Playing the telephone game, health journalism style

6 February 2009

I followed the link under a rather astounding headline to this article, which reports that:

Nearly 13 percent of people in car crashes suffer spine injuries that could lead to paralysis or death

Right in the lead paragraph.  My first thought was:  my God, I haven’t been boarding and collaring nearly enough people!  Second thought:  I need to read the journal article that inspired this summary.

The actual data are far less scary (nice of them to make that available for free, by the way).  Of patients admitted to the hospital in Wisconsin, 12.5% had a spine fracture.  Some were cervical, some were lower, but only 8% were classified as “severe.”

The actual numbers (instead of percentages) are even less worrisome.  Over the 9-year study period (1994-2002 inclusive) there were just over 20,000 cases analyzed, with just over 2500 spine fractures–about 280 a year, of which just 22 were severe.  In 2002 (the earliest year for which I can find data), there were 129,000 total crashes.  That paints an entirely different picture, doesn’t it?

Bonus note:  the accompanying editorial also includes some misreading of the article.  The author suggests that “It is highly likely that the increase in the speed limits in Wisconsin in 1995 was an important cause of the increase in the incidence of spinal fractures between 1994 and 2002.”  Take a look at Figure 2 and tell me how unlikely that conclusion is given that the increase after 1995 is followed by a dip to below 1994 levels.  Even the authors admit that part of the increase may be due to better detection as spinal CT became more common.

Share



Leave a Reply