I think the real problem is that many doctors also lack the capacity to translate statistical data into a rational evaluation of risk. Consider the infamous 'm

I think the real problem is that many doctors also lack the capacity to translate statistical data into a rational evaluation of risk. Consider the infamous 'mammogram statistics problem' as an example. So what do you do then, when nobody in the decision-making chain can be trusted to understand the data?