How many services and patients?
Among respondents to our survey it appears that around two-thirds of services have primary care staff operating within or alongside the emergency department. This is not representative, as respondents are a self-selecting group including more of those that have tried or adopted such a model. We estimate that around half of the services across the country have some form of primary care service working with the emergency department.
When we used a consistent definition and a consistent denominator of all emergency department cases we found that the proportion that could be classified as primary care cases (types that are regularly seen in general practice) was between 10% and 30%.
We identified three main operational models:
- A GP service located alongside or next to the emergency department. This is by far the most common model across the country.
- GPs working at the front of the department screening attendees and either treating or diverting to other places – effectively acting as a filter.
- GP services fully integrated into a joint operation covering the whole range of primary care and emergency services. We came across a small number of systems that offer a full primary care service to appropriate patients.
All of the services that we examined used staff to direct the patient to the correct skill or service. In the vast majority of cases this decision is made by a clinician whose role is also to ensure the patient does not have a condition that would make it unsafe for them to wait in a queue. We observed three approaches:
- A rapid decision by non-clinical reception staff – often using simple protocols.
- Rapid clinician assessment – typically by a senior nurse and typically taking less than two minutes.
- Patients wait for a full clinical assessment process, usually taking five to 15 minutes.
Most services use primary care practitioners from 8 till late. Very few use primary care staff during the 'red-eye' overnight period. About half ask them to take on responsibilities that are beyond typical general practice, such as interpreting X-Rays and a wider group of tests than are normally available to a GP surgery.
The initial reception process is critical. During our visits we observed that, where primary care clinicians and emergency care staff were at odds, this process was the main focus of discontent. Equally, where there was a feeling that the system worked well, it was clear that considerable joint work had gone into developing and refining the reception process.
Consistency across all parts of the service is important. In some cases, primary care practitioners found the work they were expected to undertake had changed out of all recognition – sometimes for just a short period – with no advance notification that this would happen. Equally some emergency departments complained that the primary care element of the service had sometimes been under-staffed so had stopped seeing most patients. Services need to work together to help each other at times of difficulty, but this requires communication and planning.
Quality of care
Most services, across both primary care providers and emergency departments, see queuing as inevitable. But this is not the case. We observed that the main reasons queues build up are poor scheduling of staff or inadequate premises that make it difficult to deploy staff effectively. This is a feature of overall management and governance. In fact, long queues can be avoided if capacity is sufficient to meet the demands of patients as they arrive, especially if a true ‘see and treat’ model is implemented.
The vast majority of primary care services use GPs, (some use nurses with primary care training) usually sessional GPs paid for each shift that they undertake with little or no long-term commitment. Services that regularly use the same individuals found that both the GPs themselves and hospital clinicians were confident that it led to more coherent and higher quality clinical decisions.
It must be remembered that the alternative to using primary care staff is that patients with primary care type presentations will often see doctors in training whose clinical skills are not yet fully developed. The result is slower treatment and often excessive investigation.
The proportion of cases that are seen by primary care professionals varies considerably, as does the expectations of those commissioning and delivering these services. There are a number of reasons for this variation – some of which reflect real underlying differences in the case mix, the operational process and model adopted or the different clinical ethos in the service. Many of the variations are also caused by the different way that such cases are counted.
In many services there is a lack of clarity over responsibility for important aspects of the scheme. There appeared to be little, if any, joint clinical or operational governance. Some organisations have started to address the problem of split accountability and the increased risks this involves, by developing closer collaboration across organisational boundaries. There are also examples of more formal collaboration through joint ventures.
Funding services in a more collaborative way encourages clinicians and managers to work together. Innovative examples of local tariffs show that it is possible to integrate urgent care while at the same time aligning the financial incentives and mitigating the economic risk to individual organisations across the healthcare system.
We need to emphasise that we found it very difficult to access information from commissioners or providers about the cost effectiveness of these services, despite initial requests for information of this kind and further reminders. It may be that this information is difficult to access or that it is not routinely collected. More work needs to be done to develop a consistent format for collecting this information to support wider comparisons across services, an issue we will address in the subsequent commissioning guide.
An Academic Review of the published evidence in this are was commissioned from Warwick University.
A narrative analysis of the data was undertaken and found:
- A GP working in the ED may result in less referrals for admission and less tests being undertaken. Cost benefits may exist but the evidence is weak.
- Redirect away from the ED has had variable results regarding future attendances and the assessments of the safety of this intervention have also revealed variable results. Whether or not a primary care appointment was made for patients being redirected from the ED some may not be kept.
- Educational interventions have not been shown to change attendance patterns.
- There is a paucity of evidence available to support the current system.