#WeNeedToTalkAboutJeremy

No one I know has any confidence that the 10 point plan developed by NHS England, Health Education England (HEE), the General Practitioners Committee (GPC) and the Royal College of General Practitioners (RCGP) in order to address primary care workforce issues in January 2015 has any chance of making a significant impact. I thought it might be worth exploring why.

The current crisis is down to too many GPs saying it's all too much, that although the job in principle is extremely worthwhile, the pressures mean increasing numbers just don't want to do it.

A way of assessing this is to look at willingness to do the job at different workloads as pay varies. If pay is high and workload manageable people will feel better about doing the job; if it is too low people may feel the job is not worth doing whatever other worthwhile elements or job satisfaction it offers.

A Commons Briefing Paper shows that between 2005 and 2013 real terms pay for GPs has dropped 22% from £137,108 to £107,331. 

The number of FTE GPs in the same period has gone from 29,248 to 32,075. However the estimated annual number of consultations has escalated from (my estimate) 267m to 340m over the same period. The workload (in consultations pa per GP) has therefore gone from 9,134 to 10,600, a 16% increase.

You will notice both measures are going in the wrong direction, meaning that the job in 2005 was 48% more worth doing than in 2013 when this is expressed as the change in Pay/Workload ratio. (Some may query my choice of 2005 as the baseline but bear in mind this was where we were after a solution to the last GP crisis was enacted and generally folk felt the crisis was resolved at that point and so this seems to me to be the most appropriate point to choose for a baseline.)

We can get a handle on the crisis by considering how pay and workload affect the proportion willing to do the job. 

Each line represents willingness to do the job for the same workload at different pay levels, with say 10% increases in workload between lines A through D. As the work stress increases there will come a point where the stress levels are such that willingness to do the job drops away rapidly as in C. In the most extreme cases where the job causes illness, work may become impossible and willingness drops out very rapidly as in D. I suspect that over the last 10 years we have moved from X to Y as real terms pay has dropped and workload shifted up a gear.

 My contention is that our current level of workload means pay vs willingness is not a linear relationship but has a steep slope, if not a cliff. I think there is good reason to think we are now at point Y on the curve C. This is very precarious for two reasons. Firstly, further drops in real pay, without a change in workload will continue to push us down the steep part of curve C. (Bear in mind GPs are generally very capable individuals who can and do switch to other parts of the economy or other economies.)
But this is not the worst thing that could happen. I suspect that secondly, like fissile material at critical mass, further increases in workload could trigger a chain reaction where loss of more GPs pushes up the workload for the remainder suddenly flicking us down vertically from Y onto curve D, in which large numbers just walk as the job becomes non-doable at that or any price. This disaster would be difficult and expensive to recover from! I think the level of anxiety in GP land at present reflects an intuition that this could happen.
So the solution to the crisis relies on finding ways to push us back up the curves to higher levels of willingness to do the job. (To be sure pay and workload are not the only reasons for doing the job as others have stressed (www.whyGP) which is why on curve A people with means would do the job voluntarily but these factors only shift the curve to the left, they don't change its shape at higher workloads: however fantastic you think General Practice is there is always a point when the stress of the job will make you unwilling to work. What makes this crisis such an issue is that the curve is likely to be already left-shifted by the intrinsic value of the job and the values of the profession compared with other jobs and yet we are still at Y.) My view is that reducing workload rather than increasing pay is likely to be more effective and you can see that from the shape of the curves - going from Y vertically to curve B is more efficient than a pay increase in getting people to stay, if I am right. The main determinant of workload is list size and so for an effective solution practices must be given the right to unilaterally close lists to contain workload, NHSE should commission to cope with the overflow and pay must not drop any further in real terms.

We are now in a position to understand why folk might be sceptical about the 10 point plan which fails to address the above adequately.
I will cover the points in turn

1 Promoting general practice 
This is a marketing campaign setting out the positive aspects of General Practice.

This will clearly make no difference to the Pay/Workload issue. It is possible that it might shift the curve to the left by convincing some to be more willing to do the job for the same degree of stress (but the curve is probably already as left shifted as it can go for the reasons listed above). This does rely on pulling the wool over many eyes in an age of social media, and them discounting their experience and the experience of the more experienced, which seems rather unlikely.

2 Improving the breadth of training 

HEE will work with partners to resource an additional year of post CCT training to candidates seeking to work in geographies, where it is hard to recruit trainees. 

This does nothing to shift the curve or address Pay/Workload once the doctor gets into General Practice and may increase work stress generally by taking a cohort out of full service delivery for a year.

3 Training Hubs  

NHS England will invest in the development of pilot training hubs, where groups of GP practices can offer inter-professional training to primary care staff, extending the skills base within general practice and developing a workforce which can meet the challenge of new ways of working.

This too will have no effect on the curves. New ways of working are likely to increase work stress at least in the transition (note there is no mention of transition funding), risking pushing us further down the slippery slope.

4 Targeted support

NHS England will work with the BMA GP Committee and the RCGP to explore a time-limited incentive scheme to offer additional financial support to GP trainees committed to working in specific areas for 3 years.

Once in a post after three years the Pay/Workload ratio is unchanged once the real work starts. Significant golden handshakes have not been very successful for post CCT GPs which rather suggests it is unlikely they will work for GPRs.

5 Investment in retainer schemes

NHS England will review the use of current retainer schemes and invest in a new national scheme, making sure it meets the needs of both GPs and practices. 

The Pay/Workload ratio is unchanged.

6 Improving the training capacity in general practice

The Government’s recent announcement that there will be an extra £1 billion for investment in new primary care infrastructure will enable increased training capacity and a more positive experience for medical students and foundation year doctors within general practice.  More broadly, NHS England will work with the BMA's GPs committee and the RCGP on the strategic direction of the primary care estate.

A nicer building does little to change the willingness to do the job. Current experience with GPR recruitment is not picking up despite the publicised investment and my City has several shiny underused buildings.

7 Incentives to remain in practice

NHS England and partners will conduct a detailed review to identify the most effective measures to encourage experienced GPs to remain within practice. Options may include a funded mentorship scheme, opportunities to develop a portfolio career towards the end of your working life, and a clearer range of career pathways.

The options suggested specifically do not address the Pay/Workload ratio. The options might decrease stress for the individual doctor but will increase stress on others as their contribution is reduced. This might help prevent things being worse than they would be otherwise but won't reverse the slide.

8 New ways of working

NHS England, HEE and others will work together to identify key workforce initiatives that are known to support general practice - including e.g. physician associates, medical assistants, clinical pharmacists, advanced practitioners (including nursing staff), healthcare assistants and care navigators.

Now this might work - but can only do so if it reduces the GP workload without decreasing pay, thus increasing the willingness to work. Unfortunately the recently released information on the Pharmacist Scheme where practices pay an increasing proportion of Pharmacist time means this is highly unlikely to happen.

9 Easy return to practice 
HEE and NHS England will publish a new induction and returner scheme.

See 2

10 Targeted investment in returners

NHS England will make available additional investment to attract GPs back into practice, increasing over time.   Targeted at the areas of greatest need, the scheme will offer resources to help with both the costs of returning and the cost of employing these staff.

Now this might work if they turn out to be serious in meeting the cost of employing these staff on a permanent basis but the numbers are not likely to be high. 

What is striking is there is absolutely no mention of pay or workload which are key to the willingness of GPs to do the job. They just don't want to talk about this Elephant.

So what will change the situation?

In order to increase general willingness to do the job workload has to be reduced. This will push us back up a curve.

I suspect too that it is this cooperation with such ineffective proposals that has contributed to a loss of confidence of a growing number of GPs in their representative bodies.

I'm afraid there really is no way out of this without increased investment in the primary care workforce (as well as buildings) which could be achieved within the NHS budget by stopping activity with poor evidence and diverting those resources. NHS Health Checks are a good place to start.

None of the current proposals reliably move us back up from Y to X and the GP contract is not set to change in a way that will address this.

Pay is set to continue it's downward drift and nothing significant is being offered to relieve the workload. Consequently the situation is set to worsen still, driving the profession down either a steeper cliff or to implosion. If JH does not realise this he is incompetent, if he does but does nothing he is either impotent or malicious - whatever #WeNeedToTalkAboutJeremy and the petition for a no confidence vote is entirely appropriate.

Mending primary care

So we know not for profit OOH is favoured by patients and that APMS practices on average deliver poorer quality care. So the evidence gives us something we should be moving towards (don't hold your breath!) - not for profit OOH and conventional primary care contracts but with an improved cost base from economies of scale while retaining an emphasis on optimising the access/continuity balance.

We desperately need to improve recruitment of GPs and nurses into a context where many work not just very hard but suicidally hard in a wholly unsustainable way. Working even harder is therefore not an option, so just paying more to do more won't work. Nor will back to work schemes without enough wanting to come back, golden hellos or unrepresentative assertions of rosiness without concrete action.

We have to get back to a non-punishing daily routine. I can still remember the only time I had a decent lunch-break in the last 4 years, which was on 16/5/2014. It was 45' long and I actually went home later in a better mood than I started the day for the first time in ages. I was still at work for 10 hours but was only actually working for 9 of them. This and better must become a daily experience if we are to attract folk.

A growing problem is that progressively more clinical and organisational responsibility is being carried by fewer and fewer shoulders. The financial and medico-legal risk run by many partners is no longer sustainable and also contributes to burn-out and early retirement. Further, the more people for whom an individual partner is carrying ultimate responsibility for, the less they are able to keep tabs on what is going on and the greater the anxiety. GPs face the doubly stressful work situation of both loss of control of their work and of control of clinical risk. We have to get back to a more even distribution.

And it is essential to guarantee any improvement going forward. No one is going to come into a profession when they think we will be back here again in 10 years when we have been here at least twice in the last 12.

Getting back to sensible working will happen because the current situation, ironically for health workers, is incompatible with their health and so will not last. The only real question is whether it's with or without a public health disaster en route. More of the same will collapse primary care as we know it and our NHS values along with it. Everyone has a limit and many of us could just walk away right now, taking experience and the training skills to produce the next generation with us, further depleting input at the bottom. Practices could tumble like dominoes as one takes out the next; some areas could be left with no primary care to speak of. The whole thing might be privatised and serious numbers of people would lose the right to primary healthcare as it becomes unaffordable as medical time is priced out of their reach.

This will eventually be OK for doctors though as price will limit activity and lunch times will be restored. Most GPs oppose this form of privatisation because they know only too well how badly large numbers of people will be affected. We could move to a US fiasco and public health and inequality go into reverse and still the overall costs escalate. Even if replaced by a comprehensive insurance system it will still be more expensive and very many will lose the kind of access to care they currently enjoy.

It's probably not too late for a more civilised route but I still have yet to see anything tangible. So what needs to happen?

Things that might help a bit or not at all are:-

Defining GP workload to limit demand. There was much talk about this around the 2003 contract negotiations and it was attempted but failed miserably. The role of the GP is constantly evolving and so as soon as you try to define it, it is out of date. It would require continuous re-working and is not worth the endless wrangling. And as we have discovered since 2003 'new resource for new work' just doesn't work.

Payment for activity. The main reason for cost effectiveness in the NHS is that doctors are NOT paid for activity; you end up with healthcare activity regardless of whether it is needed and this drives up costs. This will seriously damage NHS cost-effectiveness.

Co-payments for primary care appointments. This directly contradicts a core NHS value and discriminates against those most in need. The transaction costs are counter-productive anyway.

Cavalry. In the form of Physician Assistants or Nurse Practitioners. As effective as these can be there is no evidence they are cheaper over all and the jury is still out on the safety question when more serious disease presents and there aren't enough anyway. They increase the burden of clinical responsibility on the GP supervising (as if they have time!) their work.

Salaried Service. Many see the only hope is in a fully salaried GP service as they think the loss of autonomy will be compensated for by a contract that will have to offer acceptable working patterns. Personally, I think it is simply unaffordable in the current climate and will ultimately stifle innovation and patient advocacy, which is becoming ever more important as secondary care services return to 1990's waiting times.

So what will help?

We need a mechanism that is going to guarantee adequate, properly focussed funding of primary care and sensible work intensity. What is obvious is that no government can be trusted with this, otherwise we wouldn't be in this mess. We need a system where there is a powerful legal incentive for politicians to ensure that viable primary care continues, by containing demand and adequate resourcing. The current system provides neither.

Practices must be given absolute discretion over their list sizes. (I was pleased to see this passed as a motion at this years LMC conference and is a restriction that would be sensible to enshrine in law, Mr Osborne). By all means the practices should give warning they are about to cease taking on registrations and there should be some limitations eg they should close for a minimum of three months. It will focus minds properly on commissioning primary care if local politicians are faced with the realistic possibility of angry patients unable to register with a doctor. Currently, struggling practices are seen by NHSE primarily as the other local practices' problem when it comes to making up any deficiency in supply but not their own. Er, contractually it's not other practices' problem and nor should it be, the other patients are not registered and should not be automatically assigned. It's NHSE's (or jointly CCGs) responsibility, although practices may be willing to help, if able. The current system leads to lazy (non-)commissioning; it has to change.

The beauty of the above change is we do not need to worry about what a GP job is as the mechanism will cope with whatever it is, depending on the collective experience of coal face practitioners about a suitable list size.

Practice funding must reflect actual practice co-morbidity (NOT just age)

The current funding system actively discriminates against practices providing quality services that patients appreciate. How so? In an area where patients perceive a differential in quality between practices and vote with their feet, inevitably the high service users are the ones that switch practices. They take with them a higher than average workload but only an average capitation fee. Thus the 'good' practice becomes less profitable (greater workload per remuneration) and the 'poor' one more profitable. ie exactly the opposite outcome the patient choice agenda was supposed to achieve. In this inefficient application of NHS resource, it over-stresses popular practices while over-funding the less popular. Doh! (Current funding mechanisms do not take account of actual morbidity distribution between practices but smear an average deprivation factor over a geographical area.)

Evidence of this can be seen from QoF 2012 prevalence data from my patch of Leicester.

The coefficient of variation here is a measure of the variation of the disease prevalence distributions between seven practices serving the same geographical area of Leicester. There is, for example, a seven fold variation in the recorded prevalence of depressive illness between the practices, with the same pattern of variation for the other disease areas. Notice that there is much greater variation between practices for mental health / neurological conditions than for cardiovascular ones. As any GP will tell you the demand on GP time and emotional energy for the former is much greater than the latter.(1,2) We know also for example that depression associated with diabetes in the same patient is more time consuming for primary care than diabetes and depression in separate patients. The same is not true for hypertension and diabetes. (3). This variation, for reasons I haven't got space to go into here, is almost certainly down to prevalence variation reflecting patient choice / transfer rather than coding variation between practices.

A new funding mechanism based on actual practice multi-morbidity, assessed annually, is essential if meaningful competition and innovation is to be driven in primary care.

There must be more intelligent analysis of doctor behaviour. In exchange for discretion on list size GPs need to be willing to handle greater scrutiny over the areas that generate much secondary care cost, especially referral behaviour. Current initiatives such as the Unplanned Admissions DES simply don't work, but there are better ways.

For example there is a completely untapped seam of data which may give useful insights and potential savings here, as well as function as a useful educational resource. This is to audit GP referrals adjusted for the case load and case mix which form the background to this.

Years ago now we were struck by a 28 fold variation between different doctors in the crude count of their gynaecology referrals in our practice. I decided to adjust these referral rates by looking at the background of all consultations for gynaecology problems from which these referrals were drawn and adjusting for case load / mix and age. When you adjust for these factors the variation disappeared (moving from variation well outside 3 sigma to well within)!

Analysis of other areas such as neurology showed that less experienced GPs referred more patients for vague symptoms than experienced ones, offering an immediate educational and cost saving opportunity, that was not otherwise apparent.

I have been unable to repeat this work in SystmOne because incredibly despite 25 years of GP computing it is still impossible to create a list of all consultations by GP classified by the problem encountered even though our GPs have been recording every consultation properly!

The Doctors & Dentists Review Body (DDRB) recommendations must always be implemented fully by government with more than just a gentleman's (ha, ha) agreement to do so. Successive governments have shown themselves to be entirely untrustworthy on this. The DDRB in recent years has become a joke and ought to be experiencing some kind of existential crisis given the degree it is ignored by government. The deal that we don't take industrial action but DDRB recommendations are implemented must be restored by legal instrument and MPs must be obliged to take other peoples' review bodies as seriously as they apparently take their own.

The above combination should be able to ensure that workload is controlled at the practice level, remunerated according to need and funded adequately nationally. Importantly it will oblige engagement by reluctant politicians in an ongoing discussion about our NHS and what we want it for.

The main obstacle will have to be an acknowledgement of the repeated failure to manage primary care properly. It will be messy as I suspect a lot of practices will cease to take new registrations but the alternative will be much messier. It will cost because practices will have to be paid the same to do less to bring workload levels back from the suicidal. But it will cost less, be more equitable and a lot less messy than the alternative.

1. The workload of GPs: consultations of patients with psychological and somatic problems compared
Else M Zantinge, Peter FM Verhaak, Jan J Kerssens, Jozien M Bensing
British Journal of General Practice, vol. 55 no. 517 609-614, PubMed 16105369

2. Epidemiology and impact of multimorbidity in primary care: a retrospective cohort study
Chris Salisbury, Leigh Johnson, Sarah Purdy, Jose M Valderas, Alan A Montgomery
British Journal of General Practice, vol. 61 no. 582 e12-e21, PubMed 21401985

3. Implications of comorbidity for primary care costs in the UK: a retrospective observational study
Samuel L Brilleman, Sarah Purdy, Chris Salisbury, Frank Windmeijer, Hugh Gravelle, Sandra Hollinghurst
British Journal of General Practice, vol. 63 no. 609 e274-e282, PubMed 23540484