There've been reports on radio and TV news today (every bulletin I've witnessed, e.g.) about an Imperial College London study (published in the BMJ) of the relationship between the day of the week on which surgery takes place and the likely outcome, indicating that patients operated on towards the end of the week have less chance of survival than those operated on on Mondays or Tuesdays.
Having heard a surgeon on a BBC Radio 4 report this morning, I think the statistics need further analysis. He stated that he purposely scheduled surgery with more likelyhood of a bad outcome towards the end of the week, on the basis that he (and other medical practitoners) would be more likely to be available for emergency call-out during the weekend if emergency aftercare was needed, as they wouldn't be in theatre performing scheduled operations (as they would during the week), and intensive care units would be less busy.
If this surgeon is not atypical, the data could be skewed by the types of operation undertaken vs. day-of-the-week. Perhaps another example of confused causality?