International Journal for Quality in Health Care Advance Access originally published online on December 23, 2007
International Journal for Quality in Health Care 2008 20(2):105-114; doi:10.1093/intqhc/mzm071
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Chronically ill Australians' satisfaction with accessibility and patient-centredness
1 Centre for Primary Health Care and Equity, University of New South Wales, Sydney, New South Wales, Australia
2 Discipline of General Practice, University of Adelaide, Adelaide, South Australia, Australia
3 Faculty of Health Sciences, University of Adelaide, Adelaide, South Australia, Australia
Address reprint requests to: Upali Jayasinghe, Centre for Primary Health Care and Equity, University of New South Wales, Sydney, New South Wales, Australia. E-mail: upali.jay{at}unsw.edu.au; upalij{at}optusnet.com.au
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Objective. To evaluate the association of characteristics of patients and general practices with patient assessment of quality of care.
Design. Cross-sectional multi-practice study using the general practice assessment survey.
Settings. General practices in Australia.
Participants. Ninety-six general practices and 7505 chronic illness patients aged 
18 years.
Main outcome measures. Access of care and patient-centredness.
Results. Two factors were identified in factor analysis: ‘Access of care’ and ‘Patient-centredness’. Multilevel regression analysis showed significant associations between patients' assessments and patient and practice characteristics. Patients from smaller practices (one to three general practitioners) reported better access to care compared with larger practices. Patients from urban areas were more satisfied with patient-centredness than those from rural areas. Self-reported health status and age had a positive and home ownership, employment and education, and patients from non-English-speaking countries a negative relationship with both scores. Females were more satisfied with patient-centredness.
Conclusions. Patient assessments of quality of care and patient-centredness were strongly associated with practice and patient characteristics. This has important implications for interpreting assessments of the quality of primary care, and for policy and practice measures designed to improve this.
Keywords: general practice assessment survey, patient satisfaction with accessibility and patient-centredness, patient and practice characteristics, factor analysis, multilevel regression model, assessment of quality of primary care
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Chronically ill people have complex care needs, often requiring care from multiple providers over a long period of time and active engagement in their own care [1]. Patient assessments of care are increasingly seen as important in planning general practice services [2], especially for patients with chronic disease. Satisfied patients are more likely to follow treatment instructions and medical advice than less satisfied patients, probably because they are more likely to believe that treatment will be effective [3] and are more likely to maintain a continuing relationship with their practice [2]. However relatively few studies of patient assessment have focused on this group of patients in Australia, and none allows direct international comparisons [4].
There are two main dimensions of quality of care – access of care and patient-centredness. Accessibility can be defined as the opportunity or ease which consumers or communities are able to use services in proportion to their need [5]. Patient centred care is defined as a deliberate attempt to understand and flexibly respond to the patients' perspective – their concerns and their priorities as a whole person [6].
Patients' assessments may reflect characteristics of the practice. For example in the USA, UK and many other European countries, patients from smaller practices have reported better access to care [7–10] compared with large practices. Practices in urban areas have reported longer consultations than rural practices [11].
Differences in satisfaction across socio-demographic groups may reflect differences in needs and expectations, but may also be due to actual differences in the care received, which is not always suitable for the individual patient [12–16]. Females have reported having longer consultations [11] and better doctor-patient communication than men [16]. Older patients were more satisfied [3, 12, 14, 17, 18], placed greater emphasis on continuity of care [15], had longer consultations [11, 17] and reported better doctor-patient communication [16]. Patients speaking languages other than English reported poorer access of care [14] and longer waiting times [13], than their English speaking counterparts. Having fair to poor self reported health status was also associated with lower satisfaction with quality of care [17].
The study aims to examine variations in ‘Access of care’ and ‘Patient-centredness’ according to practice (size and geographical location) and patient characteristics (gender, age, self-reported health, home ownership, education, employment, marital status, country of birth and chronic illness) in Australian general practice.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Participants
This study was part of a larger study of general practices in Australia examining the association between organizational capacity and quality of care. It was conducted in 27 Divisions in five states and in the Australian Capital Territory between December 2003 and October 2004. The data on Division characteristics showed that 27 of 103 Divisions that agreed to participate tended to be larger and to have a lower population to general practitioner (GP) ratio than the Australian average [19]. One hundred practices were invited to participate in the study. In each practice, clinical management software was used to select a random sample of about 180 patients aged
18 years and currently being prescribed medication for type 2 diabetes, ischaemic heart disease/hypertension or moderate to severe asthma. A total of 12 544 patients attending 96 practices agreed to participate. Completed surveys were received from 7505 patients (a response rate of 60%). A priori sample size calculations on the General Practice Assessment Survey (GPAS) access score confirmed that after adjustment for clustering (our pilot patient data from the GPAS indicated intracluster correlation coefficients (ICC) of 0.17 for the GPAS Access factor) predicted that an average of 50 patients from each of 100 practices would have sufficient power (1 – β= 0.8 and 
= 0.05) to detect an effect size of 0.24 between male and female patient groups.
Ethics
Ethics approval for the study was obtained from the University of New South Wales Human Research Ethics Committee and University of Adelaide Human Research Ethics Committee. Both practice staff and patients provided full written informed consent.
Instrument
Patient satisfaction was assessed through the GPAS version 2. The GPAS is a multi-item self-report questionnaire which measures several dimensions relating to patients' assessment of general practice [20]. The psychometric properties of the GPAS have been evaluated [21] and the instrument was found to have excellent measurement properties. Bower et al. [21] used 21 items and reported three factors (‘Access of care’, ‘Patient-centredness’ and ‘Nursing’) [21, 22]. One of our main aims was to examine ‘Access of care’ and ‘Patient-centredness’. Previous research indicated that 18 items contributed to these two factors [21]. Furthermore, about half of our 96 practices did not have a nurse (the remaining factor). Therefore, we restricted our analysis to these 18 items.
Factor analysis
Factor analysis was conducted with 18 items using SPSS statistical software (version 14; SPSS, Chicago, IL, USA) with principal axis factor analysis and a number of factors was determined by the scree test and number of eigen values >1. The solution was rotated using the varimax rotation (Table 1). Only assessment items are used in the calculation of scale scores and assessment items are measured on 6 point scales (Table 1). Scale scores (ranging from 0 to 100) were computed in accord with the GPAS manual [20]. As suggested by Bower et al. [22] and also suggested by the results of the factor analysis of current Australian data and UK data [21] the two main dimensions of Access of care (access, receptionists and continuity scores) and Patient-centredness (communication, interpersonal care and knowledge of the patient scores) were calculated based on the sum of three component scale scores (ranging from 0 to 300) with a mean of 242.3 (SD = 43.3) and 241.0 (SD = 47.8) respectively (Fig. 1).
|
|
Further, we assessed the internal consistency (reliability) of the scales by investigating whether items within each subscale correlated significantly (Cronbach's
).
Multilevel models
Multilevel regression models were used with two GPAS dimensions (Access of care and Patient-centredness) as continuous dependent variables and practice and patient characteristics, as the independent variables. Multilevel analysis (with MLwiN Software [23]) adjusted for clustering of patients (level 1) within practices (level 2) [10, 11, 16, 18]. Initially, we fitted a baseline variance component model (no independent variables) for each of the response variables followed by the main model. The main model expands the baseline model by including patient and practice characteristics as fixed effects. The analysis of variance was conducted using SPSS (Table 2).
|
Data and variables
Because patients do not register with GPs in Australia, it is not possible to determine the ‘list size’ of practices accurately and thus number of GPs was used as a measure of practice size. Geographical area was defined by using the Rural, Remote and Metropolitan Area classification [24] as urban (capital cities and other metropolitan centres with populations > 100 000) and rural (large and small rural centres with populations of 10 000 to 99 999 and other rural centres with population less than 10 000). There were no remote area practices in the sample. The socio-demographic characteristics of respondents studied were gender, age, self-reported health status in the last 12 months, home ownership, education, employment, marital status and country of birth (Table 2). The chronic illness of the patient was also included in the analysis. The patient characteristics were collected using the GPAS questionnaire.
Patients were asked to indicate whether they had a chronic illness: 3264 indicated one chronic illness and 1514 more than one of the three chronic illnesses. Patients who did not indicate any chronic illness (n = 2727) were incorporated in the analysis as a separate category (‘unknown’ category).
Significance of parameters
Parameter estimates were tested by the t-value, determined by dividing the estimated coefficients by their standard errors (Tables 3 and 4) [23]. Because the two models were nested (for example, the baseline variance component model was nested within the main model because the latter was created by adding independent variables to the former), we used –2 log likelihood, known as the ‘change in the deviance’, which has a 
2 distribution to test whether the difference between the two models was statistically significant (Table 4).
|
|
Variance explained at each level
The variance explained was estimated using the baseline model and main model [18]. The baseline variance component model explained how the total variance was partitioned into variance between patients and practices (Table 4). Differences in the modelled variance indicate how much better a model can account for the variance at a specific level [18]. The formulas to calculate the proportion of variance are explained by Sixma et al. [18].
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
There were 7505 of the 12 544 consenting patients who returned the questionnaire (60% return rate). We conducted analyses comparing proportions of respondents with non-respondents for gender and age. Gender and age were available for 90 and 84% of non-respondents respectively. The gender of respondents and non-respondents were similar (P = 0.76). Twenty percent of non-respondents were younger than 40 years as against 10% of respondents (P < 0.001).
Factor analysis
Factor analysis suggested a two-factor solution which explained 57.2% of the variance. Table 1 shows the items and their loadings on each factor. The first factor was ‘Patient-centredness’ which included items from communication, interpersonal care and knowledge of the patient (Cronbach's = 0.95). The second factor was ‘Access of care’ which included items from access, receptionists' performance and continuity of care (Cronbach's = 0.86).
Table 2 shows the characteristics of the analysed sample and the results of the univariate analyses. The proportions of characteristics of the total sample and the analysed sample were identical. The mean age of respondents was 60 years (range 18–96).
After list-wise deletion of missing values of patient characteristics, 7004 (93%) cases for Access and 7052 (94%) cases for Patient-centredness from 96 practices were available for the multilevel analyses.
Table 3 shows the results of the multilevel regression analyses for each of the response variables.
Female respondents were more satisfied with Patient-centredness than males. Older chronically ill people were more satisfied with Access of care and Patient-centredness than younger patients. Patients with better self-reported health status rated both Access of care and Patient-centredness better than those with poor health (Table 3). Home ownership can be considered as one marker of economic status [25]. Home ownership was associated with a poorer assessment of Access of care and Patient-centredness. Patients with ischaemic heart disease/hypertension or asthma only had lower satisfaction with both Access and Patient-centredness than patients with two or more conditions. However, the satisfaction of patients with diabetes was not significantly different.
Well-educated patients tended to be less satisfied with Access to care and Patient-centredness than less well educated patients. Patients who were employed were less satisfied with Access of care and Patient-centredness than those who were not employed. Patients born in English-speaking countries (including Australia) tended to be more satisfied with Access of care and Patient-centredness than those born in non-English speaking countries.
Patients from smaller practices reported better accessibility of care compared with those from larger practices. Respondents were more satisfied with Patient-centredness in urban areas than in rural areas. Practice location had no association with Access of care.
Variance components
Ninety one percent of the total variance in Access of care was at the patient level, the remaining 9% variance at the practice level. For Patient-centredness the corresponding figures were 93% at patient and 7% at practice level. Access of care had the higher ICC with 0.09 (9% of the total variance is between practices) and ICC of Patient-centredness was 0.07 (Table 4). At the patient level (level 1) 8 and 3%, respectively of the variance among patients for Access of care and Patient-centredness were explained by the independent variables used in the analysis (Table 4). At the practice level (level 2), 27 and 7% of the variance among practices for Access of care and Patient-centredness were explained by the variables used in the analysis (Table 4).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Patients with chronic disease account for an increasing burden of disease and presentations in general practice in Australia [26]. This study provides the first comprehensive data on how they assess the access and quality of care provided to them by GPs.
Methodological limitations
While 103 Divisions from the five States and one Territory were approached, only 27 Divisions agreed to participate and there were no remote area practices in the sample. Practices that volunteered to participate may not be representative of all practices within these Divisions. However, the proportion of practices that were solo, or large (four or more) was similar to that reported in other studies [27]. Patients who were unable to read English were excluded from the study. Although the response rate of 60% was comparable with other studies [14], it is possible that some of those not responding may have had different views of their general practices from those who responded. For example, 20% of non-respondents were younger than 40 years compared with 10% of respondents. We adjusted for these differences in distribution between the total sample (14% from 18 to 39 age group) and respondents by giving greater weight to younger respondents. The results showed the difference between unadjusted and adjusted Access score (P = 0.21) and Patient-centredness scores (P = 0.42) were not significant. Finally there may have been other practice and patient factors important to patient assessments such as patient co-payments and availability which were not specifically measured in this study and warrant further exploration in the Australian context.
Factor analysis requires complete data on all items, but one item (‘Thinking of times when you are to see any doctor: how quickly do you usually get seen? How do you rate this?’) was answered as ‘not applicable’ or missing for 1317 (17.5%) and some of the other items also had small number of missing values, leaving 5734 complete records for factor analysis. Most (81.7–98.0%) patients rated all aspects (items) of Access of care and Patient-centredness between good and excellent, with the exception of waiting time (62.8%) and the time taken to see a particular doctor (76.0%) (Table 1). The item ‘How often do you leave your doctor's surgery with unanswered questions?’ had the lowest loadings among patient-centredness items (Table 1). Further, this item had the lowest correlation with other items and 98% expressed satisfaction with this item. If this item was deleted Cronbach's would marginally increase where as if any other item was deleted Cronbach's would decrease. However, we think it may be better to explore this item further before simply deleting it.
Patient assessment of access
Chronically ill patients were likely to assess the practice as more accessible if the practice was small. Studies in the USA, the UK and many other European countries have reported similar results [7–10]. This may reflect a preference for personal service and continuity of care [2] which may be more difficult to achieve in a large practice [10].
Patients were likely to rate access to the practice more favourably if they were older and had better self reported health status. This is consistent with other research [3, 9, 12, 14, 15, 17]. In the case of those with better health, this may reflect less frequent visits to general practice [2, 9], and therefore fewer chances to experience difficulty of access. Patients with a poor health status have higher expectations and value shorter waiting times [15]. The lower scores among higher socioeconomic status groups may be related to higher expectations [15] and those for patients from non-English-speaking countries to difficulty in accessing a GP with the same language or culture.
Patient assessment of patient-centredness
Practices in urban areas were likely to be rated as more Patient-centred. This may reflect an increased choice of provider (possibly including nursing or allied health providers) within a larger practice, and greater choice of practices in urban areas. This is the first study that we are aware of which shows better patient-centredness in urban practices.
As with Access, patients who were less than 40, employed, well-educated, had poor health, and were owner occupiers rated the Patient-centeredness of their practice lower. This may be because these patients have higher expectations (especially if the GP elects to charge above the Medicare fee which requires the patient to make a co-payment for the consultation from their own pockets as this is not covered by insurance or Medicare). Patients with a higher economic status are known to place greater emphasis on involvement in decisions and getting information [15] and apply more stringent standards in their evaluations of care [12].
Female patients rated Patient-centeredness more highly. This has been found in other research [16] and suggests that GPs may tend to be more effective communicating and recognizing signs of agreement with female patients [16].
The results of the univariate analysis (Table 2) were similar to those of the multivariate analysis (Table 3) for most of the characteristics. However, there were marked differences in satisfaction with home ownership, retirement and two or more conditions.
The univariate analysis (unadjusted scores) showed a statistically significant trend for increased satisfaction with retired in comparison to employed or unemployed and with two or more conditions in comparison to a single disease. Further, home ownership had no effect on satisfaction. The results were in contrast to those of multivariate analysis (adjusted scores). It is clear that the strong positive effect of older age on satisfaction was confounded in retired, two or more conditions and home ownership. The result shows the importance of adjustment for a range of possible confounding characteristics.
Policy and practice implications
An important implication is that while larger practices may offer greater choice of provider and services and thus respond to a wider range of patient needs, there is a need to make greater efforts to reduce waiting times and increase the ease with which patients can book with their ‘usual’ GPs. Conventional efforts to improve access tend to give greater emphasis to timeliness than to promoting continuity of provider, which is a key element in patient overall satisfaction and in the quality of care received [28].
Different chronically ill patients have different expectations, needs and perceptions and they should also receive different care. Practice location had no effect on Access of care. This highlights the role of expectations. Access may be very different in urban and rural areas, but patients may not expect the same access to care and so be equally satisfied. Patient expectations are clearly important and this may explain the lower scores for quality of care given by patients who are younger, who have higher socio-economic status or poor health, all of whom may have higher expectations of services. These groups are likely to be associated with less choice (in rural areas), lack of continuity of care and any co-payments they were required to make to see their GPs. These findings are broadly consistent with Anderson's model of access to health care, which includes the association with client characteristics, and enabling resources on use of health services [29]. Engaging consumers in their own care may be associated with these expectations and be important in efforts to improve patient assessment of their own care [30].
Lower satisfaction of ethnic minority groups and patients with poor self rated health is clearly a case of the ‘inverse care law’ in which those with the greater need receive lower quality care. It suggests that greater effort is required to support GPs and other practice staff working with poor health or culturally and linguistically diverse patients including improving access to training and to interpreters.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
This study was funded by the Australian Department of Health and Ageing. The Department of Health and Ageing was not involved in the study design, data collection, analysis or interpretation, and had no influence on the writing and submission of this article. The investigators would like to thank the participating general practices and their staff, and the participating Divisions of General Practice for their assistance in recruiting practices and assisting practices with the feedback provided to them. We would also like to thank other members of the PRACCAP study group, Jane Grimm, Edward Swan and Dr Chris Barton for their valuable contribution and Professor Nick Zwar for his comments. Our thanks also go to Sheryl Scharkie, Roy Batterham, Heidi DePaoli and Robyn Alexander for their assistance with data collection.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
- Wagner EH, Austin BT, Davis C, Hindmarsh M, Schaefer J, Bonomi A. Improving chronic illness care: translating evidence into action. Health Aff (2001) 20:64–78.
[Abstract/Free Full Text] - Baker R. Characteristics of practices, general practitioners and patients related to levels of patients' satisfaction with consultations. Br J Gen Pract (1996) 46:601–05.[Web of Science][Medline]
- Grogan S, Conner M, Norman P, Willits D, Porter I. Validation of a questionnaire measuring patient satisfaction with general practitioner services. Qual Health Care (2000) 9:210–15.
[Abstract/Free Full Text] - Cheek J, Oster C. A Qualitative Investigation of the Experiences, Perceptions and Understandings of People with a Chronic Condition (2002) Adelaide: Centre for Research into Nursing and Health Care, University of South Australia.
- Whitehead M. The Concepts and Principles of Equity and Health. (1990) Europe: World Health Organisation, Regional Office.
- Stewart M. Towards a global definition of patient centred care. BMJ (2001) 322:444–45.
[Free Full Text] - Rubin HR, Gandek B, Rogers WH, Kosinski M, McHorney CA, Ware JE. Patients' ratings of outpatient visits in different practice settings. JAMA (1993) 270:835–40.
[Abstract/Free Full Text] - Campbell JL, Ramsay J, Green J. Practice size: impact on consultation length, workload, and patient assessment of care. Br J Gen Pract (2001) 51:644–50.[Web of Science][Medline]
- Baker R, Streatfield J. What type of general practice do patients prefer? Exploration of practice characteristics influencing patient satisfaction. Br J Gen Pract (1995) 45:654–59.[Web of Science][Medline]
- Wensing M, Vedsted P, Kersnik J, Peersman W, Klingenberg A, Hearnshaw H, et al. Patient satisfaction with availability of general practice: an international comparison. Int J Qual Health Care (2002) 14:111–18.
[Abstract/Free Full Text] - Deveugele M, Derese A, van den Brink-Muinen A, Bensing J, De Maeseneer J. Consultation length in general practice: cross sectional study. BMJ (2002) 325:472–4.
[Abstract/Free Full Text] - Hall JA, Dornan MC. Patient Sociodemographic characteristcs as predictors of satisfaction with medical care: A Meta-Analysis. Soc Sci Med (1990) 30:811–18.[CrossRef][Web of Science][Medline]
- Shi L. Experience of primary care by racial and ethnic groups in the United States. Med Care (1999) 37:1068–77.[CrossRef][Web of Science][Medline]
- Campbell JL, Ramsay J, Green J. Age, gender, socioeconomic, and the ethnic differences in patients' assessments of primary health care. Qual Health Care (2001) 10:90–5.
[Abstract/Free Full Text] - Jung HP, Baerveldt C, Olesen F, Grol R, Wensing M. Patient characteristics as predictors of primary health care preferences: a systematic literature analysis. Health Expect (2003) 6:160–81.[CrossRef][Medline]
- van den Brink-Muinen A, Verhaak PFM, Bensing JM, Bahrs O, Deveugele M, Gask L, et al. Communication in general practice: differences between European countries. Fam Pract (2003) 20:478–85.
[Abstract/Free Full Text] - Barr DA. Race/ethnicity and patient satisfaction. using the approapriate method to test for perceived differences in care. J Gen Intern Med (2004) 19:937–43.[CrossRef][Web of Science][Medline]
- Sixma HJ, Spreeuwenberg PMM, van der Pasch MAA. Patient satisfaction with the general practitioner: a two-level analysis. Med Care (1998) 36:212–29.[CrossRef][Web of Science][Medline]
- Hordacre AL, Keane M, Kalucy E, Moretti C. Making the Connections. Report of the 2004–2005 Annual Survey of Divisions of General Practice (2006) Adelaide: Primary Health Care Research and Information Service, Department of General Practice, Flinders University, and Australian Government Department of Health and Ageing.
- National Primary Care Research and Development Centre. Manual. General Practice Assessment Survey (GPAS) Version 2 (2001) Manchester: University of Manchester.
- Bower P, Mead N, Roland M. What dimensions underlie patient responses to the general practice assessment survey? A factor analytic study. Fam Pract (2002) 19:489–95.
[Abstract/Free Full Text] - Bower P, Campbell S, Bojke C, Sibbald B. Team Structure, team climate and the quality of care in primary care: an observational study. Qual Saf Health Care (2003) 12:273–79.
[Abstract/Free Full Text] - Goldstein H, Rasbash J, Plewis I, et al. A user's guide to MLwiN. (1998) London: Institute of Education, University of London.
- Australian Department of Primary Industries and Energy. Rural, Remote and Metropolitan Areas (RRMA) Classification. (1994) Canberra: DPIE.
- Macintyre S, Ellaway A, Der G, Ford G, Hunt K. Do housing tenure and car access predict health because they are simply markers of income or self esteem? A Scottish study. J Epidemiol Community Health (1998) 52:657–64.[Abstract]
- Australian Institute of Health and Welfare. Chronic Diseases and Associated Risk Factors in Australia. (2001) Canberra: AIHW.
- Commonwealth Department of Health and Aged Care. General Practice in Australia. (2000) Canberra: DoHA.
- Murray M, Bodenheimer T, Rittenhouse D, Grumbach K. Improving timely access to primary care: case studies of the advanced access model. JAMA (2003) 289:1042–46.
[Abstract/Free Full Text] - Andersen RM. Revisiting the behavioural model and access to medical care: does it matter? J Health Soc Behav (1995) 36:1–10.[CrossRef][Web of Science][Medline]
- England SL, Evans J. Patients' choices and perceptions after an invitation to participate in treatment decisions. Soc Sci Med (1992) 34:1217–25.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||