Going to the doctor is, in many ways, like visiting a car sales yard. The customer has a limited knowledge of the product and the supplier may have a financial incentive to over-service or overcharge.
Of course, the main difference is trust. Most people trust their doctor to do the right thing and provide appropriate medical care. But, if you throw incentives into the mix, where does this leave patients?
Doctors’ pay
Australian GPs are paid largely through a fee-for-service system, the Medicare Benefits Schedule. Doctors can set their own fees but the government determines how much they (or their patients) will be reimbursed for a consultation.
Because the rebate for a standard consultation is fixed, it doesn’t matter if GPs spend six or 16 minutes with a patient – the payment is the same. So there’s an incentive to see a lot of patients but not necessarily provide the best quality of care.
In an attempt to address this problem, Health Minister Nicola Roxon last year announced an ambitious scheme to change the funding for diabetic patients. Rather than a simple fee-for-service, the proposal included capitation funding (where a clinic would receive a grant for each diabetic patient wanting to enrol) with a controversial pay-for-performance element.
Doctors would be rewarded with performance payments if their diabetic patients achieved certain health outcomes, such as good glucose control.
But Australian Medical Association opposed the plan and forced Roxon to abandon the scheme only a few months later. It’s now been replaced with a small pilot study, which is currently under way.
Performance payments
There are already limited elements of pay-for-performance in the Australian health system and larger initiatives are in place in the United States (the Integrated Healthcare Association’s payments P4P program) and the United Kingdom (the Quality and Outcomes Framework or QOF).
One motivation for these schemes is the recognition that as rates of chronic disease grow, and hospitals become more burdened, patients will increasingly be cared for in a general practice setting.
Currently, 80% of over 65s and more than half of over 45s have at least one chronic condition such as diabetes, heart disease, asthma and cancer.
Around 5% of all hospitalisations are potentially preventable complications of chronic disease, and over half of these are related to diabetes. So there’s a clear need for doctors and nurses to improve the care they provide in general practice.
But are performance payments the answer?
Recent research from the United Kingdom confirms GPs who perform better in diabetes care measures such as blood glucose control have fewer of their patients admitted to hospital.
But it’s still unclear whether the prospect of performance payments is driving GPs to provide better care to patients with other chronic diseases.
Incentives review
We reviewed all the available evidence on pay-for-performance for primary care doctors worldwide and found there wasn’t enough evidence to either support or oppose these schemes. The review was published last month in the Cochrane Collaboration.
Some of the studies we examined showed incentives had modest positive effects on the quality of care, such as increased rates of cervical screening and counselling for smokers. But no improvements were achieved for other outcomes.
Overall, there was a shortage of good data to enable us to effectively evaluate the schemes. There was also a high risk of bias in all the studies because doctors could choose whether or not they took part in the scheme.
Bigger picture
The current Australian government pilot incentive scheme is very limited in scope, focusing only on diabetes patients.
A more ambitious plan would be to pilot a wider-ranging pay-for-performance scheme encompassing more chronic diseases, patient enrolment, block (capitation) funding grants, and performance payments. This would, in effect, be an extension of the original, abandoned proposal.
Rather than “recruiting” GPs who want to take part in a study, a more radical approach would be to choose a pilot state and implement the scheme in full to all GPs in that state.
But first we’d need to improve the quality of data collection in all GP practices, potentially by building on the primary care collaboratives model, to ensure it could be accurately evaluated. Monitoring and evaluation is the only way to know how pay influences performance.
Whatever route the government takes on pay-for-performance in general practice, it’s important that improvements in care, rather than just the level of achievement, are rewarded. This will ensure funding goes towards raising the minimum standard of patient care.
The government must proceed with caution, but an expansion of the diabetes coordinated care pilot to other disease areas would be a good first step.