Shared decision-making is an essential - but often neglected - component of evidence-based medicine. In practice, doctors, nurses and patients encounter dilemmas which are understandable, but not insurmountable. We illustrate this by means of a complex decision concerning therapy in the advanced stage of Parkinson's disease. Evidence of effectiveness differs with each individual therapy, and, in practice, the experiences of neurologists also vary widely. In addition, the various treatment strategies all require a different sort of daily care, and have differing impacts on people's lives. It is time that we worked towards a realistic view of shared decision-making. Shared decision-making cannot be generalised to a 'one size fits all' strategy that we can choose to apply or not to apply at will. It should be a flexible instrument which, depending on the individual and the specific situation, can be put to use to deliver the best tailored care.