Some non-musical facts:
Dr. John Ioannides, an epidemiologist at the University of Ioannina, in Greece, has noted that four of the six most frequently cited epidemiological studies published in leading medical journals between 1990 and 2003 were later refuted. Demonstrating the malleability of data, Peter Austin, a medical statistician at the Institute for Clinical Evaluative Sciences, in Toronto, has retrospectively analyzed medical records of the more than ten million residents of Ontario. He showed that Sagittarians are thirty-eight per cent more likely to fracture an arm than people of other astrological signs, and Leos are fifteen per cent more likely to suffer a gastrointestinal hemorrhage. (Pisces were more prone to heart failure.)
To help strengthen epidemiological analysis, Sir Austin Bradford Hill, a British medical statistician, set out certain criteria in 1965 that indicate cause and effect. Researchers must be sure that exposure to the suspected cause precedes the development of a disease; that there is a high degree of correlation between the two; that findings are replicated in different studies in various settings; that a biological explanation exists that makes the association plausible; and that increased exposure makes development of the disease more likely.