Pay-for-performance dealt a major blow

A FAILED UK general practice incentive scheme for hypertension management has added weight to arguments against the introduction of pay-for-performance models in Australia, experts say.

Evaluation of a £1.8 billion ($2.9 billion) scheme involving nearly 480,000 patients with hypertension showed the incentives had “no discernible effect” on management or outcomes over three years. 

Substantial financial incentives – up to a quarter of a GP’s total income – did not improve blood pressure monitoring and control, or rates of stroke, myocardial infarction, renal failure, heart failure or all-cause mortality in the 358 general practices, the study found. 

The authors said the scheme may have been ineffective because some performance thresholds may be too low for the financial incentives to be effective. 

Alistair Vickery, professor of general practice at the University of Western Australia, said the results sent a word of caution to policy makers about introducing a system not backed by evidence. 

“Policies for improving health outcomes should be based on evidence, not belief, and this study provides some hard evidence,” he said. 

The results may bolster opposition to the Australian Government’s controversial $449.2 million diabetes management scheme, shelved in 2010 after medical groups strongly objected to the proposal.

The scheme involved annual payments worth an average of $10,800, which would be partly based on GPs’ performance in keeping patients healthy (MO, 19 November 2010). 

Professor Vickery said the UK results backed warnings from general practice on the diabetes scheme that improved clinical outcomes would be based on more factors than financial incentives, such as patient behaviour, self-education, socioeconomic status, access to continuity of medical care and professional training. 

Professor Simon Stewart, head of preventative health at Melbourne’s Baker IDI Heart and Diabetes Institute, agreed the results cast doubt over introducing a similar scheme in Australia. 

“Simply throwing money at a problem doesn’t necessarily correlate with good outcomes, and this is a classic case of maybe a good idea needing to be thought through a little more,” he said. 

“If someone is rewarded for putting more people through the mill – for example, more hypertensive patients – that is likely to have a decrease in quality of care because there is only so much capacity for GPs to achieve that,” he said. 

A better strategy, he believed, would be to link incentive payments for GPs to management processes as opposed to outcomes. 

GP payments could support absolute cardiovascular risk assessment, electronic systems and practice nurses to undertake routine management of patients with chronic diseases, he concluded.  

BMJ 2011; 342:d108