Lessons from the world of payments

The payment system is increasingly seen as one of the major levers to change the way healthcare is provided. With responsibility for pricing architecture passing to the NHS Commissioning Board and the pricing function moving to Monitor, now is a good time to reflect on what lessons can be learned from the experience of the NHS and elsewhere.

Reforming Payment for Healthcare in Europe to Achieve Better Value, a new report by the Nuffield Trust, and produced with support from KPMG, gathers together learning from several European countries.

Limited tool

The first lesson is that policy makers overestimate the power of payment systems to fundamentally reshape the healthcare system. The payment system will lag behind what they want it to do.

It is a blunt instrument and can only send big signals and get in the way, but it will not provide an automated mechanism that will sort out difficult issues of health system configuration.   

It may even be limited as a tool to drive efficiency if providers do not have systems that transmit price signals to individual clinical departments. The lack of stability of the tariff in the NHS is an added problem as services can swing from profitable to loss making through movements in the tariff rather than their efficiency.

A related second lesson is that, compared with other countries, policy makers in the NHS have loaded more objectives on to the payment system than other systems.

Some of these have been mutually contradictory; for example, objectives to both increase and decrease different types of admissions.

Lesson three is that each type of payment system has its own set of unintended undesirable consequences. Diagnosis-related group payment systems tend to be good at improving provider efficiency but create incentives for unwanted extra treatment.

As a result, most systems have imposed mechanisms to protect payers from increasing volumes - particularly in emergency care.  The NHS was relatively slow to respond to this. Capitation systems create incentives to under treat, and disease-specific bundled payments can create incentives to cherry pick patients.

Fee for item of service payments encourages fragmentation, over-treatment and inappropriate failure to refer.

Poor alignment

The parallel evolution of payment mechanisms for primary care, outpatients and hospitals in many countries, including in the NHS, means that there is often poor alignment between them so that hospital specialists and primary care doctors have few incentives to work together to develop models of chronic care that might manage demand.

The growth of incentives for quality adds to this complexity, although outside the US and UK their adoption has been slow. The evidence for the effectiveness of these schemes is not very impressive so far. They may suffer from the same problem of trying to micro-manage providers from afar as using targets - there is a limit to how many it is possible to have, there are unintended effects and it encourages the optimisation of individual components rather than quality across the board.

There appear to be limits to the proportion of a provider’s income that can be sensibly linked to incentives. It is much less than the proportion accounted for by the general practice quality and outcomes framework, which has demonstrated that if too much is linked it simply becomes viewed as part of the provider’s baseline income rather than an incentive.

The German way?

Lesson five is that there is more to do to improve the quality of costing and pricing. One solution is to follow Germany and use sample cost data from providers with an accredited costing system. How to properly fund capital is a difficult issue that remains unresolved.

A number of countries do not include capital in the tariff at all. In the NHS, the component of the tariff that reflects capital cost appears to be insufficient to create sufficient reserves to allow for the replacement of the estate.

Lesson six is that new payment models such as year of care tariffs - the extension of tariff models into areas where it is harder to categorise patients - is harder and takes longer than expected. There is tendency for these systems to become increasingly complex as policy makers try to address these issues. Attempts are being made in a number of countries to create year of care and other forms of bundled payment. There is a point at which it makes sense to strip away this complexity and use capitation models with the DRG system used for budget setting and for transactions between capitated providers and for elective surgery and similar short self-contained treatments or investigations.

This too has downsides and so increasingly there will be a growth of blended methods of payment that reflect value for patients rather than just the cost of individual items of care.

Nigel Edwards is a senior fellow at the King’s Fund, and former acting chief executive and director of policy at the NHS Confederation. Downloaded the report here