A study of primary care doctors in Singapore using case scenarios and patient-centred care statements through a self-administered questionnaire
Authors and affiliations
1- Yvette SL Tan FRACGP, MMed(FM), MBBS
2- Lee Gan Goh FRCGP, FCFP, MMed (int med), MBBS
3-Rob Whitley PhD National Healthcare Group Polyclinics, Singapore Department of Community, Occupational and Family Medicine, National University of Singapore Division of Social and Transcultural Psychiatry, Department of Psychiatry,McGill University Canada.
Dr Yvette S L Tan National Healthcare Group Polyclinics, Singapore Hougang Polyclinic, 89 Hougang Avenue 4. 538829Tel: +65 64898051; fax: +65 63863783; e-mail: firstname.lastname@example.org
of primary care doctors in
To assess patient-centred care (PCC) behaviour amongst doctors and explore reasons for non-patient centred behaviour .
The PCC index was 18 (PCC score range 12 – 24). Prominent reasons for a lack of patient-centredness were doctor-centred issues (53%), time-centred issues (31%) and skills issues (15%). For both common ground and holistic care themed scenarios, doctor-centred issues ranged from 46-88%. For proactive themed scenarios, time-centred issues ranged 60-65%.
Scale of patient-centeredness was determined predominantly by doctor-centred issues. However the nature of the determinants also appeared to be situation specific. Case scenarios appear to be useful for assessing the learning needs of doctors with regards to patient-centredness.
of primary care doctors in
Patient-centredness has been shown to improve outcomes of care, including patient’s and doctor’s satisfaction 1-9.
With recent emphasis on delivering quality care, patient-centredness has inevitably become one of the key features of such quality care.
Trying to teach patient-centredness in an attempt to improve care is now an important part of both undergraduate and postgraduate medical curriculum. It has become increasingly appreciated that there is a need to go beyond the teaching of basic communication skills to achieve a sustainable change towards a more patient-centred practice. Addressing the attitudes concerning the learning of such skills are thus important 12-17 .
Many studies have tried to measure patient-centredness using either a third party observation of a consultation or through doctor’s self-reporting of patient-centred knowledge 9,11,22. However, the doctor’s belief with regards to their level of patient-centredness and reasons underpinning this has received little attention. Doctors’ view on the determination of barriers to patient-centredness would enable those who teach patient-centredness to direct their resource in a more appropriate way .
This paper attempts to assess the contribution of doctor centred , time and skill factors as determinants of patient-centred care (PCC) delivery among doctors working in a cluster of nine polyclinics (primary care health centres) in Singapore. These clinics are grouped administratively as the ‘National Health Care Group Polycinics’ (NHGP) providing 60% of subsidized primary care and 10% of overall primary care in the country. In these clinics, doctors see an average of 60 patients per day, making the mean consultation time about 7.5minutes per patient.
We aimed to assess patient-centred care behaviour amongst doctors and explore reasons for non-patient centred behaviour.
Ninety general practitioners from all nine clinics in NHGP who were fully registered with the Singapore Medical Council, and actively practising at the time of the survey were invited to participate. This excluded doctors who were doing mainly administrative work and those who were on leave. Eighty-six participated.
We designed the questions based on six common case scenarios that attempted to elicit the doctors’ behaviour with regards to the various themes of patient-centredness. These were chosen from those described in current literature on PCC 3,7-9 , on the basis of being the most pertinent and applicable to the context of the primary care practised in NHGP as perceived by the authors. The three themes chosen were: finding common ground, holistic care and proactive care. Two case scenarios were dedicated to each of the PCC themes:
Each scenario question consisted of 2 parts. The first part asked how often the respondent’s own response to the given situation was similar to that suggested on a 4 point likert scale ( 4-always, 3-most often, 2-sometimes, 1-never), four being considered the desirable behaviour with regards to patient-centeredness. Anyone scoring less than 4 for the first part was directed to answer the second part of the question.
For example in the question on ‘request for referral”,
‘The father of a 5yr old boy comes demanding that you should refer him to the cardiologist because of some shortness of breath. You have made a thorough examination of the child and are convinced that it is most likely a psychological cause.
One response to a clinical scenario similar to this is to : suggest that it is probably a psychological cause and negotiate your way out of writing the referral ; while acknowledging the underlying concern of the parent. Also suggest that a referral may be considered if symptoms persist even after addressing the psychological issues.
The second part of the question explored the reasons why the respondent might choose to respond differently from what was considered desirable by asking the doctor’s agreement to three statements. The first statement addressed a doctor-centred attitude (e.g.” I should not compromise my professional judgement” in case scenario 1) . The second question addressed time as a factor ( e.g, “ I would write the referral anyway since it would probably take longer to explain why a referral is not needed” in case scenario 1) . The last question of ‘other reasons’ provided the respondents with a space for free comments. See tables 1A and 1B for the 6 case scenarios and questions asked.
The last part of the questionnaire explored on a 4 point likert scale, the doctors’ perception of global factors such as rating of his quality of care, the quality of his relationship with his patients and professional satisfaction.
The questionnaires were distributed via internal mail to all eligible doctors practising during the period 25th to 30th March 2002. Anonymity was ensured. The respondents returned the questionnaires into a sealed box that went round all the 9 clinics.
PCC score and PCC index: The PCC score is the summation of the responses from the six scenarios for each doctor. The PCC index was the median score of PCC of all the respondents.
Exploratory factor analysis was employed to study the psychometric property of the index. The underlying factors were extracted by the principal axis factoring method with Varimax rotation performed to simplify the factor structure. The cut-off for factor loading was predetermined to be 0.35.
Quality of care: Spearman correlation was carried out to ascertain the correlation between the index and doctors' perceived satisfaction, efficacy, quality of care and relationship. The data was analysed with Stata 6.0. All statistical tests were carried out at 5% level of significance.
Determinants of PCC: Qualitative analysis of the second part of the case scenarios exploring the reasons for non-patient centred behaviour was done with tabulation of factors into ‘doctor centred’, ‘time’ and ‘skills’ categories. (See figure 5). The doctor centred reasons given were then subcategorized according to their nature eg negative regard, inflexible professionalism, protection of the health system, defensive medicine, diminished personal accountability. Likewise, skills were further defined into negotiation, professional adequacy, quality of doctor-patient relationship and resourcefulness
Eighty-six of a total of ninety questionnaires were returned, making the response rate 96%.
The PCC score of all 86 respondents ranged from 12-24 with distribution of score skewed to the right. (see figure 1). The PCC index was 18.
Based on the scree plot, only 1 factor was extracted and was interpreted as a general factor in view of the loading patterns. There was no sufficient evidence to suggest that the index was formed by three underlying factors, namely common ground, proactive care and holistic approach. The PCC score was found to be positively correlated with doctors’ perceived quality of care provided (Spearman correlation 0.23 p-value 0.04). (See table 2).
There were a total of 434 responses to analyse . (See table 3). This included all who answered ‘always’ (4) and those who answered ‘most often’ or less (<=3) to the proposed response. 148(34%) of all the responses were scored 4 with ‘request for sleeping tablets’ case having the highest number, 41. There were 286 (66%) responses which scored <=3 , with the highest number, 68, from the ‘request for referral’ case.
Of the reasons for non patient- centred behaviour in the 6 case scenarios (See table 4), 55%(156) were doctor- centred issues, 33% (93) time-centred issues and 13% (37) skills issues.
For both common ground and holistic care themed scenarios, doctor-centred issues predominated. For proactive themed scenarios, time-centred issues predominated.
‘Treatment refusal’ had the most doctor-centred responses 88%(43). ‘Fragmented care’ had the most time-centred responses of 65%(20) respectively. ‘Request for referral’ had the highest skill-centred response of 29%(20).
‘Request for referral’, which had the lowest number of respondents scoring 4 (always), had 46%(31) doctor-centred , 25%(17) time-centred and 29%(20) skills-centred issues that resulted in non patient-centred behaviour.
Request for sleeping tablets which had the highest number of respondents scoring 4 (always), had 51%(22) doctor-centred, 14%(12) time-centred and 21%(9) skills centred issues that resulted in non patient-centred behaviour.
Further analysis within the various categories of responses revealed that the underlying features for a particular theme could be further defined . (See table 5)
Within doctor-centred issues, negative regard for patients predominated in the request for sick leave scenario, inflexible professionalism in request for referral, defensive medicine in request for sleeping tablets and finding someone else accountable in refusal of treatment and post discharge scenarios.
Within skills-centred issues, negotiation skills and professional inadequacy predominated in the request for referral whereas professional inadequacy alone predominated in the request for sleeping tablets.
The model of patient centred care described by Brown et al7 in 1995 to consist of 6 domains, namely, (1) Exploring illness experience and expectations, (2)understanding the whole person, (3) finding common ground, (4) health promotion, (5) enhancing the doctor-patient relationship and (6) being realistic.
A more recent review (2000) by Mead and Bower9 on the empirical literature of patient centredness proposed a five dimension conceptual framework namely the dimensions of biopsychosocial perspective; ‘patient-as-person'; sharing power and responsibility; therapeutic alliance; and ‘doctor-as-person'.
“Doctor as a person” in the Mead & Bower model emphasized the dimension of the doctor. His attitudes, personality, cultural background, transference and counter-transferance, are all hypothesised as having a significant impact on the doctor-patient relationship 9,15-21.
Of the six domains in the Brown et al model, we chose to examine patient centredness in the themes of finding common ground, the whole person (holistic care), and health promotion (proactive care) in the case scenarios that were scripted.
The results revealed that the doctors were mostly patient centred in attitude with a PCC index of 18. This was found to be significantly related with their perceived quality of care provided, suggesting that doctors associate quality care with patient- centredness.
The exploration of reasons for non patient- centredness revealed that the responses were predominantly doctor-centred attitudes, followed by time and skills issues.
Request for sick leave and treatment refusal resulted in the highest proportion of doctor centred responses. In such doctor- patient encounters, the doctor might often feel their own professionalism undermined resulting in negative feelings towards the patient. Responding with a negative regard or diminishing their own accountability would be ways to cope with such difficult situations.
Request for sleeping tablets triggered the most patient centred responses. In such a situation where the patient was seen as vulnerable, the response to want to be the patient’s advocate appeared to be natural for most doctors.
In the case of post hospital discharge and fragmented care, time seemed to be the limiting factor to providing more patient centred care .This suggested that with regards to proactive health matters, most doctors have the desirable attitude and skills needed, so that if enough time was given, they would be able to deliver the appropriate care.
With a relative short consultation time in NHGP, it was surprising to note that except for the two proactive themed case scenarios, time did not feature as the most important reason for being less patient-centred.
The majority of doctors in this study also seemed to struggle with the request for referral scenario with only 18%(15) PCC response of 4 . However, the determinants for non patient-centered behaviour appeared to be more evenly distributed among doctor- centred, skills and time factors, consistent with the fact that most doctors would consider this a tough situation .
Although communication skills has been the focus of many studies in the training of patient-centredness, it is surprising to note that in the 6 case scenarios, skills played the least important determinant. Of the cases that had relatively high skills-related issues, professional adequacy featured doubly more commonly than negotiation skills.
The results suggest there is a need to look beyond communication skills training when teaching patient-centredness. Recognizing underlying attitudes and modifying them is equally important for example when we want to try to modify the health seeking behaviour of patients who attend frequently for sick leave . Besides allowing for more consultation time, the need for attitudinal change related to doctor-centred factors must also be addressed to allow a more patient-centred approach which is desirable for the successful outcome of a consultation.
The doctor-centred attitudes that surfaced in this study such as negative regard, inflexible professionalism, concern about the health system, protecting one’s own legal interest, diminished personal accountability appear to be learned risk management tactics that protect the doctor from difficult consultations. Where have these attitudes come from? Can we postulate that these attitudes are part of the professional ego that has been acquired during the course of medical training? If so, have our senior doctors been less than ideal role models in the teaching patient centredness?7,23
There is probably a need for more training on PCC at all levels of seniority from medical students to professors. There is also a need to recognize current structural limitations on doctors’ practice eg time constraints, patient ‘s autonomy and disillusionment. Steps should be taken to overcome these.
It would have been better if the PCC score could be correlated with actual observation of patient-centeredness and other outcomes such as patient satisfaction, health utilization as in previous larger studies on PCC as this would have increased the construct validity of the scale.
We were also unable to demonstrate internal reliability of the questions as factor analysis did not support the presence of the 3 themes chosen; but only a single factor. This could possibly indicate the multifaceted nature of the Patient-centred care, which in reality may be difficult to capture in its individual components.
As with all self-reporting scales9 , there is also that social desirability bias may mask the real differences between doctors by encouraging particular responses . This we attempted to circumvent by the non threatening way the PCC statements were phrased and ensuring anonymity.
What we can say is that empirically, the scenarios appeared valid enough to extract levels of patient-centredness from the respondents; and their reasons for non patient-centred behaviour.
Most doctors in NHGP appear to have appropriate patient-centered clinical behaviour as reflected by the PCC index. The determinants that stood in the way of patient- centredness were mainly doctor-centred attitudes followed by time and skills and appeared to be situation specific. In the areas of finding common ground and holistic care, doctor centred attitudes were the main determinant for being less patient-centred than desired. In proactive care, time was the most important determinant.
The use of scenarios and statements appeared to be a useful practical tool for needs analysis in patient-centredness training.
|Click on these buttons to visit our journals|
All pages copyright ©Priory Lodge Education Ltd 1994-2004.
We thank :
All the doctors in NHGP who participated in this study.
Dr Shanta Emmanuel ( CEO, NHGP) , Dr Chong Phui Nah (chairman, NHGP research committee) and Dr Jason Cheah(chief of projects, NHG) for their support.
Ms Genediene GA Villanueva – Lim and Mr Siew Pang Chan from the Clinical Epidemiology Unit Tan Tock Seng Hospital, Singapore for the development of the questionnaire and statistical support.
Dr Elaine Tan and Dr Mah Tuck Cheong who gave comments on the 6 drafts of the questionnaire as it was taking shape as well as input on the final manuscript.
Dr Teoh Siew Har for her input during analysis.
Dr Hong Ching Ye for going through the first draft of the manuscript.
Miss Orchid Chua and Miss Xiang Lynn for helping to print the questionnaires and collecting them back from the respondents.
Competing interests : none.
Date of submission of manuscript: 02 12 2003, Accepted December 20the 2003
Word count of main text: 2451
Financial support : funded by National Healthcare Group Polyclinics Research Fund.
presentation at the 2nd Asia Pacific Forum on Quality Improvement
‘Patient-centred care : Exploring its determinants using case scenarios and statements’
Oral Presentation at WONCA
EUROPE 2003 ,
‘Patient-centred care : What determines the doctors’ clinical behaviour?’