Why do we use evidence-based policy?
The reasons for the use of evidence-based policy should be obvious to policymakers. Anyone in a position to make decisions should look to do so with the most robust evidence for them.
In some cases robust evidence may indicate that some policies are not only ineffective, but also actively harmful. For example, the Scared Straight programme was shown to increase the likelihood of recidivism and the use of Critical Incident Stress Debriefing has been shown to increase the likelihood of Post Traumatic Stress Disorder.
But ending a policy can sometimes be just as hard – if not harder – than starting one. So how can we get better at stopping practices that are ineffective or harmful?
Vinay Prasad and John Ioannidis have recently written a paper which helps provides an answer in relation to the “de-implementation of contradicted healthcare practices”. In this paper they outline a conceptual framework to guide and prioritise the process of abandoning ineffective medical practices and mitigating the risks of untested practices.
This process involves careful inspection of the prior evidence base, paying special attention to the cost effectiveness, the documented harms and the comparative benefits of an intervention above alternatives. If the evidence base is sparse, then one should look at using a trial to establish whether an intervention works with an open-minded perspective, in a similar way to ourTest, Learn, Adapt framework.
The paper by Prasad and Ioannidis highlights the importance of using evidence-based policy and serves as a good reminder of the dangers of not using evidence to inform policy.
John Ioannidis’s previous work also deserves a mention: his two classic articles on why most published results are false and on contradicted studies are both well worth reading.