American Journal of

Public

Health

Electronic Letters to:
COMMENTARY:
Leon S. Robertson
Blood and Oil: Vehicle Characteristics in Relation to Fatality Risk and Fuel Economy
Am J Public Health 2006; 96: 1906-1909

Electronic letters published:

"Data-based science"
Leon S. Robertson   (1 December 2006)
US Traffic Safety Failure -- Data-Based Science Demolishes Robertson's Denial
Leonard Evans   (28 November 2006)
Evans' rhetoric and reality
Leon S. Robertson   (20 November 2006)
Comments on Robertson's Response
Leonard Evans   (16 November 2006)
Response to Evans
Leon S. Robertson   (14 November 2006)
Traffic deaths and fuel use – overwhelming influence is public policy, not vehicle characteristics
Leonard Evans   (7 November 2006)

 
"Data-based science" 1 December 2006
Leon S. Robertson,
Research Scientist
Yale University (retired)

The NHTSA report to which Evans refers says: “MY 1996-99 pickup trucks and SUVs had, on average, higher fatality rates than MY 1996-99 passenger cars or minivans of comparable weight.” Readers may examine the report online at: http://www.nhtsa.dot.gov/cars/rules/regrev/evaluate/pdf/809662.pdf. In his book, Evans counts lives saved if percent changes in total fatalities in the U.S. had been the same as in Great Britain, Canada and Australia, as he does here with Sweden, with no adjustment for changes in number, types or miles driven of vehicles on the roads in separate countries. In his book, he also calculated rate per vehicles, which greatly reduced the differences, but did not adjust for vehicle types and mileage. In my paper critiquing Evan’s analysis, I showed the huge growth in pickup trucks and truck-based SUVs in the U.S. relative to other countries, including Sweden. Because the makes and models of vehicles in Canada are similar to those sold in the U.S., but the relative mix is different, I applied U.S. death rates to Canadian vehicles by make and model, summed the total and corrected for miles driven between the countries. The difference in Canadian and U.S. rates disappeared. (1) Evans argues that the changes in other countries are the result of less emphasis on vehicles and more on behavior, with no documentation of specific policies or expenditures in the countries mentioned. In my paper, I documented that the U.S. government spends hundreds of millions of dollars on behavior programs and the Canadian national government spends nothing. Also, the U.S. government has done little in recent years to regulate vehicle characteristics that would justify Evans claim of an overemphasis on vehicles in public policy. As I said in my initial response to Evans, changing vehicles does not preclude changing behavior as well. He would do nothing to change vehicles despite the evidence in my AJPH paper that more than half the deaths could be reduced by such changes. He has not produced a shred of “data-based science” to refute my findings.

Leon S. Robertson

1. Robertson, Leon S. Motor Vehicle Deaths: Failed Policy Analysis and Neglected Policy. JOURNAL OF PUBLIC HEALTH POLICY 27:182-189, 2006

US Traffic Safety Failure -- Data-Based Science Demolishes Robertson's Denial 28 November 2006
Leonard Evans,
Scientist, lecturer, author
Science Serving Society

The adjustments for other factors that Robertson seeks are in the complete text [1]. It is NHTSA that concludes that SUV’s have lower total fatality rates than small cars, but higher rates than large cars. [2, page 200]

The mentioned book by Elvik and Vaa is indeed good. Amazon.com couples it with mine [1] and offers a deal for purchase of both together. Amazon has 10 reviews of "Traffic Safety" that additionally support the theme of the journal reviews [3] that the inferences in the book are based on solid science.

If US traffic deaths had dropped by the same percent as Sweden's between 1966 and 2005, our 2005 total would have been 17,000 instead of 43,000. This enormous difference can be only modestly influenced (in an unknown direction) by different percent changes between the countries in factors related to vehicles (number of vehicles, how far driven, safety equipment, type, mix, etc). Indeed, Sweden's 1966 vehicles were allegedly safer than ours, providing us more opportunity to improve.

Understanding why our fatalities increase while those in other countries decline requires applying science to quantify how different factors affect total risk. The persistent belief, in the face of overwhelming contrary evidence, that vehicle factors are central is the core of "The Dramatic Failure of US Safety Policy". [1, p. 381-411]

References

1. Evans L. "Traffic Safety". Bloomfield Hills, MI: Science Serving Society; 2004. www.scienceservingsociety.com/traffic-safety.htm

2. Kahane CJ. Vehicle weight, fatality risk and crash compatibility of model year 1991-99 passenger cars and light trucks. Report DOT HS 809 662. Washington, DC: US Department of Transportation, National Highway Traffic Safety Administration; October 2003.

3. Reviews of "Traffic Safety" www.scienceservingsociety.com/ts/Reviews.htm

Evans' rhetoric and reality 20 November 2006
Leon S. Robertson,
retired Yale Research Scientist

I did not comment on Evans’ comparison of the U.S. with Sweden because I know that most readers of the Journal are knowledgeable in epidemiology and recognize that one does not compare trends in raw numbers of deaths without adjustment for other factors. Most notably in this case, Evans does not calculate rates per vehicle or per mile during a period when SUVs and trucks more than quadrupled and cars almost doubled in number on U.S. roads. Evans apparently loves only the numbers that he thinks support his rhetoric. Evans cites his own book for data from the National Highway Safety Administration rather than the original source to support his claims regarding SUV deaths. My research does not support his claims.(1) Those who are interested in more objective analysis of ways to reduce motor vehicle deaths should check out a book by Elvik and Vaa(2) rather than Evans’ biased book. 1 Robertson, LS. Motor Vehicle Deaths: Failed Policy Analysis and Neglected Policy. J Public Health Policy 2006; 27:182-189. 2 Elvik R, Vaa T The Handbook of Road Safety Measures. Amsterdam: Elsevier, 2004.

Comments on Robertson's Response 16 November 2006
Leonard Evans,
Scientist, lecturer, author
Science Serving Society

Robertson's response did not mention the dramatic contrast I presented between changes in US and Swedish traffic deaths. Between 1966 (the year of the US Highway Safety Act) and 2005, US traffic deaths declined by a modest 15% (from 50,724 to 43,443). In the same period, Swedish deaths declined by 66% (from 1,313 to 440).

How can one ignore so stark a difference? How can one escape the conclusion that so enormous a difference must be due to fundamentally different policies? There are many other countries with performances similar to Sweden's. It is the US that is aberrant. [1, p. 389-407]

If vehicle factors had changed in corresponding ways in Sweden and the US, it is far from clear whether the enormous safety difference between them would be slightly greater or slightly smaller. NHTSA finds that SUV’s have lower total fatality rates than small cars, but higher rates than large cars. [1]

As for Robertson's comment about "eyeball trends", I would invite readers to examine my book [1] to see if they do not agree with the reviewer who writes "Every page exudes a love of numbers, statistics, and analytical rigor. Evans has a clear passion for getting the right answers. For all the strong opinions he lays out, one senses that his agenda is simply to understand how to improve traffic safety." [3]

References

1. Evans L. "Traffic Safety". Bloomfield Hills, MI: Science Serving Society; 2004. www.scienceservingsociety.com/traffic-safety.htm

2. Kahane CJ. Vehicle weight, fatality risk and crash compatibility of model year 1991-99 passenger cars and light trucks. Report DOT HS 809 662. Washington, DC: US Department of Transportation, National Highway Traffic Safety Administration; October 2003.

3. Eisenberg, D. Review of "Traffic Safety" by Leonard Evans. JAMA 2005 294(6), 746–747.

Response to Evans 14 November 2006
Leon S. Robertson,
retired Yale Research Scientist

Evans claims that variations among vehicles do not explain a large portion of the difference among vehicle death rates on the basis of eyeballed trends of motor vehicle fatalities among countries, rather than epidemiological analysis, and unsupported assertion regarding “public policy” in those countries. My paper (1) clearly shows that claim to be untrue within the U.S. Elsewhere, I have analyzed data that indicate the claims in his book and in his letter regarding the reason for trends among countries are false. Indeed, the differences among vehicles, mainly the disproportionate sales of overweight vehicles in the U.S. compared to the countries that Evan’s cites, are sufficient to explain why the U.S. no longer has the lowest death rate in the world. (2)

Changing vehicles to reduce fatalities does not preclude attempts to change driver behavior and some efforts such as belt use laws and anti- alcohol campaigns have some effect. But Evans’ argument that the screening of pilots, the double-check system between pilots and co-pilots, and the aviation traffic control system can be applied to tens of millions of drivers is foolish. The false positives and false negatives in driver screening programs are enormous and denial of driving privileges to large numbers of people would probably have less political support than increased gasoline taxes. Other than occasional “back-seat drivers” who may be more distraction than help, there is no co-pilot in cars. Continuous monitoring of tens of millions of drivers on the road at any one time is hardly feasible. Leon S. Robertson

1 Robertson, LS. Blood and oil: vehicle characteristics in relation to fatality risk and fuel economy. Am J Public Health. 2006;96:1906-1909. 2 Robertson, LS. Motor Vehicle Deaths: Failed Policy Analysis and Neglected Policy. J Public Health Policy 2006; 27:182-189.

Traffic deaths and fuel use – overwhelming influence is public policy, not vehicle characteristics 7 November 2006
Leonard Evans,
Scientist, lecturer, author
Science Serving Society

My purpose is not to dispute or endorse the detailed findings, but to question their relevance to public health or national fuel use. As documented in my book "Traffic Safety" [1], vehicle characteristics are not central to either safety or fuel use.

The relative unimportance of vehicles to safety is revealed by comparing fatality time trends in different countries.[1, p. 381-8] From 1979 through 2002, Great Britain, Canada, and Australia reduced fatalities by an average of 49%, compared to 16% in the US. If US deaths had dropped by 49%, more than 16,000 fewer Americans would have died in 2002. Accumulating the differences over 1979 – 2002 shows that by merely matching the mediocre safety performance of these countries, about 200,000 fewer Americans would have died.

These trends continue. In 2005 the US recorded 43,443 traffic deaths, the highest total in 15 years, and 4,881 pedestrians deaths, the highest number in four years.[2] Sweden recorded 440 total traffic deaths in 2005 [3] – their lowest total since the 1940s, and a more than 65% reduction from their peak value. Among US states with smaller populations than Sweden, 23 recorded more deaths than Sweden, 11 more than twice as many deaths as Sweden, and one (NC with 1,534) more than three times as many.

The obsessive focus of US policy on vehicles rather than on effective countermeasures is at the core of our dramatic safety failure. [1, p. 389- 407] It is driver behavior,[4] and government policies addressing driver behavior (belt-wearing, drunk driving, running red lights, speeding, etc.) that substantially affect casualties.[1, p. 332-58]

In commercial aviation safety, the US leads the world. In 2002 there were zero fatalities.[6] Yet (for operational reasons) modern aircraft are far less crashworthy than those in earlier high-fatality years. Aviation safety succeeds by focusing on preventing crashes, not surviving them. In order to make real progress, US road safety policy needs to shift to this same focus.[7]

The introduction of fuel economy standards for vehicles (CAFE) was followed by substantial increases in

1. Total national fuel use

2. The percent of that fuel that was imported

3. The average annual travel per vehicle[1, p. 89-91]

Total fuel use is determined by economics, not details of how it is consumed. Expecting CAFE to reduce fuel use is about as sensible as expecting to cure an alcoholic by insisting he drink out of a smaller glass.

The most effective policy to reduce fuel use is to increase fuel taxes. This is politically unacceptable, but we feel we must do something, no matter how irrelevant. As a nation we are like a 300 pound patient asking a doctor to reduce our weight – but with the stipulation that the treatment must not mention diet or exercise.

For both safety and fuel use, it is government policies, not vehicle characteristics, that are crucial.

References

1. Evans L. "Traffic Safety". Bloomfield Hills, MI: Science Serving Society; 2004. www.scienceservingsociety.com/traffic-safety.htm

2. Fatality Analysis Reporting System (FARS) Web-Based Encyclopedia. Data files and procedures to analyze them at http://www fars.nhtsa.dot.gov

3. Preliminary data on Road Safety in Europe: Improvement continues in the West - Decrease of fatalities in Central and Eastern European Countries - Positive signs confirmed in the CIS. http://www.cemt.org/events/PressReleases/06acc2005.pdf

4. Evans, L. The Dramatic Failure of U.S. Safety Policy. TR News (Transportation Research Board of the National Academies), 242: 28-31, January-February, 2006. See also "The Dramatic Failure of US Safety Policy", Chapter 15 of "Traffic Safety", www.scienceservingsociety.com/traffic-safety.htm

5. Evans, L. The dominant role of driver behavior in traffic safety: Am J Public Health. 1996;86:784-785.

6. US Department of Transportation, Bureau of Statistics. Table 2-1. Transportation fatalities by mode. Available at http://www.bts.gov/publications/national_transportation_statistics/2003/html/table_02_01.html

7. Evans, L. "Vision for a safer tomorrow". Final chapter of "Traffic Safety". Complete text available at http://scienceservingsociety.com/ts/text/ch16.pdf

Copyright © 2007 by the American Public Health Association