The Scientific Doctor.

Medical professionals do not sufficiently realise how much scientific responsibility they take when certified science fails. They must make the best choice for the individual patient based on their own experience and that of their colleagues. This 'next best choice' requires basic knowledge about heuristic bias and statistics. The best solutions based on experience could then be classified according to their reliability and expected effectiveness. This requires probability and conditional probability calculations that result in Bayesian updating of the likelihood of individual cures. Practitioners could have considerable influence on medical decision-making by applying Bayesian probability. The high quality and large quantity of combined experience can ultimately be translated into treatment algorithms, to be tested in daily practice. To achieve this, we need better scientific training of practitioners and a suitable infrastructure of the professional community.