The Bradford-Hill criteria were originally developed for use with communicable diseases. Which of the criteria do you think are the most useful for exploring causation for non-communicable diseases? Explain why using an example.