Recent years have seen an upsurge of interest among public health experts in what’s known as unwarranted clinical variation. It’s a problem that arises when similar groups of patients or populations are treated for the same conditions with different medications or procedures for no clear, evidence-based reason.
One example that has got a lot of attention is the rate of antibiotic dispensing in Australia, which is almost double that of Canada’s. It’s theoretically possible that Australia has a much higher burden of bacterial infection than Canada, which would make this variation warranted. But studies have shown this is not the case. In fact, a third of antibiotics prescribed in Australian hospitals have been deemed inappropriate.
Unwarranted clinical variation plays out across different regions of the country, where people in rural and remote areas may not get the same quality cancer care as those in the cities. And it plays out across genders. Recent research using data from the Sax Institute’s 45 and Up Study shows that treatment of heart attack varies considerably between men and women, with men more likely to receive guideline-recommended treatment than women.
Unwarranted clinical variation can lead to undertreatment or overtreatment, can compromise health outcomes and wastes resources. So what can be done about it?
The paper – a synthesis of evidence from 27 publications – focuses on feedback approaches to solving the problem. The idea is to document evidence of unwarranted clinical variation with the aim of helping clinicians change their decision-making behaviours. Does this work? The authors say there is value to this approach, but ironically there is plenty of variation in the methods used and not enough evidence to suggest which is the best.
Approaches in the reviewed publications ranged from simply presenting evidence to clinicians, teams and organisations, to facilitated, tailored feedback that may be integrated into broader projects to actively engage clinicians in improving care.
One promising strategy is to incorporate health IT decision support tools into clinical practice. Two of the reviewed papers addressed the overuse of imaging in lower back pain, reporting on a tool which included mandatory peer-to-peer consultation before ordering a new test, along with quarterly practice variation reports to providers. The intervention led to a 33% decrease in the use of MRI testing.
The authors say the use of health IT in countering unwarranted clinical variation will continue to grow, and that there are other opportunities that could be exploited, such as recording patient preferences as a way of understanding and reducing variation.
They conclude that the next step is to assess the effectiveness of different approaches to providing feedback and to identify whether there is any particular method that is more likely to drive the change that is needed.