The trouble with healthcare choice
The common-sense of health reform is that offering greater choice is the way forward. The logic is that patients will choose the health services that are most responsive to their needs, and if money follows their choices, that will encourage providers to become even more responsive for fear of patients not choosing them in the future. The NHS will therefore be driven from the bottom-up to improve and providers of care (including GPs) forced to compete for their choices to create a virtuous circle of improvement.
The trouble is that this may work in an abstract, theoretical way (which is perhaps why economists like it so much), but it bears little resemblance to the reality of either the existing NHS or the proposed reforms to it.
How are choices made?
First we need to consider the environment in which most patient choices are made. They usually occur in a time-limited GP consultation or through a call-centre after a doctor appointment with little or no information to support patients in their choices. How patients are therefore meant to make their choice in these circumstances is not at all obvious – so they will tend to either ask their GP or nurse who they should choose, make a choice based on experience of another health service, or choose their provider based on other factors such as parking or accessibility through bus-routes. If a private provider is on offer, they may choose them, especially if they are going to be seen quicker than they would be publically as a result.
What this all means is that patients are effectively stepping into the dark when making choices, either asking their GP to make their choice for them, or making choices on non-clinical grounds. This is important because, if patients choose providers on the basis of them being private (and so perhaps thinking this represents better service for them), or because they have better car-parking (which is what many patients and clinicians tell me is happening) or because an appointment is available more quickly (which may not be an entirely good proxy of clinical quality), then patients are not choosing providers on the basis of what is likely to lead to the best clinical outcome for them, but on bases that may be little or not resemblance to good clinical practice. If this the case, then their choices will not drive clinical improvement.
Healthcare choice only works for patients if they are both willing and able to make choices in the way the government wants. The assumption behind policy is that patient choice is a learned transferrable skill – that we are living in a consumer society, and people want a greater say in decisions about their healthcare. But this assumption isn’t really borne out in reality. Some patients do want to choose both their treatment and who provides it, but there is little to stop them from doing this within the existing structures of the NHS where they can discuss what they would like with their GP and to come to a decision together. If patients are willing and able to choose, then they probably are already working with their GP to do so.
It may well also be that asking patients if they would like more choice before they are ill gives very different results to asking patients to choose when they actually need to – most of us tend to respond positively to the idea of more choice when we don’t actually have to act on it, but when feeling ill, or having to face difficult decisions which could have very serious consequences, we are far less likely to want to do so – this is one of the central observations of Barry Schwarz‘s ‘The Paradox of Choice’. It often turns out we want less choice in practice than we do in theory.
One area where there is the potential for GPs and patients to work more closely together is in the area of long-term disease. We need to respect the fact that patients with long-term diseases often acquire expertise about their reaction to treatments that is a valuable complement to clinical knowledge, and make sure this happens. Patients with long-term diseases acquire expert knowledge through repeated interactions with treatment and health services, but this contrasts directly with those that are making ‘one-off’ choices about secondary referral – and asks the question of how patients are meant to make choices about healthcare without either good information or experience of those health services?
What about GPs choosing instead?
So if many patients might struggle to make choices, can GPs make choices on their behalf to achieve the goals the reforms require? For this to happen GPs need information about potential providers in their local health economies, to be able to understand it, and to be able to find the most appropriate provider for the patient before them. This is a pretty demanding series of requirements. The information about the performance of particular services available to the GP will probably be difficult to come by, and even if directly comparable data is available, picking the best provider may still be difficult. It is far more likely GPs refer on an historical basis – sending patients on to services they have used in the past – as seeking out information about new providers (which may not even be available, by definition) is likely to be time-consuming and, even if the new provider appears to be better than the existing one, it requires the GP to trust a clinical team they probably do not know. Even apparently favourable statistics about waiting times, for example, may be suspicious – why is there a short wait for a service? Does it mean that other GPs know something and so are not referring to it? It is odd the policymakers have pilloried GPs for being reluctant to refer to new healthcare providers – from the perspective of a responsible GP, why would you want to send your patients to a clinical team you don’t know, and for whom you have no data?
The problem of information in healthcare choice
This leads to a bigger problem. Markets, generally speaking, work on the basis of a considerable amount of information being encapsulated by one measure – the price. In healthcare it might be possible (in theory) to create a market where price could capture a complex range of factors including clinical outcome, quality of service, waiting time, infection rate and recover time, but even if it could, we would not want to allocate care on the basis of only a few people being able to afford to see the best doctors with the highest prices. If we cannot use prices, we struggle to find an alternative indicator for high quality care because there are so many potential measures of it. The government has often suggested in the past that what we need is more information on which to make choices – but this isn’t the case. What we need is a reliable indicator of how good a provider is – and it may not be possible to construct one. And without such a measure, it is not clear how either the public or GPs are meant to make informed choices and drive improvements through them.
Could choice undermine public provision?
A further, and potentially even bigger problem is that patient choices have the potential to substantially undermine the entire basis of the NHS. Most public hospitals in the NHS are configured to provide a comprehensive range of services for their local area. Introducing competition is likely to lead to some services (such as orthopaedics, for example) facing greater new entrants. The services that face competition are those that the private and not-for-profit sector can provide, unsurprisingly, competitively. These services tend to be relatively straightforward (comparatively), avoiding high levels of initial investment. However, the public providers will still have to keep providing the complex services, or no-one will.
So would it be fine if the non-public providers did orthopaedics (and other more routine services) and the public sector was left with complex services? There are three big problems with this. First, public providers often depend on remaining comprehensive providers in order to meet their cost bases – they were designed to be big providers, and have to remain so. If they lose contracts beyond a certain threshold they become insolvent – and their closure could lead to a massive gap appearing in the provision of a health economy. This is especially likely to be true if the hospital is PFI, and so has made a commitment stretching off over the next thirty years in order to cover its capital costs.
Second, public sector provision is needed so that trainee clinicians can receive experience across clinical specialities while training. It is possible non-public providers can provide education as well, but whether it is likely to be of the standard that our teaching hospitals can give is debateable at least. Forging relationships between Universities and a range of private providers would be extremely difficult.
Third, where public care provision leaves an area entirely, we effectively lose the provider of last resort. Public care effectively comes with a guarantee – you may have to wait, but the hospital is unlikely to close without notice or without alternative care being made available. Private care comes with no such underwriting – and if the public sector has to underwrite it, which is surely the case where only private provision exists in an area, then this undermines the point of making use of the private sector in the first place.
In all then, the trouble with patient choice is that the messy reality of trying to make it work is often entirely lost in the abstract models that policymakers and economists appear to hold for it. It is not clear whether it is patients or GPs that are meant to be making choices, whether it is even possible to create a measure of provider performance that can be used to make choices, and what the cost of losing public provision of care in an area might be.
Healthcare choice sound like a great idea – but in practice it may come with a range of costs and problems that abstract analysis from policymakers and economists simply fail to take into account.