Stats, drugs and rock 'n' roll

The death of Amy Winehouse is terribly sad, if not wholly shocking.  She was a gifted, fascinating, and ultimately tragic talent, and the music world will miss her.  One thing I do find slightly annoying, though, is the claim that her death is proof of 'the curse of 27', whereby pop stars die at that age.

The famous five generally included in the '27 Club' are Brian Jones, Janis Joplin, Jimi Hendrix, Jim Morrison and Kurt Cobain, with Robert Johnson and Alan Wilson also mentioned.  Now the unfortunate Ms Winehouse will be added to the list.

But is there really anything significant going on?  A study by a group at the Centre for Public Health at Liverpool John Moores University showed that, on average, European and North American rock stars die at a younger age than an average member of the general populace of those regions.  Of 1064 pop stars examined (in 2005), 100 had died, indicating a mortality 1.7 times higher than for non-rock-stars.

North American pop stars were about twice as likely to die young as European ones, but the median age of death was 42 for North American stars and 35 for Europeans.

This means that Elvis achieved the median age of death for his regional cohort, whilst John Lennon lived five years longer.  Of course, neither of them died at 27, and neither did Sid Vicious or Buddy Holly or Eddie Cochran or Marc Bolan or Keith Moon or Freddie Mercury or countless, countless others.  Interestingly, though, the modal age of death was 27, with fourteen rock stars having met their maker at that age.

14 out of 100 is a fairly high value, though I'm not enough of a statistician to tell you if it is genuinely significant.  Turning it on its head, 86% of dead rock stars didn't die at 27.  In fact, if dying at 27 was such a curse, why did it happen to only 1.3% of the 1064 stars in question?

And furthermore, the theory all depends on whom you classify as a star, and whether their death was in any way unusual.  The roster of deceased musicians on includes all sorts of people you've never heard of, whilst another lists the '8 most infamous rock star deaths' and only one of them - Brian Jones - died at 27.

To me, the key problem is that once a 'pattern' has been spotted, people find data that fit it.  It's rather like believing horoscopes, or finding Nostradamus' writings prophetic after the event.  One particularly bizarre claim was made by an author of a book on the 'curse', who stated that, "I think it’s a continuing story. I’m still trying to process Jade Goody, the British reality star."

Jade Goody wasn't a pop star, she was an unfortunate, ordinary young woman who became a celebrity thanks to demonstrating high levels of ignorance on two versions of the same TV show.  She died of cancer.  The fact she was 27 has nothing to do with anything.

For a theory to be meaningful it has to be predictable and testable.  At the moment, all we can say is that quite a few pop stars have died aged 27, that far more have died when they weren't, and far, far more haven't died at all.  Rather than getting all mystically distracted by a spurious hypothesis, people should be looking at the real reasons Amy Winehouse passed away so far before her time, and how we might be able to stop it happening to other vulnerable or unstable young people.