The GP bonus scheme will be reformed to better link GP earnings to outcomes for patients under proposals outlined by the Department of Health.

The DH has proposed that the National Institute for Health and Clinical Excellence evaluates which indicators should be included with the quality and outcomes framework from next April.

The proposal follows criticism from academics that there is "no relationship" between the pay GPs can receive under the framework and health gain for patients.

Bonus payments make up about 15 per cent of GP earnings and total more than£1bn.

A 2005 research paper published by the York Centre for Health Economics concluded that the framework could lead to "a danger that clinical activity may be skewed towards high workload activities that are only marginally effective, to the detriment of more cost effective activities".

In a consultation launched last week the DH proposes handing responsibility for reviewing the current framework indicators and assessing new ones to NICE, which uses the quality-adjusted life year to assess the value for money of treatments.

Health minister Ben Bradshaw said: "[This] will make sure we make the best use of our annual investment in the scheme and continue to support GPs in delivering the best care possible for patients, allowing the QOF to adapt and respond to the latest medical advances."

The DH has also suggested primary care trusts should be able to adapt the range of framework indicators they pay for to suit local needs. Royal College of GPs president Steve Field said that although he would welcome a role for NICE in the framework, the college would be concerned about giving PCTs too much flexibility.

"We need to ensure we don't go backwards on inequalities," Dr Field said. "We have concerns that not all PCTs have the competence to manage this sort of programme."

DH economists have estimated that allowing PCTs to select which QOF indicators they use locally could increase the number of practices failing to meet the payment threshold on selected indicators by as much as 78 per cent.

The consultation was launched as the NHS Information Centre published details of the rate at which GPs excluded categories of patients from the QOF in 2007-08.

There was a fourfold difference between different types of indicators with GPs "excepting" around 7 per cent of patients from indicators where the patient's health had to improve by a particular measure or whose specific treatment needed to be recorded.

The exception rate for indicators that were related to processes such as compiling a disease register was much lower, at 1.7 per cent.