In many ways Payment by Results has been a tremendous success. While its implementation and roll out was not completely problem-free, the payment system did not fall over. The significant technical issues that had to be addressed – from defining the ‘product’ to designing the fixed tariff without causing instability across the NHS – were substantial.
And there have been measurable benefits from PbR. Formal national evaluation showed that it had succeeded in boosting activity, reducing lengths of stay, and encouraging more day cases rather than inpatient work. At a minimum it appears not to have reduced quality of care. The financial incentives embedded in PbR have also encouraged hospitals to improve the quality of their own data systems.
But the more we have got to know about how PbR and similar payment methods in other countries operate in practice – coupled with big changes in the economic, financial and policy context that have occurred since PbR was originally designed – the more there seems to be a case for a rethink.
As we found in our assessment of PbR published in November last year, a number of general conclusions should inform this thinking.
First, payment systems cannot do everything: they are one of many managerial, policy and financial levers that can be used to achieve change. Second, one size does not fit all. Different services need different ways of paying providers in order to meet different sets of objectives. Related to this is a third point, that any payment system needs to be flexible – to deal with unexpected shocks, or unpredicted outcomes.
There needs to be flexibility too between national rules and frameworks and local discretion and experimentation. Importantly, different types of care and different patients – from a knee replacement operation for the otherwise healthy 50-year-old to ongoing episodic care for a 75-year-old with multiple long-term needs – will require different payment approaches to give the right incentives to providers to deliver high-quality cost-effective care.
Finally, there is a need to understand more about how different payment systems operate in practice in different situations. Further developments in payment approaches will need to be supported by high-quality data and analysis or they will lack compliance and risk leading to unintended and unwanted side effects.
As with medicine, at a minimum, payment systems should do no harm. But with careful design they could do some good.