Search tips
Search criteria 


Logo of brjgenpracRCGP homepageJ R Coll Gen Pract at PubMed CentralBJGP at RCGPBJGP at RCGP
Br J Gen Pract. 2006 October 1; 56(531): 799.
PMCID: PMC1920729

Dangerous Cats and Non-Barking Dogs

Paul Hodgkin, Writes in a personal capacity

How can we control demand from primary care? PCTs have struggled for years with rising elective referrals while simultaneously knowing that that some referrals are unnecessary and others could be dealt with by communitybased services. That's why PCTs have set up CATS — Clinical Assessment and Treatment Services. CATS (and the closely related Referral Management Services and Clinical Assessment Services) are intended to ensure that referrals are appropriate and hence by implication that demand is reduced. They are usually PCT-led and often backed by local policies that force all referrals for specified specialties to be assessed initially by the service. Evidence to date shows that CATS can be effective in reducing the volume of referrals.

In a world of limited budgets and rapidly changing practice the referral system needs to change. GPs need to become more flexible and quality control needs to increase. But this is an intrinsically risky way to skin the demand management cat.

Firstly over time CATS are likely to de-skill referrers and de-motivate GPs. The problem is not that GPs may resent being forced to refer in this way — GPs have grumbled about change since time immemorial. The problem is that we will come to like it. At one level an onerous duty has been taken off our busy shoulders, but at a deeper level it is the ultimate professional emasculation. If not gatekeepers then who are we?

Secondly those responsible for CATS will find that very significant levels of clinical risk have been transferred from GPs to themselves. A typical CAT service responsible for 100 000 people may generate 600 new referrals to hospital each week or some 30 000 per year — if half the referrals are reviewed that's a sizeable number of significant, highly salient clinical decisions. And CATS are not your local friendly general practice that's been around forever. They'll appear to patients as distant bureaucratic organisations, denying taxpayers their birthright. Imagine a 55-year-old man with back pain referred by the GP for an (inappropriate) orthopaedic consultation and triaged (correctly) by the CATS to physio who is found 3 months later to have co-incidental spinal secondaries from his undiagnosed prostate cancer. Who will be sued?

Finally GPs can undermine any CATS that they have no confidence in. ‘I'd like to refer you to see a proper consultant right now but I'm afraid I now need to get permission first’. If 1 in 1000 CATS' decisions ends in litigation, that could mean a new case every 2–3 weeks for a typical service covering 100 000 people. Viewed in this light the lack of litigation about the current 15 million GP referrals per year is the dog that isn't barking — perhaps referral decisions are trivially easy to get right. Or perhaps GPs really are adding value to a complex and often ambiguous process.

So if CATS are not the answer what is? The solution is a service that focuses primarily on helping all referrers become excellent and generates referral management as a byproduct of education. In this model practices retain all responsibility for referrals with a subset of letters reviewed by a consultant working alongside a GP for 2 hours per week. GPs rotate every 2 months in order to build relationships with consultant colleagues and spread best practice. Individual GPs do a paid 2-month rotation once every 2 years so that a typical five-handed practice is updated by two specialties per year. When referral can be improved the reviewing GP phones the practice to explain. The practice retains a right of veto — and of clinical responsibility. Costs are met from savings to practice-based commissioning consortia. Over time the quality of referrals across all practices and all specialties are improved.

The new electronic booking system also offers interesting possibilities. Consultants could grade incoming referral letters using a similar system to that used on eBay between buyers and sellers. Each referral would be scored by the consultant for appropriateness on a −1, 0 or +1 scale. Over time each GP would accrue a consultant-generated score about the quality of their referral letters. Such reputations would be powerful quality drivers especially when shared with the public. At the same time clinical responsibility would remain right where it should be — with highly skilled GPs who know their patients.

PCTs are understandably exercised about controlling referrals. But they should talk with GPs before opting for the Big Game solution.

Articles from The British Journal of General Practice are provided here courtesy of Royal College of General Practitioners