recommended patient centredness as the preferred style of doctor–patient communica-
tion as a means to improve patient outcomes (Neighbour 1987; Pendleton et al. 1984;
McWhinney 1995). Further, empirical research has explored both the extent to which
consultations can be deemed to be patient centred. For example, in one classic study
Tuckett et al. (1985) analysed recorded consultations and described the interaction
between doctor and patient and a ‘meeting between experts’. Research has also
addressed whether patient centredness is predictive of outcomes such as patient satisfac-
tion, compliance and patient health status (Henbest and Stewart 1990; Savage and
Armstrong 1990). Such research has raised questions concerning both the definition of
patient centredness and its assessment which has resulted in a range of methodological
approaches. For example, some studies have used coding frames such as the Stiles verbal
response mode system (Stiles 1978) or the Roter index (Roter et al. 1997) as a means to
code whether a particular doctor is behaving in a patient centred fashion. In contrast,
other studies have used interviews with patients and doctors (Henbest and Stewart
1990) whilst some have used behavioural checklists (Byrne and Long 1976). Complicat-
ing the matter further, research studies exploring the doctor patient interaction and the
literature proposing a particular form of interaction have used a wide range of different
but related terms such as shared decision making (Elwyn et al. 1999), patient participa-
tion (Guadagnoli and Ward 1998) and patient partnership (Coulter 1999). However,
although varying in their operationalization of patient centredness, in general the con-
struct is considered to consist of three central components; namely (i) a receptiveness by
the doctor to the patient’s opinions and expectations and an effort to see the illness
through the patient’s eyes; (ii) patient involvement in the decision making and planning
of treatment; and (iii) an attention to the affective content of the consultation in terms of
the emotions of both the patient and the doctor. This framework comparable to the six
interactive components described by Levenstein and colleagues (Levenstein et al. 1986)
and is apparent in the five key dimensions described by Mead and Bower (2000) in their
comprehensive review of the patient centred literature. Finally, it is explicitly described by
Winefield and colleagues in their work comparing the effectiveness of different measures
(Winefield et al. 1996). Patient centredness is now the way in which consultations are
supposed to be managed. It emphasizes negotiation between doctor and patient and
places the interaction between the two as central. In line with this approach, research
has explored the relationship between health professional and patient with an emphasis
not on either the health professional or the patient but on the interaction between the
two in the following ways: the level of agreement between health professional and
patient and the impact of this agreement on patient outcome.
Agreement between health professional and patient
If health professional–patient communication is seen as an interaction between two
individuals then it is important to understand the extent to which these two individuals
speak the same language, share the same beliefs and agree as to the desired content and
outcome of any consultation. This is of particular relevance to general practice con-
sultations where patient and health professional perspectives are most likely to coincide.
For example, Pendleton et al. (1984) argued that the central tasks of a general practice
DOCTOR–PATIENT COMMUNICATION 93