There have been a number of papers that have argued that have identified a post 9/11 uptick in traffic fatalities, theoretically based on a rise in driving, itself motivated by a fear/dread of flying (most notably, Gerd Gigerenzer, 2004, “Dread risk, September 11, and fatal traffic accidents,” in Psychological Science).
A new paper, “Driving Under the Influence (of Stress): Evidence of a Regional Increase in Impaired Driving and Traffic Fatalities After the September 11 Terrorist Attacks,” by Alexander J. Rothman, et al., also in Psychological Science, comes to a rather different conclusion.
“Although we confirmed that U.S. domestic air travel decreased significantly following September 11,” the authors write, “our analyses did not support the claim that there were notable increases in driving miles and in traffic fatalities across the United States after that date. In fact, total U.S. driving miles in the post-September months in 2001 did not differ significantly from total U.S. driving miles in the same months in 1999 and 2000, and the observed increase in total U.S. driving miles in October through December 2001 appears normative when examined within broader historical trends. The number of fatal traffic accidents in the United States did increase, albeit only marginally, in the 3 months following September 11, but there was no evidence of an overall increase in traffic fatalities.”
They did find one change amidst the data, however: “We did obtain evidence that the terrorist attacks had a systematic, but localized, effect on traffic fatalities.” The “localized” effect was on the Northeast, the region arguably the most directly impacted by the September 11th attacks. “Our analysis revealed a significant increase in traffic fatalities in the Northeast in the final 3 months of 2001.”
They continue: “To examine regional differences in traffic fatalities further, we used alcohol- or drug-related citations and reckless-driving citations as two behavioral indicators of psychological distress… [W]e found a significant increase in the number of alcohol- or drug-related citations issued in connection with such accidents during the last 3 months of 2001, but only in the Northeast. The concurrent regional increases in traffic fatalities and in alcohol- or drug-related citations lend support to our second hypothesis—namely, that behaviors impairing the quality of driving increased in those regions most affected by the terrorist attacks, and may have contributed to the observed elevation in percentage of traffic fatalities. This effect is consistent with other findings indicating that exposure to traumatic events is associated with an increased use of psychoactive substances, especially alcohol (e.g., Chilcoat & Menard, 2003; Pfefferbaum & Doughty, 2001)…”
Interestingly, they found the rate was the effect was highest in New York State, though they caution that “that meaningful operationalization of geographic proximity can be complicated and remains a task that is beyond the scope of this article.”
So, if correct, the study implies that it wasn’t a mere affect of people driving more miles to avoid airplane travel, but that their behavior on the road had in some way changed (one Israeli study found a similar increase in fatal crashes in the days following suicide bombings). My initial instinct was to think that a rise in drunk driving crashes might make sense from the perspective that more traffic enforcement officials were pulled off the roads and put into other duties in the wake of 9/11, although that wouldn’t necessarily explain the rise in citations. Another issue is to break down more specifically what kinds of roads people were driving on after 9/11, as Michael Sivak and Michael Flannagan have done, although, interestingly, this seems to potentially add weight to this study: Sivak and Flannagan found “the largest increase [in driving] occurred on local roads, not interstate highways that would be the main alternative to flying. Local roads, both urban and rural, accounted for 45 percent of the increase in traffic deaths.” Presumably, people swapping out flights for long-haul driving would be on those interstate highways, not local roads. As Rothman and his colleagues caution, “the rates of fatal traffic accidents, and hence fatalities, may have increased in the Northeast after the attacks as a result of more people driving in unfamiliar areas because of road closures and detours.”
Lastly, one can’t be certain that aggressive driving or impaired driving is a sign of “psychological distress.” Still, the pattern, localized in time and place, seems very real and suggestive.
(Horn honk to Shirl)