Journal of Health Services Research & Policy >
Volume 13, Number 4 >
Pp. 233-239
Original research |
National Primary Care Research and Development Centre, University of Manchester, Manchester;
1 Department of Health Economics, University of Birmingham, Birmingham, UK
Correspondence to: bonnie.sibbald{at}manchester.ac.uk
Objectives: To assess the likely impact on patients and local health economies of shifting specialist care from hospitals to the community in 30 demonstration sites in England.
Methods: The evaluation comprised: interviews with service providers at 30 sites, supplemented by interviews with commissioners, GPs and hospital doctors at 12 sites; economic case studies in six sites; and patient surveys at 30 sites plus at nine conventional outpatient services. Outcomes comprised: staff views of service organization and development, impact on primary and secondary care, and benefits for patients; cost per consultation and cost per patient in new services compared to estimates of the price of services if undertaken by hospitals; patients' views of waiting time, access, quality (technical and interpersonal), coordination and satisfaction.
Results: New services required high initial investment in staff, premises and equipment, and the support of hospital consultants. Most new services were added to existing hospital services so expanded capacity. Patient reported waiting times (6.7 versus 10.1 weeks; p = 0.001); technical quality of care (96.2 versus 94.5; p < 0.001), overall satisfaction (88.2 versus 85.4; p = 0.04); and access (72.2 versus 65.8; p = 0.001) were significantly better for new compared to conventional services but there was no significant difference in coordination or interpersonal quality of care. Some service providers expressed concerns about service quality. New services dealt with less complex conditions and undercut the price tariff applied to hospitals thus providing a cost saving to commissioners. There was some concern that expansion of new services might destabilize hospitals.
Conclusions: Moving specialist care into the community can improve patient access, particularly when new services are added to existing hospital services. Wider impacts on health care quality, capacity and cost merit closer scrutiny before rollout.
 | Introduction |
|---|
 Top Introduction MethodsResultsDiscussionAcknowledgementsReferences |
|---|
‘Closer to Home’ services are intended to substitute community for hospital care rather than simply move existing specialists into community settings. This is to be accomplished by making greater use of specially trained community staff, such as general practitioners (GPs), nurses and other Practitioners with Special Interests (PwSIs), and through the increased provision of diagnostic and treatment facilities in community sites. In 2006, five demonstration sites in each of six specialties (n = 30) (dermatology; ear nose and throat surgery; general surgery; gynaecology; orthopaedics; and urology) were selected by the Department of Health to show how this could be achieved.
Previous research suggested that services provided by PwSIs could improve access to specialist care and reduce demand on hospitals.4,5 There was a risk, however, that service quality might decline and costs increase. The overall impact on local health economies, particularly the opportunity costs in hospitals, was unknown. We therefore conducted an evaluation of ‘Closer to Home’ demonstration services between September 2006 and May 2007 to: describe service organization and the factors which facilitated or impeded service development; estimate the likely impact on local health economies; and assess patients' views about service access, quality and coordination. Key findings are summarized below with full details available in the final project report.6
| Methods |
|---|
|
TopIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Interviews with service managers covered their perceptions of: service organization; factors affecting service development; impacts on primary and secondary care; and benefits for patients, practitioners and the NHS. Additional interviews (n = 58) covering the same topics were conducted at two sites in each of the six specialties with: service commissioners; GPs referring to the new service; ‘Closer to Home’ providers; and a hospital consultant in the same specialty. Interviews were audio-taped and subjected to thematic analysis.
Information provided by service managers was used to group services by type, after the framework described by Sibbald et al.4 Information from all interviews was used to identify common factors affecting service development and perceived impacts on patients, professionals and local health economies.
Economic case studies were conducted at six sites, one in each specialty. Patient-level data (monthly summaries at one site) covering a 12-month period (six months in one site) were used to quantify service capacity and type of clinical activity. Cost data provided by each site were used to estimate the cost per consultation and cost per patient. National tariff prices for 2006–2007 were used to estimate what demonstration services would have cost commissioners if undertaken in hospitals.7 Sensitivity analyses of costs to different assumptions about service configuration and patient flows were carried out. Differences in distance travelled by patients were estimated using the first part of patients' postcodes and Automobile Association and Google Maps route planner software.
Questionnaires were mailed to the last 50 patients seen in each demonstration service (1500 total) of which 783 (52%) were completed. Questionnaires were also mailed to the last 50 patients seen in nine conventional outpatient services (450 total), matched by specialty to demonstration services, of which 275 (61%) were completed. The questionnaire, based on Parchman et al., covered the domains of service access, quality of care (interpersonal and technical) and coordination of care.8 Overall measures for each domain were created by summing responses to relevant items and expressing the total on scales ranging from 0 to 100, with higher scores representing better outcomes. Internal reliability, was good for access (Cronbach's alpha 0.76) and interpersonal quality (0.90) but only moderate for technical quality (0.60) and coordination (0.55). Patients were also asked to report on: waiting time for specialist care; overall satisfaction with care (standardized to same scale as other measures); and provide information on their sex, age, ethnicity, housing tenure, general health and number of co-morbid conditions.
Multilevel regression, using Stata version 9, was used to examine the relationship of service type (three types of demonstration services plus conventional services) to access, quality, coordination and satisfaction after controlling for specialty and patient characteristics. A random effects model was applied in which the within- and between-site relationships with patient characteristics were allowed to differ. Post-estimation tests were used to compare: all demonstration services to conventional services and differences among demonstration services. We used propensity score methods to check the robustness of the findings.9,10
| Results |
|---|
|
TopIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Three types of demonstration services were identified: transfer to primary care; relocation into the community; and redesign of hospital care (Table 1). The most common types entailed a transfer of activity from hospital consultant teams to community PwSIs or a relocation of hospital clinicians to community settings. There was wide variation in terms of how long services had been operating (from 3 months to 27 years), and the volume of patients seen in the previous four weeks (from 3 to 1200).
|
Service location was determined as much by site availability as by patient need. Service providers pointed to: the additional costs of equipping multiple sites; the challenges of ensuring equipment was properly maintained and decontaminated; and lost clinical time in travelling among sites. Economies of scale were sometimes achieved by co-locating a number of new services in the same community facility.
Care was provided in most services by PwSIs. At the time of the study there was no national scheme for accrediting such workers, so training was developed locally. While some sites had no difficulty recruiting PwSIs, others were constrained by a lack of funds and/or a shortage of candidates. An added difficulty was that ‘Closer to Home’ sites were not accredited as venues for training doctors. There was concern that PwSIs might mismanage patients or delay necessary hospital treatment (Box 1). Robust arrangements for service management were identified as key to addressing these concerns, including clear written guidelines governing the role of PwSIs.
| Box 1 Quality and safety concerns of service providers ‘Risks are quite honestly that patients will slip through the net on the basis that either your protocols aren't tight enough, your equipment isn't good enough, your staff are not trained enough. If you haven't thought those through or you have difficulties in any one of those areas, the patient could end up suffering as a result.’ (Urology sonographer) ‘Consultants are doing dermatology full-time, they've had full-time training for 4 or 5 years, they are just 10 times more expert than GPwSIs [General Practitioners with Special Interests]. I think GPwSIs can deal with simple dermatological problems or diagnostically simple problems, but I don't think there is any comparison with their skills.’ (Dermatology consultant) ‘I don't believe there are skilled GPwSIs in the community who could do the job of a hospital specialist ... In order for GPs, or nurse specialists, to work effectively, they should always have the safety net of working alongside a consultant. ...You can educate them and make them smarter and that is part of the continuing medical education, but you cannot turn a GP into a specialist.’ (ENT consultant) ‘As we expanded we realized we needed the consultants to come out more often because it was clear there were limits to the number of patients we could see on our own without adequate back-up from the consultants... I imagine there are fewer clinical governance issues in something like dermatology where you wouldn't expect to come up against something life-threatening, whereas with ENT, although it's not common, it happens and is not rare... Sometimes it's about having that experience to know what's right and what's not right, and in ENT that difference can be quite subtle.’ (ENT GPwSI) ‘I think when we first started we questioned what if we miss something, what if there's a cancer ... I mean you do pick them up, but sometimes it's very hard to decide whether they go straight to the "top" or straight to the [Closer to Home] service. So missing something, that was one of the major risks everyone felt.’ (Orthopaedic manager) Text in italics shows quotes recorded in fieldnotes at time of interview
|
The factors said by service providers to facilitate service implementation included: learning from similar services elsewhere; local champions and continuity of leadership; positive prior working relationships among key stakeholders; and stakeholder consultation. This last factor was critical to success. Services required the support of at least one consultant. Prior consultation with other health care providers, notably GPs and community nurses, whose work might be affected was also important.
Patients were generally consulted after, not before, new services were introduced. The main benefits for them were said to be shorter waiting times for specialist care and improved access due to fewer difficulties with car parking.
Community providers could bid to provide services at lower cost than the national tariff fixed for hospitals. Service providers expressed concern that moving simple cases out of hospitals would leave hospitals with a more complex case-mix, but no adjustment in the tariff to reflect this (Box 2). Consultants were concerned that ‘Closer to Home’ expansion might leave some hospitals without sufficient work to sustain particular specialist services (Box 3).
| Box 2 Service providers' views of impact on hospital case-mix and income ‘The GPSI service filters out the "easy" stuff, in some respects, and we get left with the more complex stuff, and that's always going to be an issue... I don't have a problem with it, so long as we are funded for it... For example, one of the GPwSIs just phoned, he's seen a guy yesterday in the clinic and wants me to sort him out. Now, I will sort him out, that's fine. He's actually going to go onto my waiting list. So we'll do the operation, but we'll never see the tariff that would have come with that patient if he'd been seen in outpatients.’ (ENT consultant) ‘It has obviously meant a huge reduction in referrals to secondary care. It has meant that secondary care can meet their targets. It's meant that secondary care have the appropriate patients now, and we've managed to fill our "Take or Pay" contract, (which basically means that if you don't use it, you still have to pay!), so we've managed to fill this with appropriate patients that aren't rejected, and have met our targets as well.’ (Orthopaedics manager) ‘The fact that [Closer to Home practitioner] and his colleagues are doing 2000 of these procedures here on our behalf, means that we are freed up to do other more complex things. If [Closer to Home practitioner] and his colleagues were to all resign tomorrow, we couldn't cope. Our system couldn't cope and all the targets and achievements we're making (we're achieving all the government's targets at every level) we wouldn't be able to meet these any more: the outpatient targets, the cancer targets, the waiting list targets.’ (Urology consultant) ‘ENT which is fairly "cheap and cheerful" is actually what subsidizes the preventive dementia and renal care, and things for which there is no income. The worry is that if GPs are allowed to "cherry-pick" out the services from secondary care that they want to do in primary care, this is ultimately to the detriment of the health economy, because we (secondary care) can't then afford to treat people who need expensive things doing to them for which there is no income stream.' (ENT consultant) ‘Well I think if you look at the whole tariff situation, there are certain procedures which cost far more than tariff, because we tend to get the more complex cases, so the other areas that are under-tariff will then subsidize that. If we moved all the simpler cases out, and we just do the complex ones, it is going to cost us more, because we're likely to use up tariff.’ (General surgery manager) ‘The danger to the Trust [of Closer to Home service] is that if you take away from outpatients clinical problems that can be dealt with very easily and quickly, for which you are paid quite a lot of money, then the Trust in a way can lose out. This is because if you have been referred patients who take a lot of time and are difficult, you have to put a lot more effort into those, so the Trust can be losing bits of its service, which reward it quite well for not doing a lot, and be left with complex cases which are more difficult and expensive to treat...The benefit to the Trust is that it has more time to deal with the more complex issues, but those more complex issues have got to be remunerated at a higher tariff than the easy ones.’ (Urology consultant) Text in italics shows quotes recorded in fieldnotes at time of interview
|
| Box 3 Service providers' concerns about hospital reconfiguration ‘I went to a meeting at the Royal College recently, and there was a lot of concern expressed about the prospect of hospitals closing if more and more services are moved out into the community. There was a lot of discussion, and a lot of people felt vulnerable, because a lot of it is the diagnostics stuff, the bread and butter, that keeps us going. However, in my speciality we're lucky in that we've got obstetrics, and obstetrics is never going to go away, GPs are scared of it, so I think at the moment we're likely to nearly always have a job.’ (Gynaecology consultant) Consultants have been the biggest challenge – more resistant to the idea of services moving out into the community for reasons of ‘professional pride’; ‘it goes against their medical model ’. Their biggest concern is with maintaining their jobs and status and are worried that hospital services will close. ‘They feel they've lost all control.’ (Orthopaedics manager) ‘...hospital departments closing down is inevitable’ (but not undesirable). (Urology consultant) ‘There is no doubt in my view that in the future the specialists who work in a hospital will mainly be doing operative procedures and seeing patients with very difficult problems requiring specialist care. If all the outpatients are being taken away and being seen in the community, whether by a specialist or by non-medical staff, that means that the consultants will have to change, the secondary care hospitals will have to change. The funding will go, there will just be money for the patient who lives in the community, so hospitals will either have to close or will have to amalgamate. We would therefore see some of the hospitals as non-sustainable simply because of the funding being pulled out, and if the funding is removed there will be less nurses, less junior staff and you could get into a situation which would not be sustainable.’ (ENT consultant) Text in italics shows quotes recorded in fieldnotes at time of interview; other text is taken from contemporaneous fieldnote
|
Table 2 shows the type and volume of activity, and the estimated cost per patient, in the six case study sites. From the commissioner perspective, ‘Closer to Home’ services were estimated to cost less than they would have cost under national tariff prices if they had been provided in hospitals, principally because the new services saw patients with less complex conditions. Sensitivity analyses showed this finding was robust to alternative assumptions about key aspects of service configuration and patient flow (Appendix 1, www.jhsrp.rsmjournals.com/cgi/content/full/13/4/233/DC1). In some sites, the loss of income to hospitals was not viewed as problematic because the new service was small in scale or introduced to provide additional capacity. In other sites, the impact was reported to have led to some reconfiguration of hospital services.
|
There were one to four services of each type (Transfer, Relocation, Redesign) in each clinical specialty with the exception of Redesign which did not occur in two specialities. Patient characteristics differed in minor respects across services. Table 3 summarizes the differences in patients' views of the services by service type. Patients reported shorter waits to attend demonstration services (p = 0.001) and found them more accessible (p = 0.001) than conventional services. Interpersonal quality of care did not vary significantly by service type but technical quality was rated higher at new services compared to controls (p < 0.001). Service coordination did not differ across services. Overall satisfaction was rated higher at new services compared to controls (p = 0.037). No aspects of care differed significantly by type of demonstration service. Sensitivity analyses confirmed that demonstration services were associated with shorter waiting times and improved accessibility but found no other significant differences (Appendix 2, www.jhsrp.rsmjournals.com/cgi/content/full/13/4/233/DC1).
|
| Discussion |
|---|
|
TopIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Wider health economies may be affected in three ways. First hospitals may need to reconfigure services if large volumes of care are moved out. While some hospitals struggling to meet current demand may benefit through the addition of ‘Closer to Home’ services, for other hospitals the loss of business may provoke closure of some services. Tariffs need adjustment to reflect the altered mix of patients seen in outpatient clinics if hospitals are to compete on an equal footing with ‘Closer to Home’ services.
Second is the possibility that rates of referral for specialist care might increase as services become more accessible.4 It was unclear whether the added capacity generated by ‘Closer to Home’ services was directed to unmet need or lowered the threshold for referral.
Third, patients accessed some ‘Closer to Home’ services without referral from a GP which might risk inappropriate use. As noted above, the overall cost effectiveness of health systems like the NHS derive partly from their division into primary and secondary care with the former governing access to the latter.1
The findings need to be interpreted within the limitations of the study. As policy-makers wanted evidence rapidly, the study had to be based on the perspectives of service providers, commissioners and patients. While such evidence has limited scientific rigor, the alternative was for policy to be made without the benefit of any research.
The patient survey did not match study and control sites for factors that may affect outcome (e.g. hospital size, location and performance) or adjust for non-response bias. However, the differences in access observed between demonstration and conventional services were large and were robust when subjected to sensitivity analyses, suggesting they were not an artefact of uncontrolled factors. Cost comparisons between demonstration and conventional services used national tariff data to estimate costs in the latter without adjustment for case-mix, which differs between services. Also, objective measures of quality of care were lacking. For these reasons, ‘Closer to Home’ services require more robust scientific evaluation.
The challenges of setting up ‘Closer to Home’ services which were observed are likely to apply to other sites seeking to move specialist care from hospitals to the community. Our findings point to the high initial investment needed in terms of service design, consultation with key stakeholders, training and accreditation of staff working in new roles, and finding and equipping suitable venues for service delivery. Whether this initial investment pays off in terms of improved quality of care and health outcomes for patients has yet to be determined.
| Acknowledgements |
|---|
|
TopIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
| Footnotes |
|---|
| References |
|---|
|
TopIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
