Of course, 37,485 prescription drug fatalities is bad. Likewise, a doubling of prescription drug deaths in a decade is bad. And, we should note, tens of thousands of traffic fatalities per year are also bad. But it is also worth noting that the two forms of death have nothing to do with each other, save for their potential to create a catchy journalistic hook.
So congratulations to the LA Times for finding that hook! Had the paper simply reported that close to 40K of our fellow citizens died of prescription drug overdoses in 2009, Americans would have been all, "Meh, what time is the game on today?" But by tying it to the well known and highly visible but completely unrelated auto death toll, Americans are all, "I drive a car! Am I next to die, from Oxycontin?" (Then they take Oxycontin anyhow, because mmm, so good.)
The real impact of the false statistical hook comes in all the re-reporting of this story by all the other media outlets (hi!). A quick check of Google News shows that virtually everyone who picked up this story went with the "drug deaths exceed car deaths" hook, rather than simply reporting on the drug death epidemic itself. There is a whiff of implication, then, that the drug death epidemic is important only as it relates to the death toll from other causes. The death horse race is what is really important. We cover death just like politics!
Not that this is a bad thing. The LA Times successfully got its story spread around the media, and as a result, many readers will probably become somewhat more cognizant of the dangers of prescription drugs. It's just fun to take a moment and realize that we, as readers, are so childlike and easily distracted that we must have our important news forced into the middle of a ball of journalistic cookie dough to make it palatable to us. Just like a prescription drug.