Thanks, Marijke, for your appreciation. The scenario that you describe is inevitable in the way we practise health care. We base it on research that uncovers the effects of drugs or interventions on groups of people. Nothing wrong with that, because we have no other way to know whether statins for example, lower cholesterol. The problem lies in the translation of group based results to a given individual. While medicine pretends that this translation is possible, as statisticians we know that it isn't.