Measuring patient-centredness: a comparison of three observation-based instruments

https://doi.org/10.1016/S0738-3991(99)00092-0Get rights and content

Abstract

The aim of this study was to assess the reliability and validity of three different observation-based measures of patient-centredness. The three face-valid instruments were each applied to the same sample of 55 videotaped GP consultations. Associations were explored with consultation ‘input’ variables (e.g. patient and doctor demographic characteristics, patient health status) and ‘process’ variables (e.g. consultation length). The three measures demonstrated varying levels of inter-rater reliability. Reliability was proportional to training requirements. Differences in construct validity of the three measures were evident and their concurrent validity was relatively low. Researchers must exercise caution in their choice of measurement method because of differences in how the concept of ‘patient-centredness’ is operationalised. Greater conceptual specificity and simplification are required for meaningful, reliable measurement. The implications for research, and for assessing the quality of individual doctors’ ‘interpersonal’ care are discussed.

Introduction

Central to current UK reforms of the National Health Service is the issue of ‘quality of care’ [1]. The advent of concepts such as ‘clinical governance’ (where primary care professionals are directly responsible for the quality of care delivered by newly established Primary Care Groups) highlights the need for valid and reliable measures of ‘quality’ [2]. While undoubtedly important, measures of clinical performance alone do not capture the complexities of providing high quality general practice care. The doctor–patient relationship, the concept of patient-as-person and the intrinsic therapeutic nature of the consultation itself have been core themes in the professional development of general practice [3]. Measures of the quality of interpersonal aspects of care are therefore also required as adjuncts to those of clinical performance.

The concept of ‘patient-centredness’ is increasingly regarded as a proxy for high quality interpersonal care. However, while strong moral and clinical justifications have been advanced [4], [5], there is a lack of consistent empirical evidence that patient-centredness is associated with improved outcomes. In part, this lack of consistency may be due to the fact that patient-centredness has been defined and measured in various ways: for example, as a professional attitude [5], [6], a set of knowledge [7], and in terms of consultation behaviours [8], [9]. Even studies adopting the latter behaviourist approach have utilised a wide variety of different methods for measuring the patient-centredness of medical consultations.

Only one study to date has explored the concurrent validity of different measures. Winefield et al. [10] conceptualise the patient-centred consultation as one in which the doctor is (a) receptive to the patient’s opinions and expectations, (b) involves the patient in decision-making about treatment, and (c) is attentive to cues of the affective relationship with the patient. Using transcripts of 210 audiotaped primary care consultations, the authors found a low correlation between two different measures of patient-centredness (one a ‘global’ scale for rating doctors’ patient-centred behaviour, and the second, a more intensive method based on coding doctor and patient verbal behaviours). Furthermore, while psychosocial or ‘complex’ consultations scored high in patient-centredness on both measures, there was less consistency between the measures in respect of their associations with patient age, acquaintance with the doctor, consultation length or patient and doctor satisfaction.

These findings suggest that studies employing different methods of quantifying patient-centredness may not actually be measuring the same domain of care. Such methodological inconsistency confounds cross-study comparison and hampers development of a body of evidence concerning the determinants and outcomes of patient-centredness. Building on the work of Winefield et al. [10], the aims of the present study were to (1) assess the inter-rater reliability and concurrent validity of three different observation-based measures of patient-centredness, (2) compare construct validity of the measures through associations with consultation variables commonly measured in other empirical studies, and (3) discuss feasibility of the measures for use in research and quality assessment.

Section snippets

Study sample

The study sample initially consisted of 72 videotaped consultations with adult patients involving 24 volunteer GPs working from 13 UK practices (in Birmingham, Manchester and Exeter). The sample was later reduced to 55 consultations after reliability problems were identified in respect of one of the patient-centredness measures (see Results section for further details).

The videotaping of consultations was undertaken as part of an on-going, European-wide study of doctor–patient communication

Analyses

The intra-class correlation coefficient (R), a measure of actual agreement rather than association, was used to examine the inter-rater reliability of each patient-centredness measure [20]. Associations with consultation input and process variables were examined using Spearman correlations (r).

Inter-rater reliability

Inter-rater reliability was calculated for both the Euro-communication rating scale and the Roter-based measure using data from four observers (A, B, C and D) who, prior to this study, had each coded the same sample of twenty videotaped consultations (the ‘reliability sample’). None of the 20 reliability sample consultations was among the 72 selected for inclusion in this study.

Reliability analyses showed that the internal consistency of the five-dimension Euro-communication rating scale was

Reliability

The reliability of any observation-based measure is potentially confounded by a number of factors: bias in the sample, the experience and training of observers, and the degree of subjective judgement required when applying the measure.

The GPs who volunteered for this study were not representative of UK general practitioners: 19 (79%) were GP vocational trainers or taught general practice medicine in universities. It is therefore likely that the communication skills of this group of GPs are

Implications

The utility of a measure of patient-centredness for research and quality assessment depends on its validity, reliability and feasibility, and a trade-off between the three criteria is often necessary. While no universally agreed definition exists, there is broad consensus that patient-centredness involves the doctor being open and responsive to the concerns and needs of patients (including needs for information and participation in decision-making). Concerns and needs vary between individuals

Acknowledgements

Data were collected as part of the ‘Euro-communication’ study, which is being co-ordinated by the Netherlands Institute for Primary Healthcare (NIVEL), with funding provided by BIOMED 2 and Pfizer UK. Thanks are due to all the GPs and patients who participated in the UK-arm of ‘Euro-communication’ and to Catherine Burns, Catriona Buckley and Emma Polyanszky who helped with the videotape coding. This work was undertaken as part of the research programme of the National Primary Care Research and

References (28)

  • M. Lipkin et al.

    The medical interview: a core curriculum for residencies in internal medicine

    Ann Int Med

    (1984)
  • P. Byrne et al.

    Doctors talking to patients

    (1976)
  • H. Winefield et al.

    The search for valid and reliable measures of patient-centredness

    Psychol Health

    (1996)
  • T. Coleman et al.

    Comparison of video-recorded consultations with those in which patients’ consent is withheld

    Br J Gen Pract

    (1998)
  • Cited by (186)

    View all citing articles on Scopus
    View full text