Striving for 'right first time' referrals
Our raison d’etre for setting up a clinically led referral management service (RMS) was not to save money but to improve the quality and patient experience by getting patients to the right place at the right time.
We launched the service in April 2011 and most patients have rated it as ‘excellent’ in our first year.
We have also forged strong relationships with secondary care colleagues by jointly producing more than 200 local care pathways. And we are now producing good commissioning intelligence for our CCG.
The concept of Kernow RMS was developed by a team of local GPs in 2010, including Dr Mike Ellis, chair of the then Carrick PBC consortium, and LMC chair Dr Phil Dommett, and has been impressively driven forward by manager Peter Stokes.
We realised our RMS needed the buy-in of all practices to succeed. So senior GPs who worked up the plans promoted the idea of a new centre that, on practices’ behalf, would manage all patients into their first outpatient appointments. The RMS would also create new care pathways to ensure patients got to the right place, at the right time, with the right tests – first time.
All 58 practices referring into Royal Cornwall Hospital in Truro signed up. The RMS would serve about 450,000 patients and manage about 110,000 referrals per year.
A small GP community interest company was formed that would be the glue to hold the concept together. All practices would hold a £1 single share, giving the RMS a strong mandate for change.
When GPs approached Cornwall and Isles of Scilly PCT for funding, they discovered the PCT was also considering referral management as the NHS began to get to grips with how it would meet the elective care QIPP challenge. Once GPs and the PCT got their thinking aligned, the RMS took off from there.
The PCT agreed to fund the RMS, using our PBC savings and the Choose and Book budget in the first year. The PCT would ‘own’ the service and employ the staff (18 patient choice advisers, nine GP ‘sifters’ and six managers and administrators), but the clinical direction would be set by the CIC acting as an umbrella board.
This board consists of five GP locality leads, one of whom chairs the board, an LMC representative, and the PCT’s director of commissioning.
In terms of how the day-to-day model works for RMS, call handlers book referrals in and offer patients choice, where referrals comply with local guidelines but refer cases to GP sifters that:
•fall outside local guidance
•are on the ‘limited clinical benefit’ list
•might require upgrading to the two-week rule
•have information missing, for instance no X-rays.
The GP sifter will then look at such referrals and decide if they should be allowed or returned. Only GP sifters can reject a referral for clinical reasons. Call handlers can only make a referral but also return some to surgeries where administrative detail is missing – for instance, no patient contact number.
The GP sifter will also look at all referrals for one specialty on a rolling basis to ensure they are meeting quality standards. So at the moment, for example, we are focusing on rheumatology so the GP sifters are looking at all referrals for this area.
The first task for the RMS was to engage with secondary care clinicians to jointly write local guidelines to encourage GPs to produce ‘right-first-time’ referrals.
Take gynaecology – women with heavy menstrual bleeding were being referred even though they hadn’t undergone an internal scan, hadn’t had an FBC, or been encouraged to try contraception methods that were known to help with the problem. GPs were generally keeping up to date but some did not feel confident to take the level of steps on NICE guidelines that a secondary care clinician might expect the GP should take. We hoped local guidelines produced collaboratively by primary and secondary care doctors would give GPs confidence to try out various steps or investigations first.
Hospital doctors were initially suspicious that we were trying to restrict their workflow but they came on board once we explained that we were trying to free up their time so they could do their job properly, and help them reposition their services in the face of competition from other providers.
Once written, drafts are shared between RMS GPs, the RMS board and secondary care. A hospital manager finally assesses any impact to ensure a whole outpatient department will not suddenly be put at risk. Final guidelines are then published on our website for GPs, in clear and simple language, generally kept to one page, so they can be used in a consultation. They are also publicly available for patients to read.
The right pace
The RMS decided to target the top 13 specialty referral areas, and to publish no more than three new specialty guidelines per month. That did lead to a slight tension initially with the PCT, which asked the board to go faster to meet the QIPP challenge. The board rightly told the PCT that there would be no point, as GPs would feel swamped, and the pace of change needed to be acceptable to GPs in order to deliver a change of culture. It was refreshing that the PCT accepted our reasoning.
When guidelines are first published, referrals are ‘pseudo-sifted’ by the RMS for about a month, to allow for a period of GP education and also to ensure we don’t suddenly have empty hospital clinics. During that time, the Choose and Book advisers put referrals to one side for GP sifters to go through during their weekly session. Where we question the need for a referral, we will email a GP explaining that we’ve let it go through, but suggest that next time they might want, for example, to do some bloods beforehand.
Once the sifting goes ‘live’, if we decide to reject a referral, we do so via Choose and Book. The GP will see our reason, and also a copy of the relevant guidance. All referrals are managed using Choose and Book and practice staff have been brilliant at embracing this. We don’t, however, handle any referrals for two-week-waits: they go direct to the hospital.
We occasionally see a referral that we think fits the criteria for the two-week cancer rule and when we do, we ring the referring GP, ask them to reconsider the priority and then fax the referral to the hospital.
Before the RMS was introduced, referral rates for first outpatient attendances had been going up 5% year on year; in the past year, they have declined by about 5%. We know that GPs are looking more closely in house at what needs to be referred, and it’s this that has led to the drop, not the RMS rejecting hundreds of referrals.
Secondary care consultants also tell us that referral quality is far higher, patients are arriving with the right tests already done, and conversion rates – the number of patients referred who need a procedure – are higher. For one obstetrics/gynaecology surgeon, the pay-off was immediate. Usually 14% of patients would be booked straight to operation – in her first fully sifted clinic, bookings rose to about 80%. For the patient and hospital that’s fantastic because the patient doesn’t need to go back two or three times before an operation. We have had a 6% cancer pick-up rate on the small number of referrals that we suggest are upgraded.
The call handlers – who the GP sifters work with – do an impressive job in offering true patient choice. They also talk to the patient about things such as parking or flight times for helicopters: a number of our patients live on the Isles of Scilly. The call handlers are not interested in who’s running a provider, they just want patients to be seen in the right place at the time of the patient’s choosing, if possible. We know our work has spurred GPs to set up their own in-house peer review processes.
As a GP sifter, I get paid the BMA sessional rate and spend half a day a week on the task. The GP sifters meet about once a month with a batch of referrals, to discuss how we would treat each one and why.
The biggest contentions are around ‘procedures of limited clinical benefit’, guidelines for which are set by the PCT’s professional executive committee. If GPs feel strongly enough that their referral is warranted, they can submit it to a special panel. However, we might have received a referral that says ‘Please see this 55-year-old lady with varicose veins, I think she needs to have them sorted out.’ They miss out that she has already had compression stockings, is on regular analgesia, her leg swells up, and she has venous eczema. So education of GPs is just as much about how to word referrals as it is about improving knowledge of best practice guidelines. I think the clinicians most receptive to our feedback are younger GPs who are used to being peer reviewed.
We now have a wealth of data about high, low and average referrers; where patients choose to go; true slot unavailability and pathway issues. We never point the finger at people and say ‘you’re doing this wrong’. We publish data to practices about their own rates and when CCG budgets go live next year, GPs will know where they are on the curve. We will, as a GP community, need to decide what to do with that.
In the short-term, the CCG (which now has a representative on our board) has told us the RMS is a cornerstone of elective care that will sit under its commissioning wing. There is yet to be a debate about where we will sit in the long-term.
I am genuinely proud of what is happening here. It’s all about improving the patient pathway, and though it might sound trite, if you improve the quality of your referrals, you inevitably decrease referrals: saving money and improving the experience for patients.
Dr Rob White is a GP in St Agnes, Cornwall, and a GP sifter for Kernow Referral Management Service
⦁ Initiative A GP-owned community interest company that employs GP sifters and referral advisers to ensure GP referrals are appropriate. The model follows some 200 guidelines jointly produced by local primary and secondary care clinicans
⦁ Outcomes An inevitable consequence of improving the quality of referrals and aiming for ‘right first time’ has been a 5% decrease in the number of referrals – largely as a result of practices looking internally at their own referrals. Consultants are also reporting that more appropriate cases are coming through their clinics
⦁ Staff 18 patient choice advisers, 9 GP sifters and 6 managers/administrators