See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/14062353

Conversation analysis: A new model of research

in doctor-patient communication
ARTICLE in JOURNAL OF THE ROYAL SOCIETY OF MEDICINE MAY 1997
Impact Factor: 2.12 Source: PubMed

CITATIONS

READS

14

45

1 AUTHOR:
Anssi Perkyl
University of Helsinki
57 PUBLICATIONS 621 CITATIONS
SEE PROFILE

Available from: Anssi Perkyl

Retrieved on: 11 November 2015

JOURNAL OF THE ROYAL SOCIETY OF MEDICINE

Volume 90

April 1997

Conversation analysis: a new model of research in

doctor-patient communication
A Perakyla PhD
FORUM FOR MEDICAL COMMUNICATION, 16 MAY 1995

J R Soc Med 1997;90:205-208

In his much cited 1983 review, David Pendletont strongly

criticized existing research on the process of medical
consultation. What he meant by process research was
research that focuses on what happens during the
consultation. Pendleton argued that consultation process
descriptions have been 'like the listing of ingredients in a
cake without the analysis which shows how to put the
ingredients together'. The research that has appeared since
Pendleton's critical review indicates that the difficulties are
far from overcome. Most of the analyses of doctors' and
patients' activities in consultation have applied 'aggregation
techniques'2, which operate by coding and counting the
frequency of a small number of behaviours such as
'information giving', 'social conversation', 'positive talk'
or 'negative talk'3. This approach has drawn attention to
various types of activity within the consultation and their
possible association with the outcome, for example patient
satisfaction. What it seriously lacks, however, is exactly the
kind of insight that Pendleton called for 13 years ago an
analysis of 'consultation process in terms of social
interaction'1 and an ensuing understanding of the
consultation as a sequentially organized event.
Around the time that Pendleton published his critical
review, conversation analytical studies of a new kind began
to appear. Reports from Heath4'5, Frankel6'7 and West8'9
raised the hope that conversation analysis could meet
exactly the challenge spelled out by Pendleton. (Refs 10-12
offer examples of later developments of conversation
analytical studies on medical interaction.) In this paper, I
will demonstrate the possibilities that conversation analysis
opens up for research on doctor-patient communication.
CONVERSATION ANALYSIS AS A WAY OF
DESCRIBING THE SEQUENTIAL ORGANIZATION
OF INTERACTION
Through a detailed qualitative examination of tape
recordings and transcripts of real-life interaction, conversation analysis (CA) reveals how day-to-day human
activities are composed and organizedl4 17. The aim of
such studies of interaction between client and professional
is to describe the particular ways in which both parties
University of Helsinki, Department of Sociology, PO Box 18, 00014 University of
Helsinki, Finland

accomplish their tasks through the sequential organization of

interaction1 8-2. In medical consultations, central tasks

include elicitation of the patient's complaint, the history,
the physical examination, diagnosis and advice (how these
are delivered and received), and information about further
examinations22'23. All of them can be subjected to a
detailed analysis. To demonstrate how conversation
analysis can shed light on accomplishment of the
physician's and the patient's tasks I present results from
a study on medical consultations in Finnish primary health
care. A central medical task that we have focused upon is
the delivery of diagnostic information2+26.
THE DELIVERY OF DIAGNOSTIC INFORMATION

In our corpus of 65 videotaped primary health care

consultations, we have identified two different ways in
which doctors tell patients their diagnosis. Here I discuss
the two types of utterance, focusing on the ways in which
communication of the diagnosis involves a balancing act
between authority and intersubjectivity i.e. a balance
between orientation to the doctor as a possessor of
exclusive knowledge and orientation to the patient as an
informed participant.
All the extracts in this paper are presented with a
simplified transcription system (see Box 1). In the first
example, the diagnosis is delivered through what we call a
'straight factual assertion'.
Box 1 Transcription symbols

The pulse
Dr: pulse [can be felt

I
P:

Underlining means emphasis

Square brackets mean overlapping
talk

[Thank you.

(0.2)

Numbers in parentheses indicate

silences measured in tenths of
seconds
((The doctor is examining Descriptions of non-verbal events are
the X-ray))
given in double parentheses
Single arrows mark diagnostic
statements
Double arrows mark reports of
observations

205

JOURNAL OF THE

ROYAL

SOCIETY OF MEDICINE

Extract 1
1
2
3
4

Dr: >
>

(The doctor looks into the patient's ear)

There's still an infection
in the auditory canal. I'll prescribe
kind of drops for it

The doctor makes a statement in the classical format 'X

is Y'. The utterance offers the diagnosis as a direct
description of reality.
In the other type of diagnostic utterance, doctors
describe specific observations and frame them as evidence for
their diagnostic statements the 'evidence formulating
pattern', of which extract 2 offers an example:
Extract 2
(The doctor has just examined the patient's foot)
1 Dr:
2

3 Dr: =>

Okay. Fine do put on your,

(0.2)
The pulse [can be felt

I
5 Dr: =>
6
->
7
->

[Thank you.
there in your foot so,
there's no, in any case no real
circulation proble[m

8 P:
9

[Yes I don't understand then

[really I was wondering whether] I should

4 P:

10 Dr: ->

I
[is involved.

The main difference between extracts 1 and 2 is that, in

the latter, the doctor describes some of his observations
(the pulse can be felt) before the delivery of the diagnostic
statement and frames his observations as reasons for, or
evidence of, the diagnostic conclusion. Thus, the doctor
treats the patient as an understanding recipient of medical
reasoning.

By contrast, when employing the 'straight factual assertion'

format, doctors indicate that the diagnosis is unproblematic and
must be taken for granted27; they do nothing to make the
medical reasoning transparent to the patient.
Thus, straight factual assertions seem to involve a strong
orientation to the doctor's authority, whereas the evidenceformulating pattern seems to work towards establishing an
intersubjective understanding of some aspects of the
diagnostic process. However, a closer sequential analysis
will show that there are elements of intersubjectivity and
authority in both types of diagnostic utterance.

206

LOCAL CONTEXT OF THE DIAGNOSTIC

STATEMENTS
'Straight factual assertions' are regularly used when the
reasons or grounds for the diagnosis are obvious either on
physical examination or from some medical document such
as an X-ray.

Volume 90

April 1 997

In extract 3 the patient has hurt his finger when lifting a

stone in a lake. Our extract is from the first follow-up visit
after an X-ray was taken of the finger. The patient has
brought the X-ray to the consultation. Before the diagnostic
statement, the doctor examines it against the illuminated
screen while the patient volunteers his own reflections on
the circumstances of the accident.
Extract 3
(6.2)

2
3
4
5

Dr:
P:
Dr:

(Dr switches off the illuminated screen and

returns to his seat. He holds the X-ray picture
in his hand in front of him.)
Luckily the bone is quite intact,
Yeah,
So within a week it should get better
with that splint.

In line 2, when producing the diagnostic statement

'luckily the bone is quite intact', the doctor holds the X-ray
in his hand, so that the picture is between himself and the
patient. The diagnosis can be heard as a description of the
X-ray. Thus, the evidence of the diagnostic conclusion the Xray picture is prominently present in the activity context.
The grounds of the diagnostic conclusion are, literally,
visible to both participants. Equally importantly, however,
the patient avoids any active use of the evidence: when the
doctor examines the X-ray pictures against the illuminated
screen the patient merely glances at the picture. In this way,
the patient acknowledges the evidential source of the
diagnosis but does not presume to examine or interpret that
evidence.
The physical presence of evidence either from
medical documents or from physical examination is a
feature in all our cases of 'straight factual assertions'.
Therefore, the doctors' use of such assertions in the
delivery of the diagnosis does not involve a claim of
unconditional authority at the cost of intersubjectivity. The
doctors create a context where authority and intersubjectivity are blended.
In what context, then, is the evidence-formulating
pattern used for delivery of the diagnosis? Two issues seem
to be concerned here. One of them is the problematic
stature of the diagnosis, as involving a choice between
serious and non-serious possibilities. Doctors may display
evidence in a context where they rule out more serious
diagnostic possibilities, which the patients might have
indicated that they were worried about (example 2).
Secondly, and perhaps more importantly, use of the
evidence-formulating patterns is related to the nontransparent nature of the evidence that is, to the
inferential distance between the events seen and undergone
by the patient on one hand, and the diagnostic conclusion
on the other. In extract 3, illustrating straight factual
assertion, the evidence was in a single X-ray picture. For

JOURNAL OF THE ROYAL SOCIETY OF MEDICINE

members of our culture, it is obvious that an X-ray shows

whether a bone is broken or not.
By contrast, the patient in extract 2 has complained
about an intense pain in her foot-pain that went away on
its own but still worried her. During his examination
(looking at and palpating the foot), the doctor has not
reported any findings or observations. After the examination, the doctor tells the patient that 'no real circulation
problem is involved'. By formulating the evidence 'the
pulse can be felt' the doctor demonstrates to the patient not
only something he was doing (i.e. checking the pulse) but
also that there was a positive finding that can be used to rule
out a potentially worrying disease.
In sum, the evidence formulation bridges the potential
inferential gap between the events during the examination
and the diagnostic conclusion. Without this bridging device,
it could have remained unclear for the patient which, if any,
events or observations in the examination support the
doctor's diagnostic choice.
PATIENTS' RESPONSES TO DIAGNOSTIC
STATEMENTS
When diagnoses are delivered by straight factual assertions
the patient's response is typically slight26. In extract 3 we
see an example: after the diagnostic statement the patient
produces a minimal 'Yeah'. Through this token response he
acknowledges the diagnosis and the doctor's authority in

diagnostic reasoning.
Diagnoses that employ the evidence-formulating pattern
often elicit more elaborate responses, as in extract 4:
1 Dr: =>

As tapping on the vertebrae didn't cause any

pain

=>

3
->
->
4
5
6
7 P:
8 Dr:

and there aren't (yet) any actual reflection

symptoms
in your legs it corresponds with a muscle
complication so it's only whether
you have been exposed to a draught
or has it otherwise
Oh yes,
got irrita[ted

9 P:
10

11

April 1 997

'experiential' nature: she describes a bodily sensation to

which she and she only has access. Thus, the patient's
evidence is of a 'subjective' character, whereas the doctor's
is 'objective' (lines 1-3).
In sum, the evidence-formulating pattern differs from
the straight factual assertion in making relevant the patient's
comment on the diagnosis; but, even in this further
discussion, the doctor's authority is preserved.
CONCLUSION

In this paper I have tried to show how conversation analysis

offers a rigorous method of determining how the
'ingredients of the cake'1 come together in medical
consultations. Thereby, it becomes possible to point out
different ways of accomplishing given medical tasks. I have
looked at two ways of delivering the diagnosis and shown
how each has its typical environment of occurrence.
Nevertheless, there is always an element of choice: each
doctor is free to decide whether or not, and in which ways,
to describe the evidence for the diagnosis. The choice that is
made has implications for the patient's response. Conversation analysis can point out and clarify the parameters within
which these choices are made.

Acknowledgments The research was financially supported by the Academy of Finland and the Finnish
Foundation for Alcohol Studies. I gratefully acknowledge
the helpful comments of John Heritage, Doug Maynard,
Johanna Ruusuvuori, David Silverman and Marja-Leena

Sorjonen.
REFERENCES

Extract 4
2

Volume 90

[It couldn't be from somewhere inside

then as it is a burning feeling there so it couldn't
be in the kidneys or somewhere (that pain)

In her response to the doctor's diagnosis, the patient

both formulates her own diagnostic suggestions and offers
evidence to support them, thus treating herself as an
independent agent in the realm of medical reasoning.
However, through the use of a question format in her
diagnostic suggestions, she displays an expectation that it is
the doctor who will ultimately diagnose the trouble.
Moreover, the evidence she produces in line 10 is of an

1 Pendleton D. Doctor-patient communication: a review. In: Pendleton

D, Hasler J, eds. Doctor-Patient Communication. London: Academic
Press, 1983:5-56
2 Stiles WB, Putnam SM. Analysis of verbal and non-verbal behaviour in
doctor-patient encounters. In: Stewart M, Roter D, eds. Communicating
with Medical Patients. Newbury Park: Sage, 1989:211-22
3 Roter DL, Hall JA. Doctors Talking with Patients/Patients Talking with
Doctors. Improving Communication in Medical Visits. Westport, CT: Auburn
House, 1992
4 Heath C. The opening sequence in doctor-patient interaction. In:
Atkinson P, Heath C, eds. Medical Work. Realities and Routines.
Westmead: Gover, 1981:71-90
5 Heath C. Preserving the consultation: medical record cards and
professional conduct. Sociol Hlth Illness 1982;4:56-74
6 Frankel R. The laying on of hands: aspects of the organization of gaze,
touch and talk in a medical encounter. In: Fisher S, Todd AD, eds. The
Social Organization of Doctor-Patient Communication. Washington, D.C.:
Center for Applied Linguistics, 1983
7 Frankel R. From sentence to sequence: understanding the medical
encounters through microinteractional analysis. Discourse Processes
1985;7: 135-70
an analysis of queries and replies
8 West C. "Ask me no questions
in physician-patient dialogues. In: Fisher S, Todd AD, eds. The Social
Organization of Doctor-Patient Communication. Washington, D.C.: Center
for Applied Linguistics, 1983:75-106

207

JOURNAL OF THE ROYAL SOCIETY

OF MEDICINE

9 West C. Routine Complications: Troubles in Talk between Doctors and

Patients. Bloomington, IN: Indiana University Press, 1984
10 Heath C. Body Movement and Speech in Medical Interaction. Cambridge:
Cambridge University Press, 1986
11 ten Have P. Talk and institution: a reconsideration of the "asymmetry"
of doctor-patient interaction. In: Boden D, Zimmerman DH, eds. Talk
and Social Structure. Studies in Ethnomethodology and Conversation Analysis.
Cambridge: Polity Press, 1991:138-63
12 Frankel R. Some answers about questions in clinical interviews. In:
Morris GH, Chenail RJ, eds. The Talk of the Clinic: Explorations in the
Analysis of Medical and Therapeutic Discourse. Hillsdale, NJ: Erlbaum,
1995:233-58
13 Pomerantz A, Ende J, Erickson F. Precepting conversations in a
general medical clinic. In: Morris GH, Chenail RJ, eds. The Talk of the
Clinic: Explorations in the Analysis of Medical and Therapeutic Discourse.
Hillsdale, NJ: Erlbaum, 1995:151-70
14 Sacks H. Introduction. In: Jefferson G, ed. Lectures on Conversation.
Oxford: Blackwell, 1992
15 SchegloffEA. Introduction. In: Sacks H, ed. Lectures on Conversation, vol 1.

Oxford: Blackwell, 1992:ix-lxii

16 Heritage J. Garfinkel and Ethnomethodology. London: Polity Press, 1984
17 Heritage J. Conversation analysis: methodological aspects. In:
Quatshoff UM, ed. Aspects of Oral Communication. Berlin: Walter De
Gruyter: 391-418
18 Heritage J, Greatbatch D. On the institutional character of institutional
talk: the case of news interviews. In: Boden D, Zimmerman DH, eds.
Talk and Social Structure. Studies in Ethnomethodology and Conversation
Analysis. Cambridge: Polity Press, 1991:93-137

208

Volume 90

April 1997

19 Zimmerman DH, Boden D. Structure-inaction: an introduction. In:

Boden D, Zimmerman DH, eds. Talk and Social Structure. Studies in
Ethnomethodology and Conversation Analysis. Cambridge: Polity Press,
1991 :3-2 1
20 Drew P, Heritage J. Introduction: analyzing talk at work. In: Drew P,
Heritage J, eds. Talk at Work: Interaction in Institutional Settings.
Cambridge: Cambridge University Press, 1992:3-65
21 Silverman D. Discourses of Counselling: HIV Counselling as Social
Interaction. London: Sage, 1997
22 Byrne PS, Long BEL. Doctors Talking to Patients: A Study of the
Verbal Behaviours of Doctors in the Consultation. London: HMSO,
1976
23 ten Have P. The consultation as a genre. In: Torode B, ed. Text and
Talk as Social Practice. Dordrecht: Foris. 1989:115-135
24 Maynard DW. The perspective-display series and the delivery and
receipt of diagnostic news. In: Boden D, Zimmerman DH, eds. Talk
and Social Structure. Studies in Ethnomethodology and Conversation Analysis.
Cambridge: Polity Press, 1991:164-94
25 Maynard DW. On clinicians co-implicating recipients' perspective in
the delivery of diagnostic news. In: Drew P, Heritage J, eds. Talk at
Work. Interaction in Institutional Settings. Cambridge: Cambridge
University Press, 1992:331-58
26 Heath C. The delivery and reception of diagnosis in the general
practice consultation. In: Drew P, Heritage J, eds. Talk at Work.
Interaction in Institutional Settings. Cambridge: Cambridge University
Press, 1992:235-67
27 Pomerantz A. Giving a source or basis: the practice in conversation of
telling 'How I know'. J Pragmatics 1984;8:607-25