This site is intended for health professionals only

Using professional carers to cut emergency admissions

Dr John Havard explains how his consortium set up a simple scheme to support vulnerable patients at home and avoid a costly hospital admission.

GPs know that elderly patients who live alone are vulnerable to falls, neglect and poor medication compliance, and it does not take much for an acute hospital admission to occur. A relatively trivial infection like a UTI can lead to mild confusion and unsteadiness and a couple of days in hospital may seem a reasonable option if there are no relatives to help.

But the reality is that the patient often becomes twice as confused in hospital, increasing the chance of falls and serious morbidity. Even if this hazard is avoided they still have little chance of a prompt return home because, once they are well, a home visit is needed to identify the need for grab rails, ramps as well as carpet and bathroom hazards. This is depressing and potentially ‘institutionalising', when often all that is needed is someone living in the house to ensure the patient eats and drinks and takes their antibiotics until normality is restored.

This was recognised by the Commissioning Ideals Alliance (CIA) consortium, and the solution we came up with was a scheme to put a live-in professional carer into the home of a vulnerable patient for 72 hours. Carers are with the patient within three hours of the request being made at a cost of £370 compared with up to £2,500 for admission.

In the period March 2009 to June 2010 we believe we have saved around £46,000 excluding training costs.

Getting started

In 2009 a group of 8 GPs representing 50,000 patients from the CIA consortium sat down for a brainstorming session. We wanted to identify what we could do better and where the gaps in patient services were. One common source of frustration was people being admitted to hospital for primarily social rather than medical reasons.

As well as the more obvious drawbacks of unnecessary admissions, there was also anecdotal evidence of patients who came out in a worse state than when they went in, having fallen over in the unfamiliar surroundings of the hospital and broken their hip for example.

So we wanted to come up with a simple scheme under which someone could be kept out of hospital, looked after round the clock by a carer in their house who would be there when they needed to be fed, washed, and so on.

In the early stage we did some research with the emergency admissions unit at Ipswich Hospital to see whether they recognised that people were coming into hospital unnecessarily. But when you look at a discharge summary it will always list a good medical reason for admission, for example ‘confusion, UTI'. They may have just needed a bit of support but it is difficult to read that into the summary. However anecdotally there was good evidence that this was a common problem.

There were three phases to our project. The first was in primary care where previously, if we had a patient who needed support because of the risk of falls, confusion and so on, we had no alternative but to send them to hospital. This was typically an elderly patient with an infection.

Now the GP can leave the house after invoking InstantCARE, knowing that the patient will be professionally looked after by a responsible carer who will come and live with them for three days and contact the practice if unforeseen problems occur.

At this stage we wanted to develop the idea ourselves but then contract someone else to deliver it for us. Luckily Saxmundham is home to the largest national live-in care agency, Christies Care, with 800-900 full time live in carers on their books, so going from the concept to reality was fairly straightforward. Knowing from the start who would be managing this service meant we could involve them in the evolution of the service from the beginning.

At that point they were offering live-in carers for seven day stretches whereas we wanted to have someone there for 72 hours. We discussed this with the agency, with whom we had a good relationship, and negotiated three day packages. We feel this is long enough to be a vital lead-in time for other services (like community nurses or social workers) who may need to introduce or amend an existing care package.

The plan received strong PCT support from the outset and we are grateful to the Deputy Director of Commissioning at NHS Suffolk Andrew McDonald for his sustained support.


Next steps

We agreed with the PCT that we could use £30k from our freed up resources under PBC for this project.

We have agreed with Christies Care a fee of £370 per 72hr period of work along with some simple contractual obligations.

These are:

To provide a trained carer to the patients house within three hours of request.

To have the carer live-in for 72 hrs and to undertake the necessary caring, cleaning and cooking for the patient over this time.

To accurately record and remind the patient about taking prescribed medication without a duty to administer.

(These patients are self medicating – the carer just reminds the patient to take it and records that it has or has not been taken)

If the patient recovers within a day or two then the residual time is set aside for restoring social structures in the community and for the carer to observe that the patient is selfcaring competently. We may be able to negotiate to move the carer on to another patient but there is absolutely no contractual right here.

The contract with Christies Care stipulates a short time-frame from initiation to carer arrival as well as some paperwork to monitor medication compliance and of course satisfaction.

After about six months we decided to try to capture the patients who were going directly to hospital after dialling 999 following a fall. These patients were often admitted and impacted on our unscheduled care budget. We also felt that these patients were being assessed well, yet still being admitted because there was no social support at home. Now after A+E assessment, if it is recognised that all they need is a period of care at home rather than an admission, the InstantCARER will accompany them home and stay for up to 72 hours.

At the moment we have someone sitting in A & E for the whole weekend, but this was mainly to prove that there was a need and admissions could be avoided in this way. We now intend to have someone on the end of the phone which is much more efficient- provided they are quick enough to meet the four hour target.

We've done a trial of a full week with the carer in A & E from 9-5 every day in which we had four cases. In many of those the carer just needed to go home with the patient - having avoided the admission - and make sure the existing carers could pick up the need. None of these patients needed 72 hours of care and most involved just one night. The usual carers decided that the patient was back to normal and the InstantCARER could stand down.


Where InstantCARE is invoked from the GP's surgery or patient's home, it can be hard to prove that an admission has been avoided, but where the patient was brought into A&E it is much more clear cut since the service is only invoked at the point of admission.

Here is a breakdown of our costs and savings:


Instant Care Costs

Care initiated by practices from surgery or patient's home – 12 periods of care

Total: £4,200

IHT (Ipswich Hospital Trust) Weekend A & E Project

26 weekends from November 09 (when this phase of the project started) to June 2010 = 28 weekends @ £375/weekend

Total: £10,500

Total costs of care in both phases of the project: £14,700

Add training costs: £6,881

Total cost Instant Care to June 2010: £21,581


Savings Generated (period to June 2010)

IHT Weekend A & E project to June 2010 - 12 patients had carer

Practice-initiated care periods – also 12 carers

Total avoided cost of 24 Admissions @ £2,539 each = £60,936

Less Instant Care costs: £14,700

Savings March 2009 to June 2010 (excluding training costs) = £46,236

Lessons learned

Work had to be done to persuade colleagues that this was a better option than admission. We had to make the paperwork simple and the contact process straightforward since it is really very easy to admit a patient. GPs have to be actively encouraged to take the harder route although the evidence suggests their patients love it.

We are now working towards having dedicated InstantCARERS who will be on the end of the phone. We're recruiting around Ipswich Hospital at the moment to find carers who would be willing to do a short term stints– perhaps those whose children have grown up and have a bit of time on their hands. We have arranged a one week training course run by Christies Care at their training centre in Saxmundham to ensure the carer is fully competent and professional.

As well as the immediate benefits of avoiding admissions we also believe this will help develop a sense of community. We envisage perhaps a group of friends who will undertake to provide a carer when required and will sort out amongst themselves who will go. As well as payment for their time we will need to train them as stated above and ensure they undergo CRB checks.

We also want to tackle day nought discharges where people are admitted purely because it's not practical to send them home at 3:00am. Instead, a carer could come to the hospital, take them home, put them to bed and contact the GP in the morning if need be.

We're saving the cost of the admission, and the patient would usually much prefer to go home rather than spend the night in hospital.

We wanted InstantCARE to work for the whole 320,000 catchment area of Ipswich hospital so any patient who comes in is eligible, and this meant getting the agreement of all the PBC groups in the area. It wouldn't work if the hospital had to find out who was or wasn't covered by the service. It has to be easy for them or they won't come on board.

This small example is a rallying call for unified budgets and integrated care. The bigger picture implies that these cases are really more social than medical since they were successfully managed at home. However, presently the medical discharge summaries look very medical and social care will reasonably argue that an acute medical admission was required.

If, however, the patient was managed at home with antibiotics, pushing fluids and a diet they enjoyed at a time they preferred, then the situation is very different.

Dr John Havard is a GP in Saxmundham, Suffolk and chair of the Commissioning Ideals Alliance consortium