In healthcare, robots are increasingly being used to provide a high standard of care in the near future. When machines interact with humans, we need to ensure that these machines take into account patient autonomy. Autonomy can be defined as negative autonomy and positive autonomy. We present a moral reasoning system that takes into account this twofold approach of autonomy. In simulation experiments, the system matches the decision of the judge in a number of law cases about medical ethical decisions. This may be useful in applications where robots need to constrain the negative autonomy of a person to stimulate positive autonomy, for example when attempting to pursue a patient to make a healthier choice.