Pulse's new campaign – A right to refer – is opposing crude restrictions on referrals and promoting peer-led schemes. Here, GPs present two case studies of how they are working together to control their referral rates – without the need to resort to blanket bans or referral management centres
Producing a referral guide for each practice
Dr Jeff Cotterill, Rugby CCG
We had high referrals in certain specialities – particularly in trauma and orthopaedics. GP referrals to orthopaedics were at one point almost double the national average.
In the days of practise-based commissioning we collected referral data from GPs themselves, rather than getting it second hand via providers. We got GPs to record not only the referrals, but the reason for referral. The problem seemed to be that people were simply not sure what to do.
Nobody can be an expert in everything. If you'd showed people some sensible things to do – like pre-referral investigations and treatments to do before you refer – then afterwards you could put in a threshold, and perhaps a time delay, before referral.
What we did
We put together a document for GPs that outlines the main reasons for referral, including a summary of what you should do before you refer, what you should investigate, and roughly at what point you ought to be referring. Rather than rely on a referral management centre we decided that we'd try and do this ourselves.
We set up a system where there is an expert partner in each practice in each speciality – particularly for the high-referring ones like orthopaedics. GPs send their problems – for example, hips, knees and backs – through the expert partner, who will then build up the expertise to decide whether or not to refer themselves.
So you're not trying to ask everybody to be really good at everything. There was one practice where everybody worked as an individual rather than a team. We also asked GPs to monitor their referral rates and give those to us on a fortnightly basis. We produce rough quotas, called ‘monthly expected activity', to give them an idea. They're not strict quotas – if you go over one month it doesn't matter.
When we first started this, we noticed great variation in high- and low-referring practices. We looked at two big low-referring practices, which turned out to be my practice and the practice of the CCG chair. We knew we were referring at safe levels – there were no adverse incidents during several years of relatively low referrals – so we thought that was a realistic target to go for rather than picking it arbitrarily. So we set up the referrals guide, recommended a strategy of internal rather than external referrals, and also set practices a loose target which we felt was evidence-based from local knowledge as being safe.
We also have two GPSIs in musculoskeletal problems – so we made their expertise available across the clinical commissioning group for advice via email or a telephone call. So, for example, if a small practice doesn't do all the joint injections, we can send someone in to teach them how to do it.
What we learned
I think the main lesson was that ownership at practice level is crucial. Don't let the referral management centre do it for you – do it yourself. GPs don't have to be an expert in everything, but parcel out your highest referral specialities among your partners and let them develop their expertise in their area. We think the only way to do it safely is to have an expert GP with the patient in front of them who is used to seeing that sort of problem. In Rugby, even the smallest practices have been able to do that. Training, motivation and having one or two local champions have also been really helpful.
According to real-time monitoring on our system, our GP referrals are 25% down on the contracted target based on a value set against last year's activity.
Total outpatient referrals are down by 7% – the impact is a bit less with the local hospital doing things with outpatient follow-ups and other referrals. There have been no serious adverse events. We have had the odd moan from patients like ‘you're saving the Government money' – ‘that's the whole point,' I tell them.
The guide has been in development for two years, while real-time monitoring of referrals started at the beginning of this financial year. This was when GP engagement really kicked in, when we felt, ‘this is real now'. We're not playing at it – we've really got to do this.
We don't think we can squeeze any more out of GP referrals, so now we're looking at non GP-first referrals and the follow-ups in the hospital. With a 25% reduction in GP referrals and a 7% reduction in total outpatients, it's time to change our focus.
We've looked at orthopaedics and ophthalmology, as well as areas in which the patient is likely to be seen twice on the same day. One example of this is with glaucoma follow-up – they may see the consultant for pressures and drug advice, but they may also see the visual field tester, and you can charge two follow-ups for that.
There is very variable practice across the country in the way you interpret payment by results rules, and they're not as scientific as we would like. We are also looking at A&E and non-elective admissions.
Dr Jeff Cotterill is a GP in Rugby and an executive member of Rugby CCG
Enlisting consultants to create GP care plans
Dr John Havard, East Federation CCG, Suffolk
We wanted to find out why some people were referring much more than others, and were thinking, how can we independently and objectively make sure that we keep the standard up? Using GPSIs to advise on referrals can be a bit arbitrary. Using referral management centres possibly run by nurses is just inflammatory. But I thought if I set up a system where it was done by the very consultant you were referring to, and also sell it as going back in time, not forward – this is what consultants used to do 30 years ago – then it could work.
What we did
Advice Letter Listing (ALL) ensures that a consultant sees every referral letter and appropriate attachments on Choose and Book and responds to it within three working days. The response may be:
• a traditional outpatient appointment in the correct clinic
• a community clinic appointment if the patient consents
• a community management plan to support the GP in managing the patient in the community
• a request for further tests or information to improve the new to follow-up ratio.
The system allows our consultant colleagues to advise on future management where appropriate – we started with two schemes, urology and gynaecology.
First, I send the referral to the specialist with the scan. The specialist may then say: ‘Everything is fine.' I then share that with the patient, we read it together and they do not have to go to hospital to see a registrar. The person who has given the advice has obviously seen the letter and read the investigation. The surgeons do it remotely in their downtime in theatre when they have access to Choose and Book.
Every speciality in Ipswich hospital is now covered by this service. A management plan is a way of learning. It's a way for consultants to get closer to us and actually make us stronger together. For the trusts who have the most to lose, management plans are making the relationship with GPs stronger, because we are closer to them, taking advice, sending people back if it doesn't work. We're setting these people up as our local specialists and developing a good relationship with them. In the long term, it's a far better way of doing it.
What we learned
We need to be sure we are referring in the patient's best interest and not simply for our own reassurance. Sometimes consultants can reassure us remotely by phone or email if they are enabled to do so. Peer review by GPs has definite merits, but a review by the local specialist must be better. We reserve peer review for the community management plans we receive back, so we are all aware of what our local consultants feel can be safely managed in the community.
But I don't think everyone will want to do this. It needs very close management both at the GP and consultant end. It provides much more personal care than going to hospital, but you have to do it, see it and feel it to realise it's not simply about cost cutting.
I've had emails from consultants saying ‘it's about money'. We have had robust debates with GPs. There's a lot of undoing of traditional behaviours, but the overwhelming thing is that this is better for patients. The consultants we did the pilot with have been fantastic, because they've seen the benefit. With others it's more difficult, because although the PCT has paid some money to the hospital trust to allow for the reviews rather than referral payments, the consultants don't see that as it's lost somewhere in the system.
A pilot for gynaecology and urology referrals led to a 23% reduction in outpatient appointments, with all patients receiving written community management plans. As part of the pilot, I also did a remote-advice project using an ENT specialist at the Royal National Throat, Nose and Ear Hospital in London. Our referrals went down by nearly 30% – mostly as GPs were thinking much harder about them. But it produced the lowest number of community management plans (10%), probably because there was no local knowledge of clinical alternatives, so the plans will now be run from Ipswich.
Some 23% of gynaecology patients and 24% of urology patients were managed in the community by their GPs. Other advantages were that consultants were able to determine if a community clinic costing 80% of tariff was appropriate. The only cost of the service is paying the consultants. Our gynaecology and urology referrals have dropped by 23%, and we were making 1,000 referrals per year in these two specialities. With an average outpatient appointment costing £175, this means a saving of around £39,000 in those two areas alone. In Suffolk, GPs make 64,838 referrals for the total number of specialities covered by ALL, which equates to £11,346,825 in outpatient appointments. If we assume a 20% reduction in line with the pilots, the saving in outpatient appointments would be £2,269,365 (less a 5% deflection back to hospital). After costs of £646,920, we can make a saving of £1,508,977.
If the scheme was extended or adopted nationwide, it could save a quarter of a billion pounds a year in outpatient costs. But it's not just the outpatient costs – once someone gets to the hospital, they can get all sorts of other charges and procedures that may be unnecessary.
The critical thing is that the referral is not a request for somebody to be seen in a secondary care institution, it's advice on how best to manage a patient – which may involve going to the hospital or not. Let's give consultants the choice. When we refer, the patient has an expectation of an appointment, as does the GP. But if we say: ‘We're going to ask the best way for us to look after you', if they come back with advice, no one's let down. QIPP begins with a ‘Q'. If you get the quality right, you find you save money almost accidentally.
Dr John Havard is a GP in Saxmundham, Suffolk, and an executive board member of East Federation CCG
Click here to find out more about Pulse's campaign and register your support