Performance incentives for hospitals seem to have little effect

13 February 2018

Do financial incentives help to improve hospitals’ performance? A large-scale study into the American ‘value-based purchasing’ programme for hospitals reveals that rewarding results has not led to better quality of care in the United States. Its effects are ‘limited’ and ‘disappointing’. This was the conclusion reached by Igna Bonfrer and her colleagues in an article published in The British Medical Journal (BMJ). Bonfrer worked at the Harvard T.H. Chan School of Public Health for a year with funding provided through NWO’s Rubicon programme.

Medical personal checking the medical devices. Photo: ShutterstockPhoto: Shutterstock

Healthcare does not always lead to better health for patients due to the suboptimal organisation of healthcare systems. Because of that, rewards which depend on health care outcomes have been devised in the United States and elsewhere, in the hope that this would lead to better care. Have they helped? A study conducted at Harvard University into 1189 American hospitals revealed that long-term participation in a programme of that nature made hardly any difference in terms of process improvement or mortality.

The research results related to a ten-year period involving data from 1.4 million Medicare patients – a social insurance programme dating from 1965 run by the American federal government – aged 65 and older. There are growing calls to reward desirable healthcare outcomes in the Netherlands too. 'This means our findings are important for the future of Dutch healthcare,' says Igna Bonfrer.

Chronic heart failure

Bonfrer compared 214 American hospitals which voluntarily began rewarding better healthcare outcomes many years ago with 975 hospitals which started much later, when Obamacare was introduced. She looked at clinical process scores, patient satisfaction and mortality in a number of disorders, such as chronic heart failure. Although processes and outcomes improved during these ten years, the improvements in the hospitals which had voluntarily started earlier were no greater than those in hospitals which had started using value-based payments much later.

Bonfrer: 'And that is disappointing for such an immense programme. We suspect that the limited size of the reward played a role in this – a maximum of two percent of the total Medicare revenue and only for a limited set of disorders: a pittance. At the same time, a large number of indicators were used to measure outcomes. It would probably be better to increase the reward and reduce the set of indicators to just a handful of those which are important to patients.'

The way in which the system of value-based purchasing was introduced in hospitals in the United States in one fell swoop was unwise. 'It would seem that no lessons had been learned from the voluntary programme. It’s better to start by experimenting on a small scale to see what does and doesn’t work. That’s what we’re now going to do in the Netherlands with a number of Value-Based Health Care initiatives.'

Bonfrer is therefore arguing for a number of carefully designed experiments based on knowledge gleaned from abroad in order to be able to assess whether it would be wise to extend these activities nationally.

Further information

Igna Bonfrer (1986) conducted the project ‘Pay-for-performance to improve the quality of health care in high and low income countries’ at Harvard University with a grant from the Rubicon programme. She is a university lecturer in Global Heath Economics at Erasmus University Rotterdam and the Harvard T.H. Chan School of Public Health.



Source: NWO