I would like to comment, but I have not read all the threads as funny as the nonsense is I would like to give my opinion of the stats.
As Simon said in a long winded way..................there are too many variables to get a reasonable sample of data to extrapolate any meaningful stats.
One example of this is...............how do we know if the person would have died anyway with blood?
Personally any stats like this would never be credible to an experienced analyst, or statistition.