Refer to sender

Demand management has had a bad press lately, especially from GPs who feel that meddlesome primary care trust administrators have no business querying their clinical referrals.

So it is counterintuitive to find a PCT where all but one of 29 practices have voluntarily signed up to a clinical assessment service in which every referral is scrutinised. Even more notable is that the system for managing this is none other than the equally unpopular choose and book.

The PCT in question is Kingston in south-west London where the clinical assessment service was commissioned jointly by the PCT and local GPs, with the doctors working through a new agency: the Kingston Co-operative Initiative (KCI).

'It was an interesting one to cut our joint commissioning teeth on,' says Paula Head, director of quality and planning at Kingston PCT. 'It certainly tested our governance processes. They came out well.'

Local GPs funded the initiative with cash derived from their general medical services contracts, and employed two of their number as part-time medical directors before electing an eight-member board. In time, the founders hope it will become a not-for-profit agency operating under a specialist personal medical services contract.

Dr Charles Alessi, one of the medical directors of KCI and a GP at the Churchill Medical Centre in Kingston, says: 'We are 'of' primary care and work for primary care. Our remit is to take forward all practice-based commissioning and that means ensuring we have systems in place for the benefit of our patients.'

The clinical assessment service is KCI's first venture and started in June. It is a novel use of choose and book. Referrals arrive electronically via the system ? more than 1,000 in the first month. Dr Alessi and co-director Phil Moore (who also chairs Kingston PCT's professional executive committee), sometimes helped out by other KCI board members, review them, usually within 24 hours.

'The vast majority of the referrals are perfectly good,' says Dr Alessi before revealing that in the first month one in six needed to go back to the GP. Among them were one-line referrals or those where investigations were incomplete. In cases where another community-based service such as a GP with a special interest might be more appropriate, KCI pointed this out.

Dr Alessi does not like the term 'rejection' ? the official word in choose and book for a referral that is sent back to the GP for any reason.

'We always take referrals back to the practice with advice,' he says. 'We will ask &Quot;Have you considered this test?&Quot; or &Quot;Do you know about this alternative service?&Quot;. The decision remains the GP's although in the vast majority of cases they are happy to take our advice.'

When the service accepts a referral it is handed over to an administrative team run by Thamesdoc, a local out-of-hours service with offices in Kingston PCT's headquarters. 'They have the infrastructure and the staff to call patients at the time of day they are likely to be available,' explains Dr Moore.

Where patients have already discussed and made the choice about where they want to be treated, an operator rings to help find a convenient appointment. Where the choice remains to be made, the operator takes the patient through the whole process.

Early results are marked. Data from the NHS-wide clearing service ClearNET shows that the number of referrals to the main hospital provider fell by 500 ? around 25 per cent ? in the Kingston Clinical Assessment Service's (KCAS) first month. 'Of course it is too early to tell what the long-term effect is likely to be but there is a real difference in the numbers,' says Dr Alessi.

The next phase was getting under way in early July and involved assessment of consultant-to-consultant referrals ? a hugely controversial area. The PCT has a service-level agreement (SLA) with Kingston Hospital trust which says that all such referrals, bar some specific exclusions such as cancer, should be reviewed by the service.

In May 2006, Drs Alessi and Moore went to the hospital to explain the service and were shown to a room packed with largely hostile consultants. 'There were over 100 of them. They were lining the walls. Who says clinical engagement is dead?' remarks Dr Alessi. 'The consultants are not happy about it. It is a change in behaviour and a change in culture.'

Sylvia Kennedy, director of operations at the trust, admits that consultants were worried. 'They were concerned about their patients being delayed and who was responsible for them while a decision was made,' she says.

'Dr Moore and Dr Alessi explained how the system would work and why it was necessary and appropriate. They answered questions where they could and admitted there were things that they had not thought of when they couldn't answer.'

The result is that some intra-trust referrals are now being assessed by the service, she says. A list of exclusions has been drafted and continues to be refined between the PCT and the acute trust. Currently it is a paper-based system but in time Ms Kennedy hopes it will become automated.She says the key thing is the dialogue the initiative has generated. Furthermore, the work will feed into other clinical assessment centres; all five local PCTs have now set up SLAs with Kingston asking to assess intra-trust referrals.

'People are talking to each other. We need to capitalise on that as we move to other things such as redesigning pathways,' says Ms Kennedy.

Ms Head adds: 'Yes it is controversial, but I think it has fostered engagement between GPs and consultants. It could be the controversy of the issue has brought them together to deal with it.'

There is another set of exclusions that could also benefit from KCAS assessment: those for which there is a clinical evidence base saying that they are not terribly effective. The list includes grommets for glue ear, many dilation and curettage procedures and much varicose vein surgery.

Dr Alessi explains. 'To date, if they have been referred by GPs then they have been performed by the surgeons. Now with KCAS if we see referrals we can send them on to public health, who will screen them and either refer them back to the GP or ask for more information.' He believes this is the first time the NHS has picked up and implemented evidence on exclusions.

Dr Alessi is bullish about the potential for the clinical assessment service and the reasons for its success. 'This is not a demand management service in the classical sense,' he says. KCI is not part of the PCT and therefore has no role in performance management. 'It is there to support the practice and help it identify what it is spending.'

Ms Head agrees. 'It is fundamentally different. It is GP owned, GP run and managed and GP supported. We have facilitated its development and hopefully will enable that distance to be maintained.'

If Dr Alessi is proved right, the service will rapidly improve the quality of referrals. He is hoping that its role will be time limited. 'I don't want to do this forever. I hope we put ourselves out of business by improving the quality of referrals so much that we are not needed,' he says.

Certainly, the PCT will be keeping close tabs on it. Ms Head enthuses: 'That's the beauty of the system. We will know what's going on and in what specialties with what kind of patients. We will know what happened to our patients. We will know what's going to our GPs with special interests or what is going to the hospital that should be going to them. Because it is based in choose and book, we will have sophisticated data that will help us in our planning.'

Referral assessment schemes are just the tip of the iceberg, however. Much more challenging and potentially interesting to KCI and others is practice-based commissioning. As Dr Alessi puts it: 'GPs are interested in PBC. The whole basis of it is to ensure that the needs and aspirations of patients are taken forward.'

PBC budgets are not pooled in KCI ? although practices may choose to pool parts of their budgets as PBC gets under way. Its role is rather to foster innovation and help GPs understand the processes and limits within which they must work.

KCI has already developed the Kingston Education and Support Service by forming a partnership with the SW London Improvement Academy, the local organisation that mopped up Modernisation Agency cash when the latter became defunct last year.

The education and support service offers support to all practices and, says Dr Alessi, because it is independent of the PCT its advice is being accepted with open arms.

A first round of visits is almost complete, with each practice receiving advice on data management in preparation for PBC. The next round of visits will be to advise and train practices about commissioning intentions, including development of a commissioning template.

The template does not exist yet. 'It will have to satisfy the practices so that they are clear on what they are doing,' says Dr Moore. 'It will have to satisfy the PCT so that its clear practice commissioning intentions are within national and local priorities. It has to be able to spell out to providers any change that they will need to take account of.'

It has to do all this as well as show how it will benefit patients and improve care pathways ? crucially, without stifling innovation.

The final piece of the KCI jigsaw is a provider and treatment service, which Dr Alessi hopes will be up and running in the autumn. 'We would hope to see the PCT stop managing GPs with a special interest,' he says. 'We'd like to transfer ownership of existing services where the clinical governance is already in place to KCI. We do not believe PCTs should be doing this.'

Everyone involved in KCI is eager to stress that it is early days and so too soon to say how successful it is, although Ms Head says she is already seeing evidence of a change in referral patterns. There are no plans in place for an external evaluation, although Dr Moore is working on this.

Getting this far has not been without its headaches and teething problems, he says, adding: 'I can see how this will work for elective care. The interesting and really challenging part will be to see if we can extend this approach to urgent care.'