Cell Phone Laws Fail to Reduce Crashes
Following along the same lines as Monday’s post, let’s talk about one specific cause of distracted driving, cell phone use, and whether efforts to curb the danger it causes are working.
Currently 7 states (CA, CT, NY, NJ, OR, UT, WA) and Washington DC have bans on hand-held cell phone use for all drivers. Another 17 states (AR, CO, DE, IL, IN, KS, LA, ME, MD, MN, NE, NC, RI, TN, TX, VA, WV) have partial bans covering drivers under certain circumstances, such as drivers under 18 or drivers in school and construction zones. For a map of these bans, click here.
Recently released research indicates that these bans may not be effective
The Highway Loss Data Institute (HLDI) released a report a couple of weeks ago (Jan. 29, 2010) saying that laws against talking on a hand-held cellphones while driving do reduce the number of drivers talking on hand-held cellphones, but they have failed to reduce the number of crashes on our roads. Links to the full report, plus the press release kit and what appear to be slides from the director’s presentation to an industry group in Washington DC are all here.
The HLDI tracked collisions that resulted in insurance claims in four areas that were early adopters of hand-held cellphone bans: New York state (2001), Washington DC (2004), Connecticut (2005), and California (2008). They compared the these claims to claims from crashes in neighboring states that did not have hand-held cell phone bans at the time. These control groups help to rule out other factors that could have caused an increase or decrease in the number of crashes in a particular area, like winter driving conditions or high unemployment keeping workers at home. Collisions were tracked per area for up to 36 months before the ban, and up to 28 months after the ban.
On the right is the graph for New York. Collision claims are tracked for 22 months before the ban and 24 months after the ban, and compared to collision claims in Connecticut (this was before Connecticut’s ban), Massachusetts, and Pennsylvania. As you can see, there is variation in the number of reports by month, but New York seems to vary in the same way the surrounding states do. And the cell phone ban (blue vertical line in the center of the graph) did not decrease the frequency of insurance claims for auto accidents in New York (blue) compared to before the ban or compared to surrounding states without bans (averaged for all three in red) at least not within the two years after the ban was put in place. The graphs for the other states and Washington DC look similar (check them out in the original data linked to above). Similar rates of collision reports before, and very little change after.
Why the disparity between the expected result and the actual result?
According to the press release the Insurance Institute for Highway Safety (IIHS, partner to the HLDI) had enough data on hand-held cell phone use in collisions to say with confidence that driving while taking on a hand-held phone was a major distraction, and that it contributed to a significant number of collisions resulting in insurance claims. The press release also said that all four areas reported significant decreases in observed hand-held cellphone use by drivers after the bans went into effect, but no decrease in observed use in neighboring states without bans. The question is, who was observing and how reliable is the information? Police? Paid surveillance? They don’t say. But for now we are lead to believe that people are obeying the laws, although it is possible that some drivers are still taking on hand-held phones and the survey did not catch them.
So what’s the deal? If hand-held cell phone use was contributing to a large number of crashes, and now fewer people are taking on hand-held phones while driving, why hasn’t there been a decrease in the number of crashes?
One possible answer could be the method used to calculate collision frequency. The survey only included newer model year cars, that is cars that were 0-3 years old in the year surveyed (the year 2008 data contains only cars from model years 2007, 2008 and 2009). Obviously there are more than just new cars on the roads, and as the IIHS pointed out in an earlier publication, the driving behavior of drivers in older model cars can be different from drivers in newer models. Think teens. They are often given the old car when Mom and Dad buy a newer one, so a sample of newer model cars is unlikely to include teenage drivers, who may be more likely to chat on a cell phone while driving. The survey also didn’t collect data on the causes of individual crashes, so it can’t take into account hands free cell phone use or other distractions like texting. Is talking on a speaker phone just as distracting as talking on a hand-held? Are people texting in their laps instead of talking with the phone to their ear, and so avoiding being observed using a cell phone? We don’t know, but it’s another possibility.
Clearly this study did not turn out as the IIHS expected, but being good scientists they released the data anyway. Now they are looking for the reason(s) for the difference between the results they expected and the results they got.