | doctor |
P: | patient |
? | rising intonation |
. | a stop indicates the micro-interval between segments of speech |
(.) | a stop in round brackets indicates a pause of less than one second |
(2) | a number in round brackets indicates a pause timed to the nearest second |
:: | a colon indicates lengthening of a sound with additional colons indicating further lengthening |
no | underlining indicates a word spoken louder than those around it |
pres- | a dash immediately after an item indicates that the speaker has broken off before completing the utterance |
[ | square brackets on successive lines indicate the beginning of simultaneous |
[ | speech |
= = | equals signs indicate that there is no interval between adjacent utterances |
{mhm} | curly brackets enclose brief utterances made during another speaker's turn, which show acknowledgement of what is being said |
((sighs)) | text in round brackets indicates either contextual information or non-verbal vocalisations |
(?guess) | text in round brackets preceded by a question mark indicates an educated guess |
hh/ha | indicate units of laughter |
The next step was to analyse the turn-by-turn organisation of the interaction in consultations, paying close attention to practices that facilitated disclosure and discussion of patients' problems and those that inhibited them. While conducting the conversation analysis, the researchers remained attentive to the smallest observable elements of communicative practice, including pauses, silences, and simultaneous speech, as well as grammar, vocabulary, and intonation, making possible a level of comparison between face-to-face and telephone consultations that other methods do not afford. See Box 2 for a worked example of this method.
Box 2. Illustration of analytical method.
Transcription of consultation
Initiating the session
1 | P: | hello? |
2 | D: | hello . it's Dr Macintyre here from the health centre . I got a message to phone |
3 | | |
4 | P: | oh right . yeah . who . who's this? |
5 | D: | about Leslie Kirkness |
6 | P: | yes . that's right . aye |
7 | | (1) |
8 | D: | it's Dr Macintyre here |
9 | P: | right . e:rm . what it is is . I've just come back from holiday this morning |
10 | | |
11 | | (.) |
12 | D: | aha |
13 | P: | and . my ankle . from m- . well ((several words unclear)) for another reason is . is really swollen up |
14 | | |
15 | D: | aha |
16 | P: | and . I seem to have . the . the . the (?likes) o'mosquito bites and everything |
17 | | |
18 | D: | aha |
19 | P: | but it is really swollen and pretty painful |
Gathering information
20 | D: | and where were you on holiday? |
21 | P: | erm . we were in Greece . e- in . er . an island |
22 | D: | right |
23 | | (.) |
24 | D: | and it's just the one ankle that's swollen? |
25 | P: | yes |
26 | D: | right |
27 | P: | it's just when I got off the plane it . it really . it was throbbing a bit and . I noticed it was swollen |
28 | | |
29 | D: | aha |
30 | | (.) |
31 | P: | I thought I'd better just find out what's wrong wi'it |
32 | | (1) |
Explanation and planning
33 | D: | yeah . well . I think probably we want to have a look at that . erm |
34 | | ((inbreath)) right . if we (.) if I get you to come down at erm . |
35 | | eleven fifty? |
36 | | (.) |
37 | P: | yes . that sounds [alright |
38 | D: | [is that okay? . and er we'll have a look at you |
39 | | then . not giving you very long is it . I'm giving you five minutes . |
40 | | right . will you pop down in about five minutes |
41 | P: | yeah . I'll just come across then . [(?doctor who? |
42 | D: | [okay . that's great |
43 | P: | Dr Macintyre? . eh |
44 | D: | it's Dr Darwin |
45 | P: | Darwin . [right . okay |
46 | D: | [okay . right |
Closing the session
The first communicative task for doctor and patient is to find a means of initiating the medical discussion.17 In the transcribed example, after the interaction has been launched through an exchange of greetings (lines 1–2), the doctor (D) makes two statements that implicitly suggest that problem disclosure is expected (lines 2–3). One of these is his self-identification as a medical practitioner, the other a reference to the earlier call through which the patient (P) has shown that he wishes to speak to a doctor. However, the patient does not appear to understand that his call to the health centre is being returned (line 4) and there is a breakdown in communication which is revealed by the unusually long one second pause (line 7).18 It is likely that the interruption of the flow of talk arises because the doctor has expected the patient to continue his turn (line 6) with problem disclosure. A clue to this is the doctor's repetition of his self-identification as a medical practitioner (line 8), which this time results in the immediate disclosure of a concern by the patient. The doctor encourages the disclosure (lines 9–19) by using the type of acknowledgement token which indicates that he is listening and wishes to hear more (lines 12, 15 and 18).19 There is no evidence that the disclosure is curtailed, and the transition to problem discussion (line 20) only takes place when the patient has shown that disclosure is complete by repeating and intensifying his account of current symptoms (line 19).20
In face-to-face consultations, a doctor can gather additional information about patients' concerns by combining verbal and physical means,21 but over the telephone only verbal means are available. Here the doctor chooses to gather additional information by asking two questions (lines 20 and 24). The first of these is a request for new information about the circumstances in which the problem arose, and the second a restricted checking question designed to confirm details of the disclosure. The patient provides two pieces of factual information in response to the first question (line 21) and, in response to the second question, not only confirms the information embedded in the question but also goes on to expand his account of symptoms and the context in which they have been experienced (lines 27–28). When the patient finishes this turn the doctor gives him a further opportunity to develop his account again by using the acknowledgement token ‘aha’ (line 29), which encourages the current speaker to continue, and also by leaving the floor briefly open (line 30). The patient adds only one more point, a justification of his visit (line 31), and the floor is again left open for one second (line 32). The extent of this pause makes it clear that the patient has nothing further to add and that transition either to further questioning or the explanation and planning stage of the consultation is now a possibility. The doctor takes the latter course, following up on the problem by inviting the patient to attend a face-to-face consultation (lines 33–46) rather than attempting to make a diagnosis over the telephone. After mutual confirmation of the arrangements for this follow-up consultation (lines 45–46), there is an immediate transition to closing (line 47).
Only one concern is raised by the patient in this consultation and the doctor does not enquire about additional concerns after the first presenting problem has been dealt with. In some consultations the absence of such an enquiry could reduce safety but in this case there is evidence that it is appropriate. The use of the phrase ‘what it is is’ by the patient at the outset of the problem disclosure (line 9) indicates that he only wishes to raise one concern and, even if this were not the case, by arranging a follow-up face-to-face consultation the doctor provides an opportunity for the disclosure of additional concerns later.
The constant comparative method was used to identify both recurrent patterns of communicative practice and deviant cases, and regular team meetings were held, during which there was orderly and meticulous discussion leading to modification or verification of interim findings. In addition, when the analysis of transcriptions was complete, a random sample of consultations that had not been transcribed was re-audited, in order to check the validity of the study findings.15,16