Dr Oliver Bernath is currently working with several PBC groups gearing up for real budgets. We asked him to share his key starting points
1. You must get the hospitals on board
PCTs have failed to put a cap on hospital spend and activity, so why should we assume GPs will have any greater success? PBC groups are even smaller organisations with less managerial capability and less statutory power than a PCT.
You cannot control what a hospital does. For example, it might decide to have a purge on waiting lists and do 13 months of work in 12 months – costing GPs an extra 10%. There is no legal way to stop them doing that.
In the real budget schemes we are getting off the ground we are tackling this by developing models where GPs will not take on the risk and accountability on their own. They share it with the hospital and possibly the PCT's provider arm (see box, page 36). The budgets will be devolved to GPs and hospitals together.
Both will share the incentives and risk and will work together to make pathways more efficient. Any underspends or savings will be split between the GPs and the hospitals.
Hospitals run at a very low profit margin and if you take away some of their income and put it into the community it takes them a long time to reduce their fixed costs. Clearly healthcare has to come out of hospitals in the long term, but this way the loss of income would be more gradual so they remain financially viable.
Clinical governance benefits too – hospitals and GPs are incentivised to share data, form joint teams and work in different ways to benefit from consultants' expertise, such as making a phone call, without having to pay the full outpatient tariff.
2. Budgets must be managed at a cluster level
The new funding allocation formula that has just been introduced functions well at a PCT level, or even for PBC clusters. But it gives higher budgets to those with lots of secondary care use, so for individual practices it has the effect of penalising GPs who keep their patients out of hospital.
The current formula would work for real budgets only if it were devolved down to a cluster level of, say, 10 practices or more. That would mean the natural variations and anomalies between practices could even out and the group could decide how to break down the budget to individual practices. This is one of the reasons why it is important for GPs to work together.
Controlling colleagues' behaviour is the crux of real budgets. GPs work very much in isolation. They don't really know each other's clinical styles. The culture is: ‘It's not my business how my colleague treats patients', so it's very difficult to come to an agreed minimum standard of quality and healthcare expenditure. But that's where we have to get to.
GPs resented it when performance management information was handed down to them from the Department of Health or PCTs and they were told they were doing things badly or spending too much. But if they are working together within the group they are keen to see how to do it differently.
GPs I work with have found it refreshing to have this open discussion. We hold performance management meetings every month. Many variations are natural. For example, a woman GP may see more female patients so have a higher rate of gynaecology referrals, but for other specialties it may come down to differences in clinical practice. I've found that GPs enjoy the discussions about referrals, and how others may have done something differently. Initial fears about intrusion into their professional practice disappear when they are striving towards the same goal.
If performance management is not at the heart of real budgets, the whole thing will become just a money-saving exercise – which will disengage every clinician.
3. Hard budgets should incentivise, rather than transfer risk
Hard budgets should be about incentivising GPs rather than transferring risk from PCTs to practices. GPs could not bear the financial risk of taking over the total healthcare budget of their population, as practices don't have the capital reserves. For example: say a practice has 2,000 patients per GP and an average total healthcare budget of around £1,400 per patient. A 10% overspend would expose each GP to a potential personal loss of £280,000 – which is unacceptable. Private risk underwriters are one option but that is not a blanket solution for the NHS.
The current ‘insurance pool' for the NHS is 50 million patients. Split this into smaller pools and the underwriting costs increase and will eat up any savings made.
The problem with the current payment systems is that they don't incentivise spending decisions to be made in the interests of patients and taxpayers . Payment by Results is about ‘do more, earn more'. GPs receive a fixed amount per patient; community health providers get a block budget. No-one is rewarded or penalised for the health benefits they deliver to patients or the value they achieve for the taxpayer.
Incentivising GPs has to strike a balance between covering the additional costs of achieving ambitious goals and offering a meaningful reward. The risk of incurring a penalty for failing to achieve goals should be big enough to be felt by the individual GP but not to threaten the viability of the practice or the livelihood of any individual. So, the level of risk – in terms of the running costs to the GP in creating efficiencies – ought to be in the range of £5-£10,000, depending on circumstances. So, for example, a well-established GP whose personal finances are stable could prefer to take on a larger risk in return for greater rewards. A young GP who has just bought into a partnership, moved to the practice's area and had to take out a mortgage may wish to keep to a much lower level of risk. The GP would recoup those costs only by making sufficient savings.
The incentive scheme must allow practitioners to personally earn more for excellent performance (excellent clinical outcomes, high patient satisfaction and high value for money) and for taking on the risk of a penalty. The size of the penalty must be reflected in the potential size of the reward. If an incentive scheme could end up paying out excessive rewards to GPs, a cap could be agreed that would limit how much could be taken as personal income. Any rewards above such a cap would have to be reinvested in patient care.
Oliver Bernath is managing director of Integrated Health PartnersDr Oliver Bernath Some initial questions to ask your PCT
What is the baseline of the budget?
Does it include any growth assumptions or underlying trends?
Does it include reserves – is the PCT keeping some in its back pocket?
Is it a truly devolved budget or is it just part of a budget?
If the PCT topslices money for innovation funds then keeps a reserve, this sets the budget artificially low and gives an unrealistic savings target for GPs.
It is obviously important to have a reserve, but it makes more sense for the PCT to give as much as they can as a budget and for the GP groups to ensure they keep back some of it as a reserve.
Define what a saving is, and what it is not, in advance.
Different models for different areas
Establish what will happen about performance management.
As PCTs would be the ‘paying customer', if they continued to manage the performance of the PBC
groups, that would be a conflict of interest. Therefore GPs would have to handle any issues of performance management themselves.
Whoever manages performance would also need to ensure they have accurate data in a quick enough time to be able to make relevant operational changes.
One to one
We are currently working with several clusters and most of these deal mainly with one hospital.
Both the hospital and the PBC clusters would sign up to the PBC budget and any underspend would remain within the joint venture, split between the GPs and the hospital.
The joint venture allows resources to be shifted around. For example, for hip joint replacements time spent in hospital could safely be cut from five or seven days to one or two, creating substantial savings. The hospital could use its share of the savings to cover some of its fixed costs until it is able to scale down these costs.
So a one-to-one match between the PBC group and the hospital – a typical rural model – is fairly straightforward.
Several clusters refer to same hospital
There is still an incentive for the hospital to be involved as a venture creates more capacity for it to increase activities it can perform for other clusters.
Clusters refer to a number of different hospitals.
Where practices have several hospitals to refer to – typical of London – the joint venture concept becomes more complex.
Joint budget accountability could be created between practices and all local A&Es and hospitals, but limited to emergency care, including admissions. Elective care may be organised in a pathway-specific joint venture while the overall budget accountability remains with the GP practice groups.