The structure of the network – which corresponds to the causes of SPADs – and the strength of each cause is judged by experts. So far we have used our own knowledge to build the model – to complete it we need some experts. There is data about the occurrence of SPADs, but of course that is aggregated – it does not breakdown the contribution of each cause. It is clear that there will never be enough data for this, so we need to rely on expert judgement. However, the data can be used to validate some of these judgements and to calibrate the model.