Washington Post Writer Horrendously wrong, misleading readers!

Washington Post staff writer/blogger Lenny Bernstein has made a fool of himself with an article formerly titled "Cities with bike share programs see rise in cyclist head injuries".

Streetsblog USA captured the headline before it was altered

The basic premise of the article is that for rider safety, bike share programs should provide helmets with the bikes. Lenny "backs up" this assertion by citing Public Bicycle Share Programs and Head Injuries by Janessa M. Graves, Barry Pless, Lynne Moore, Avery B. Nathens, Garth Hunte, and Frederick P. Rivara

As many have been quick to point out, Lenny doesn't seem to understand how statistics work. Lenny says:

In the first study of its kind, researchers from Washington State University and elsewhere found a 14 percent greater risk of head injuries to cyclists associated with cities that have bike share programs. In fact, when they compared raw head injury data for cyclists in five cities before and after they added bike share programs, the researchers found a 7.8 percent increase in the number of head injuries to cyclists.

When they looked at the same data for five cities that don’t have bike share programs, the number of head injuries had declined a bit, by 2.3 percent.

Sounds scary. The point seems clear, if you are in a city with bike share, the number of injuries, and especially head injuries, is on the rise! Problem being, that some people can actually read and understand academic papers, and Lenny was quickly shot down by Kay Teschke, a University of British Columbia public health professor:

In fact, the study showed that all injuries, including head injuries, decreased in the 5 BIKE SHARE cities after implementation. All injuries went down from 757 to 545 per year (down 28%). Head injuries went down from 319 to 273 per year (down 14%). Moderate to severe head injuries also declined from 162 to 119 per year (though some were unclassified as to severity). 

 In comparison, in the 5 NON-bike share cities in the matched time periods, all injuries increased slightly from 932 to 953 per year (up [2%]). Head injuries in the NON-bike share cities decreased slightly from 356 to 342 per year (down 4%). Moderate to severe head injuries increased from 181 to 192 per year (though once again some were unclassified as to severity).  

 ...

I did correct for the 2-year time period pre implementation. If you check table 2 of the scientific paper, you will see I used the actual numbers reported for the post-implementation 12 month period, but for the pre implementation period, I divided all the numbers by 2 to account for the 24 month period. 

Oh, it seems really clear that Lenny may be intentionally leaving out the numbers to make the percentages more frightening. Cities with bike share saw a huge DECREASE in injuries overall. It seems that head injuries decreased at a slower rate, meaning that as a percentage of total injuries, head injuries increased.

So to me it is pretty cut and dry, Lenny didn't understand a statistical analysis, and so focused on a increased percentage to "make his point". Well, after being so obviously wrong, the article must have been taken down. Nope! Lenny changed his title to "Proportion of head injuries rises in cities with bike share programs"

Oh Lenny

Rather than admit that he was wrong, Lenny is desperately holding onto his original premise and hoping you don't actually read his references. This is such appalling BS that is clearly meant to catch headlines through scare tactics.

So let us be clear. Bike share lessened the number of injuries within the 5 cities studied. The cause of this decrease is debatable, but the decrease in ALL INJURIES in very apparent. Lenny still argues that the article is valid because the percentages are increasing, but then falsely asserts that it means you are more likely to get a head injury in bike share cities than in non-bike share cities.

For more terrible reporting, follow Lenny on the twitter box @LennyMBernstein