Commentary: Nudging hospitals to better quality

All developed countries are searching hard to eke more value out of the money spent on healthcare. First in line for analysis is hospital care—the most costly part of healthcare, accounting for 50% of costs in the NHS in England in 20011-12. One tool for controlling costs is the way that healthcare is paid for. Payments matter: they influence the services on offer, the viability of hospitals, the cost effectiveness of treatments, and how physicians practise medicine.

Western countries have been developing payments for hospital care over the past 20 years in similar ways. Busse and colleagues set out how and why payments for the care of homogeneous groups of patients (diagnosis related groups, DRGs) have evolved, their broad impact, and future likely direction.1

In the United States DRG payments curbed growth in Medicare hospital expenditures, partly through reduced inpatient stays.2 In Europe (including England) DRG payments have also been associated with reductions in the length of stay (and assumed technical efficiency), but admissions have increased as well. So the overall effect on efficiency depends on whether the additional admissions in European countries and the reductions in length of stay in the US were value for money. Firm evidence on both counts is limited.

Busse and colleagues explain how DRG payments are being developed across Europe and the US to cover more services, to promote quality, and to extend to whole pathways of care across several providers.

The NHS in England is pursuing all these routes. Payment by healthcare resource group (the English equivalent of DRGs) now covers about 60% of acute hospitals’ income. This includes elective, emergency, maternity, and outpatient care. The NHS is now seeking to extend the tariff payment system into mental health services. A mental health classification system has been developed and is now mandatory across the NHS but national prices for these case mix groupings have not yet been developed.

“Pay for performance” incentives have also been added to the payment system. The Commissioning for Quality and Innovation (CQUIN) system now operating in all hospitals, and “best practice” tariffs now cover treatments such as cataract surgery, hip and knee replacement, major trauma, and diabetic ketoacidosis and hypoglcaemia. These tariffs are not priced according to national average cost but are based on a national estimate of the cost of delivering best practice care. Similarly, hospitals are not paid at all for the care required as a result of “never events”—serious adverse events that should not occur if safe protocols are followed. Hospitals are also not paid for some readmissions. More controversially, hospitals are paid only 30% of a national tariff for emergency admissions beyond a baseline level occurring in 2008-09.3

But the most interesting area of development is perhaps payments across providers (including primary and social care) to encourage better coordination and preventive care, particularly for people with long term conditions. Some pilot schemes internationally are experimenting with giving groups of collaborating providers a “capitated payment” (a lump sum for a year of care for an individual) and allowing the providers to work out how it is spent and divided among them.4 In the UK, for example, there will be a year of care tariff for people with cystic fibrosis covering most care for patients by specialised care centres and networks of outreach facilities. Worries that this may encourage perverse effects, such as providers skimping on care to save money, remain, but the better scrutiny of quality now possible should help reduce this risk. Capitation budgets link payments to a provider’s ability to attract and retain patients, which works well in systems where patients can choose their provider and there are many to select from. However, in countries with less choice and competition capitation payment systems will have many of the problems of block budgets.

No one payment type is ideal—all have their strengths and weaknesses—and what is needed is an intelligent blend.5 4 To find that blend will take time, political nous, investment in information infrastructure, and careful evaluation. Achieving better quality at lower cost is clearly too complex an objective to be achieved through the design of a payment system alone. But payments are important for policy makers for two reasons: they can be altered and implemented quickly and they have an immediate impact. For these reasons, we should expect more evolution over time, not less.