Family Practice Advance Access originally published online on May 15, 2006
Family Practice 2006 23(4):453-460; doi:10.1093/fampra/cml020
The impact of standalone call centres and GP cooperatives on access to after hours GP care: a before and after study adjusted for secular trend
a Program Evaluation Unit, School of Population Health, University of Melbourne Victoria, Australia
b Epworth Hospital Bridge Road, Richmond, Victory, Australia 3121
Correspondence to David Dunt, Director, Program Evaluation Unit, School of Population Health, University of Melbourne, Level 4 207 Bouverie Street, Victoria 3010, Australia; Email: d.dunt{at}unimelb.edu.au
Received 27 January 2005; Accepted 3 April 2006.
 |
Abstract |
|---|
 Top Abstract BackgroundMethodologyResultsDiscussionConclusionAuthor contributionReferences |
|---|
Background. The After Hours Primary Medical Trials were initiated by the Australian government to redress difficulties in after hours (AH) GP care in areas of high need. The study's objective is to study the impact of two standalone call centres and one GP cooperative offering comprehensive services, in improving consumer access to services for residents of a defined geographic area.
Methods. A pre–post design was used to evaluate their impact after adjusting for secular trend at a national level. Access was considered in terms of availability, accessibility, affordability, acceptability and responsiveness of care. Unmet need and ease of obtaining AH telephone professional medical advice were also considered. Pre-trial and post-trial telephone surveys of two separate random samples of approximately 350 households using AH services in each trial area as well as in a national sample outside the trial areas.
Results. Consumer acceptability and affordability increased in residents in the area served by the GP cooperative. Access, however measured, did not improve in either of the standalone call centre areas. Reduction in unmet need approached but did not achieve statistical significance in most but not all trial areas.
Conclusions. Improvements in access in the GP cooperative conformed to expectations based on current and pre-existing AH care arrangements put in place. Absence of improvements in access in the standalone call centres did not conform to expectations but may be partly explained by the reductions in consumer acceptability, following introduction of telephone triage systems reported elsewhere.
Keywords. After hours care, access, GP, evaluation.
|
Background |
|---|
|
TopAbstractBackgroundMethodologyResultsDiscussionConclusionAuthor contributionReferences |
|---|
There has been concern expressed in recent years by governments, the medical profession and consumers about the delivery of general practitioner (GP) after hours (AH) medical care both in Australia and other countries. Traditionally, GPs provided this AH care to their own patients. However, for some decades there has been a decline in own-practice care and an increase, as a result, in deputising services arising from changes in society and the way general practice is organised and financed.
In the UK, one study conducted in 1994 indicated that over 75% of GPs had consented to use a deputising service, although not all did so.1 New initiatives to improve care, facilitating the establishment of GP cooperatives, were also mounted in the UK in 1995. These involved changes in legislation, permitting GPs to transfer care to other GP principals listed on the relevant Family Health Services Authority and government funding to meet start up and maintenance costs for the infrastructure of these cooperatives. A later initiative was the establishment of NHS Direct—a national call centre for the provision of medical advice over the telephone by nurses using clinical protocols. Following a recent review, NHS Direct will in fact become the point of first contact for people accessing AH services in the UK.
In 1992, Denmark initiated reforms aimed at transferring working hours and costs from out-of-hours to daytime.2 County-wide AH services replaced the deputising services, cooperatives and own-practice arrangements that were then in place. These services delivered GP telephone triage services (care provided by telephone alone), AH clinics, if necessary, and home visits only if other arrangements were inappropriate.
Alongside other initiatives in Australia, the Commonwealth Government funded the After Hours Primary Medical Care Trials (AHPMCTs). The goals of the AHPMCTs were to remove or reduce obstacles to the provision and use of effective AH care through the coordination of high quality services.3 Within these broad overall goals, each AHPMCT was able to pursue local goals and objectives and take up particular care arrangements that would allow them to best meet local difficulties in AH service provision.
For four of the five AHPMCTs, telephone triage was the only service innovation. Two of these four were standalone call centres; the other two were embedded in different forms of GP AH care arrangements. The fifth AHPMCT consisted of a GP cooperative offering comprehensive services. The five AHPMCTs represent two models of AH care (first, telephone triage and advice services; second, GP cooperatives) of the six models identified in a recent structured review of the effects of different models of AH primary medical care.4
This study aims to determine the impact of the AHPMCTs on consumer access (including affordability and acceptability amongst other parameters—see Table 1), which constitutes an important goal of the AHPMCTs and of AH care more generally. More precisely, it studies the impact on these service variables for three of the five AHPMCTs for which increase in consumer access to AH care was the primary goal. While consumer acceptability (patient satisfaction) has been studied to a certain extent, consumer access has been much less so.4 The study also aims to study consumer access at a whole of population or regional level. While whole of population studies are typically common in the study of health services, only a few of these types of studies have been conducted relating to AH GP services.2,5
|
Other papers will evaluate the effects of the AHPMCTs on other objectives such as the level of service utilisation and service mix.6
The three AHPMCTs included in this study were the two standalone call centres and the Multiservice GP cooperative, as follows.3
Stand-alone call centres: These two stand-alone call centres were staffed by nurses using proprietary health call centre software aimed at providing accessible advice and promoting more appropriate AH service use. The first was a Statewide call centre studied in metro and non-metro areas7 and the second was a Regional call centre studied in inner metropolitan and rural satellite areas.8,9 AH services that existed in these areas before the trials (GP own-patient care, deputising services, Emergency department (ED) and ambulance) continued during the trials. The call centres provided new service arrangements in addition to these pre-existing services.
Multiservice GP cooperative (GP cooperative model):10 Four types of services were offered—an AH GP clinic within the ED of the regional hospital, a telephone triage service, GP home visits and a transport service. Two-thirds of GPs in the area participated in the cooperative. Pre-trial, a deputising service provided AH services in the area. It ceased to operate during the trial period.
The AHPMCTs commenced operation in late 1999 to mid-2000 and were assessed over a 12-month period. The AH services together with the regional context of the trials and their organisational settings are summarised in Table 2.
|
|
Methodology |
|---|
|
TopAbstractBackgroundMethodologyResultsDiscussionConclusionAuthor contributionReferences |
|---|
Study design
The framework for the national evaluation was conceived as ‘multiple trials’ with common questions and hypotheses, rather than as a ‘multicentred’ trial with common protocols so as to constitute a community trial.3 This reflected the underlying goals of the AHPMCTs, which were to meet local circumstances and difficulties and to investigate different models and types of AH care that were appropriate to these. The individual trials were therefore studied separately and were not directly compared. Their relative success is discussed.
A pre–post design was used to detect changes in relevant variables in the trial area across the 12-month study period and to examine whether these were consistent with the trial objectives of the AHPMCTs. To take account of secular trend at a national level, results were adjusted for changes in the relevant variables across the study period in the rest of Australia outside of AHPMCTs' areas (National comparator). More local period effects, which could only be detected within the framework of a community trial will not therefore be able to be detected.
Data collection
The regional impacts were measured using telephone surveys of separate random samples of user households in trial areas and the National comparator at the beginning of the trial period (pre-trial—November–December 1999) and repeated twelve months later (post-trial). The surveys were conducted by a commercial market survey firm under the direction of the evaluation team. Random samples of all households (with publicly listed telephone numbers) within the designated trial areas were drawn from the latest version of the electronic white pages, with each selected household receiving a letter explaining the survey prior to contact by telephone. Once contact had been established with a selected household, interviewers asked to speak to someone ‘aged 18 years or over who would know about the use of medical services’ in the household (the informant). If the informant agreed to participate, the status of the household as having used or not used any AH medical services (GP clinic or home visit, ED visit, ambulance usage) during the past year was established. Only the results for user households are reported in this study. All forms of GP usage, not just AHPMCT usage (where relevant) were thus considered.
Questions in the interview schedule covered a number of topics relating to AH care including opinions about community access (including acceptability and affordability) as well as questions relating to households' and informants' characteristics. Access was considered in terms of the five dimensions outlined by Penchansky and Thomas—see Table 1.11 Some access questions around these five dimensions were derived from McKinley et al in either original or modified form12 and other questions in similar format were added to ensure all five dimensions were covered. It was not possible to identify validated questions relating to two areas of AH service use. These were in regard to levels of perceived unmet need for AH services (i.e. whether a member of the household felt they needed AH medical care but had not sought it) and ease of access to professional medical (i.e. clinical) advice over the telephone. Questions regarding these were constructed for the study after being trialled during the pilot stage of the survey.
To better understand the extent of impact on levels of access, the volume of calls to the telephone triage system in each site (as well as clinic consultations, home visits and taxi trips in the Multiservice GP cooperative) is also presented.
Analysis
Sample size was estimated at 350 user households per trial area based on a hypothesised 15% change in dichotomised access measures (80% power, 0.05 level of significance). Thus, 350 users were interviewed in each trial area and the National comparator. Since response rates varied across these sites, the numbers of users approached varied in each site—see Table 3.
|
Ratings for each dimension of access were dichotomised (difficulties reported versus no difficulties reported). If a respondent rated access as difficult on one or more questions within a dimension, then the (dichotomised) dimension was rated as ‘difficulties reported’. The unmet need and professional telephone medical advice variables were similarly dichotomised. These dichotomised scores were considered as dependent variables for multiple logistic regression analysis.
Adjustment for secular trend was undertaken as follows. The principal independent variables in this analysis were: time (pre/post), intervention status (trial area/national comparator) and trial effect (time*intervention status). In addition, three further binary variables (gender, country of birth and Aboriginal and Torres Strait Islander status of the informant) were included as possible confounders. This followed comparison of 10 characteristics of study population at pre- and post-trial in each trial area (metro and non-metro separately where appropriate) to examine if changes had occurred across the trial period for any reason including as a result of trial impact—see Table 4.
|
Informant's financial status (financially disadvantaged—health care card and no private health insurance/others), the presence of children less than 12 years old in the household and the presence of a person with a chronic illness or disability requiring AH care in the household were also included in the regression because of their past association with access, and in the case of disability, with satisfaction.13
Multivariate logistic regression was conducted with SPSS for Windows version 12.0.1. Odds ratios, adjusted for the independent effects of other confounding variables were calculated. No adjustment was made for multiple comparisons as the study was hypothesis-generating not hypothesis-testing in nature.
Relevant contextual factors in interpreting results
It should be noted that the service in the metro area of the Statewide call centre had been operating for around five months before the pre-trial survey was conducted and important effects on access in the metro area during these early months may not have been detected. A GP AH walk-in clinic operated for part of the trial period of the Regional call centre in its central metro area. It was located adjacent to the ED of a regional hospital but closed due to lack of demand. In addition, as a startup service, the Regional call centre, experienced some delays and operational difficulties leading to lower call volumes.3,8,9 There were also problems with the previous deputising service operating in the area of the Multiservice GP cooperative.
|
Results |
|---|
|
TopAbstractBackgroundMethodologyResultsDiscussionConclusionAuthor contributionReferences |
|---|
Participation rates
Participation rates as a percentage of contacted households at pre- and post-trial are shown in Table 3.
Statewide call centre—see Tables 4 and 5 and Figure 1
The only significant difference in socio-demographic characteristics was a significantly lower proportion of informants born in Australia post-trial compared with pre-trial in the metro area—see Table 4. Approximately 50 AH calls/month/10,000 head of population were made in the metro area over the first 9 months of operation of the service. No call centre monitoring data is available for the non-metro area.3
|
|
After adjusting for secular trend, levels of difficulty in availability, accessibility, acceptability and responsiveness did not change across the trial period in both metro and non-metro regional populations. Level of difficulty in affordability significantly increased in the non-metro area though not in the metro area.
Unmet need though did not significantly decrease in either area across the trial period though it approached doing so in the metro but not the non-metro area. This may relate to the fact that significantly higher levels of unmet need (when compared to the National comparator) at baseline, only existed in the metro area. There was no change in level of difficulty getting professional medical advice over the telephone in either area.
Regional call centre—see Tables 4 and 5 and Figure 1
There were no significant differences in the socio-demographic characteristics of respondents between samples pre-trial and post-trial in both the central metro and satellite areas of the Regional call centre—Table 4. There were 18 and 6 AH calls/month/10 000 head of population in the metro and satellite areas respectively during the first 12 months of operation.3
The trial had no significant impact on levels of all five access dimensions in both central metro and satellite regional populations.
There were significantly higher levels of unmet need in both metro and non-metro areas at baseline compared to the National Comparator. Unmet need did not significantly decrease in either area during the trial period though it approached doing so in both metro and non-metro areas. There was no change in the level of difficulty getting professional medical advice over the telephone in either area.
Call centres (together)
When call centre areas were considered together (metro, non-metro and all), there was no significant change in levels in any of the access dimensions. Unmet need decreased significantly in metro call centre trial areas (95% CI 0.40–0.98) but not in non-metro call centre areas (95% CI 0.46–1.14). Level of difficulty in getting professional medical telephone advice also did not change when call centre areas were considered together (metro, non-metro and all).
Multiservice GP cooperative—see Tables 4 and 5 and Figure 1
The only significant differences in the sociodemographic characteristics of samples were that a significantly higher proportion of respondents were female, born in Australia and of Aboriginal and Torres Strait Islander background post- compared with pre-trial—see Table 4. There were 13 569 recorded contacts with the trial services during the first 12 months of operations (52 taxi trips, 11 315 clinic consultations, 2,093 telephone triage calls, 102 home visits). There were 21 AH phone calls/month/10 000 head of population.3
The trial had a significantly positive impact on affordability and acceptability in the regional population. Levels of difficulty for availability, accessibility and responsiveness decreased in the trial area across the study period—however, none of these changes were significant.
There was a significantly higher level of unmet need pre-trial compared to the National comparator. Unmet need did not significantly decrease during the trial period though it approached doing so. There was no change in level of difficulty getting professional medical advice over the telephone.
|
Discussion |
|---|
|
TopAbstractBackgroundMethodologyResultsDiscussionConclusionAuthor contributionReferences |
|---|
In appraising the study's findings, it should be noted that the National comparator group cannot be considered as a control group. The programme was based on a pre–post design, using the National comparator exclusively to take into account secular trend (i.e. to detect and control for period effects occurring at the national level). It was also based on self-reported rather than objective data on access and illness.
The effects reported here apply to a whole population of residents living in the trial area and their use of all AH services. In the case of the Multiservice Cooperative, this will include other GPs inside and outside the trial area. Magnitude of the effects of AHPMCTs that are estimated this way will be smaller than if their effects alone had been measured. The magnitude of effects will also be affected by the extent of take-up of the services—the number of callers using the telephone services. The telephone triage call rate in the three AHPMCTs varied—6 and 50 AH calls/month/10 000 population in the Regional call centre (non-metro area) and Statewide call centre (metro area), respectively. This uptake of telephone triage, with the exception of the Regional call centre (non-metro area) should have been sufficient prima facie to produce an effect on access levels.
Standalone call centres: Access, as measured across the five dimensions did not increase in either metro or non-metro populations of both standalone call centres. While the trials were well targeted (higher need at baseline in three of the four areas), reduction in unmet need approached, but did not achieve statistical significance in the metro and non-metro areas of the Standalone call centre and the metro area of the Regional call centre. (It achieved significance when call centres in both metro areas were considered together.) Getting professional medical (i.e. clinical) advice over the telephone did not become easier for users for both standalone call centres. In brief, while there were the suggestions of some beneficial effects (on unmet need) there was no evidence of increases in regional access to AH care (however measured) in the two standalone call centres. It should be noted that estimates of the magnitude of these effects may have been reduced as follows. First, the service in the metro area of the Statewide call centre had been operating for around five months before the pre-trial survey was conducted and important effects then may not have been detected. Second, as a startup service, the Regional call centre experienced some delays and operational difficulties leading to lower call volumes.3,8,9
As noted in the Introduction, only consumer acceptability (patient satisfaction) of the five dimensions of access has been studied to any extent in the research literature. Consequently, only the results for consumer acceptability (for standalone call centres, no increase) can be related to this literature.4 One of the most consistent results of the structured literature review was that telephone triage was associated with reduced patient satisfaction (consumer acceptability) where it replaces an in-person consultation.4 For example, following the Danish reforms described in the Introduction, delivering GP telephone triage services alongside other services, Christensen reported that satisfaction of patients in various counties using the AH service was significantly lower in 1992 and 1995 (though less so) than in 1992.2 Two other studies have also found that there was reduced satisfaction with telephone consultation.14,15 The study of callers of NHS Direct however found that the advice offered by nurses at the first wave NHS Direct sites was well received by most callers, achieving levels of satisfaction comparable with other telephone advice services.16 The study however was without controls.
Complaints in this research literature concerning telephone advice were greatest when the patient was expecting a home visit. However, this was not the context of telephone triage use in the AHPMCTs; so this should not have a strong effect. Nevertheless, dissatisfaction with telephone advice may explain in part why the standalone call centres did not increase consumer acceptability of AH care, as was expected.17 It is unclear whether this dissatisfaction relates to the advice being provided by telephonic advice and not face-to-face advice or the advice being provided by a nurse who is not associated with a doctor, or both.
GP Cooperative Model: Unlike call centre areas, access for two of the five dimensions (affordability and acceptability) increased during the trial period, (alongside reduction in unmet need in this area which approached, but did not achieve statistical significance). The increase in affordability is clearly linked to its free-to-the-public method of payment which did not exist in the deputising service that existed in the trial area pre-trial. The increase in acceptability seems most probably linked to the AH clinic service that was the most important and used of the multiple services offered by this AHPMCT, more so than the telephone triage service, for example. The clinic's success in turn related to the support of two-thirds of this rural community's GPs, its visible presence as part of the area's only hospital and its favourable comparison with the problems experienced with the previous deputising service.
These results indicate a higher level of satisfaction for GP cooperatives compared with deputising services than previously reported. A validated postal questionnaire survey of two overlapping services, a cooperative and a deputising service, found little difference in overall satisfaction.14 Another study using the same study design and same validated questionnaire compared patient satisfaction for a GP cooperative, a deputising service and practice-based AH care. It also found that overall levels of satisfaction did not vary in these three AH arrangements.15 The greater level of acceptability with the GP cooperative than the previous deputising services may relate to some extent to non-generalisable situational factors in the trial area.
While the trials were again well targeted (higher need at baseline), unmet need did not significantly decrease during the trial period (though it approached doing so).
It is noteworthy that no service reported any effects on three of the multidimensional measures of access—accessibility, availability and responsiveness nor telephone professional medical advice.
|
Conclusion |
|---|
|
TopAbstractBackgroundMethodologyResultsDiscussionConclusionAuthor contributionReferences |
|---|
The Multiservice GP cooperative model was more successful in achieving its consumer access objectives than the standalone call centre model. Consequently, results for the Multiservice GP cooperative model conformed more to expectations than the standalone call centre model. The absence of improvements in access in the latter may be partly explained by the reduced levels of AH telephone services consistently revealed in the literature. The study does not provide strong evidence for continuation in funding of standalone call centres for the purposes of increased access for users. They would need to make changes in their methods of operation such as better integration with mainstream services as has occurred in the UK, to achieve this. Alternatively, they would need to demonstrate other beneficial effects.
Competing interests
There are no competing interests.
|
Author contribution |
|---|
|
TopAbstractBackgroundMethodologyResultsDiscussionConclusionAuthor contributionReferences |
|---|
DD made a substantial contribution to conception and design, acquisition of data, analysis and interpretation of data and was involved in drafting the manuscript. SED made a substantial contribution to acquisition of data, analysis and interpretation of data and was involved in revising the manuscript critically for important intellectual content. MK made a substantial contribution to data analysis and was involved in revising the manuscript critically for important intellectual content. MM made a substantial contribution to conception and design and was involved in revising the manuscript critically for important intellectual content.
|
Acknowledgments |
|---|
The study was funded by the Commonwealth Department of Health and Ageing (DoHA). It owned intellectual property of the study but granted to The University of Melbourne a non-exclusive, royalty-free licence to use the study for academic purposes such as publishing in peer-refereed journals. DD is funded by The University of Melbourne. SED was funded by DoHA funding for the study and more recently by an Australian Research Council scholarship. MK is funded by an NHMRC Mid Career Development Grant. MM was funded by DoHA funding during the period of the study.
|
Notes |
|---|
Dunt D, Day SE, Kelaher M and Montalto M. The impact of standalone call centres and GP cooperatives on access to after hours GP care: a before and after study adjusted for secular trend. Family Practice 2006; 23: 453–460.
|
References |
|---|
|
TopAbstractBackgroundMethodologyResultsDiscussionConclusionAuthor contributionReferences |
|---|
1 Hallam L. (1994) Primary medical care outside normal working hours: review of published work. Br Med J 308:249–253.
2 Christensen MB and Olesen F. (1998) Out of hours service in Denmark: evaluation five years after reform. Br Med J 316:1502–1505.
3 After Hours Primary Medical Care Trials—National Evaluation Report. Available at: http://www.dhac.gov.au/pcd/programs/ahpmc/publications/ahpmctne.htm.
4 Leibowitz R, Day S, Dunt D. (2003) A systematic review of the effect of different models of after hours primary medical care services on clinical outcome, medical workload, and patient and GP satisfaction. Fam Pract 20:311–317.
5 Hansen BL and Munck A. (1998) Out-of-hours service in Denmark: the effect of a structural change. Br J Gen Pract 48:1497–1499.[ISI][Medline]
6 Dunt D, Day S, Kelaher M, Feldman P, Montalto M. (2005) The impact on service usage of different models of after-hours GP services: a population-based perspective. Aust N Z Health Policy 2:30.[CrossRef]
7 Turner VF, Bentley PJ, Hodgson SA, et al. (2002) Telephone triage in Western Australia. Med J Aust 176:100–103.[Medline]
8 Bolton P and Thompson L. (2001) The reasons for, and the lessons learned from the closure of the Canterbury GP After Hours Service. Aust Health Rev 24:66–73.[Medline]
9 Bolton P, Gannon S, Aro D. (2002) Health Connect: a trial of an after hours telephone triage service. Aust Health Rev 25:95–103.[Medline]
10 After Hours Care Project. Final Report. Available at: http://www.hudgp.org.au/reports/afthrcare/default.asp.
11 Penshansky R and Thomas JW. (1981) The concept of Access. Med Care 19:127–140.[ISI][Medline]
12 McKinley RK, Manku-Scott T, Hastings AM, French DP, Baker R. (1997) Reliability, validity of a new measure of patient satisfaction with out of hours primary medical care in the United Kingdom: development of a patient questionnaire. Br Med J 314:193–198.
13 Glynn LG, Byrne M, Newell J, Murphy AW. (2004) The effect of health status on patients' satisfaction with out-of-hours care provided by a family doctor co-operative. Fam Pract 21:677–683.
14 Salisbury C. (1997) Postal survey of patients' satisfaction with a general practice out-of-hours cooperative. Br Med J 314:1594–1598.
15 Shipman C, Payne F, Hooper R, Dale J. (2000) Patient satisfaction with out-of-hours services; how do GP co-operatives compare with deputizing and practice-based arrangements? J Pub Health Med 22:149–154.
16 O'Cathain A, Munro JF, Nicholl JP, Knowles E. (2000) How helpful is NHS Direct? Postal survey of callers. Br Med J 320:1035.
17 Roland M. (2002) Nurse-led telephone advice. Med J Aust 176:96.[Medline]
18 Department of Primary Industries and Energy and Department of Human Services and Health. Rural, Remote and Metropolitan Areas Classification. 1991 Census Edition. Canberra, 1994.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||