Welcome to the Truck Safety Coalition

To view all of the content on this site you will need to download the newest Macromedia Flash Player

SIZE AND WEIGHT

ADVOCATES FOR HIGHWAY AND AUTO SAFETY TELLS CONGRESS THE LARGE TRUCK CRASH CAUSATION STUDY IS SEVERELY FLAWED

Tuesday, March 28, 2006 202-408-1711

Advocates for Highway and Auto Safety wrote to key Congressional transportation leaders yesterday reviewing major defects in the final report of the Large Truck Crash
Causation Study (LTCCS) sent to Congress in March 2006 by the Federal Motor Carrier Safety Administration (FMCSA). Advocates’ letter stresses that FMCSA in the final
LTCCS report has refused to heed any of the expert advice provided to the agency over the past several years about serious deficiencies in research design and data collection
used to produce the report. These flaws were repeatedly pointed out to the agency in reports issued by the National Academy of Sciences, the Centers for Disease Control, and Advocates for Highway and Auto Safety.

The full text of the letter follows:

March 27, 2006
(Dear Senator/Representative):
I am writing to convey Advocates for Highway and Auto Safety’s views about an important study affecting truck safety that was released last week by the Federal Motor Carrier Safety Administration (FMCSA). FMCSA has completed the study, entitled Report to Congress on the Large Truck Crash Causation Study, March 2006 that the agency was mandated to conduct pursuant to Section 224 of the Motor Carrier Safety Improvement Act of 1999, Pub. L. 106-159
(Dec. 9, 1999). The Large Truck Crash Causation Study (LTCCS) has been subject to severe criticism over the
past several years by the special oversight committee of the Transportation Research Board, National Academy of Sciences (NAS); by the Centers for Disease Control (CDC) which reviewed the LTCCS as directed in the Fiscal Year 2000 appropriations Conference Report No. 108-10;
and by Advocates for Highway and Auto Safety (Advocates) which presented a comprehensive, detailed evaluation of the major research design and data collection failures of the study in a technical paper and presentation at the 2005 International Truck and Bus Safety and Security
Symposium, November 14-16, 2005. Each of these three critical evaluations of the LTCCS essentially reached the same conclusion: the study was deeply flawed, could not provide credible explanations of why truck crashes occur,
and was based on numerous mistakes in both research design and data collection.

FMCSA has repeatedly refused to respond directly to these criticisms in public forums and does not acknowledge any of the major failings of the LTCCS that have been identified by government oversight reports or by private organizations. In fact, the final congressional report just
transmitted to the House and the Senate does not discuss, or even reference, the critical evaluations of the LTCCS performed by NAS, CDC, or Advocates. A careful reading of the final report clearly shows that no significant changes have been made to the methodology or findings of the study in response to these critical reviews. The report continues to rely on the same flawed research design and data collection approach developed several years ago that has beenrepeatedly shown to be invalid in major ways.
Moreover, new statements about the research method of the LTCCS have been made in the final congressional report that reveal even more serious abuses of valid data collection methods. For example, the report states that when drivers of any involved trucks or passenger vehicles in the crashes that were investigated were unavailable for interviews because those drivers had been severely or fatally injured, the researchers selected “surrogate,” that is, substitute “drivers” who presumably were given the circumstances about each crash and asked how they would have reacted or performed. This compounds what already is a fundamental reliance in the LTCCS on
interview, bystander, and even hearsay “data” that FMCSA has admitted is the major source of information about how crashes occurred. Relying on involved party or witness interview characterizations about why or how crashes occurred is widely recognized in the scientific research community as unreliable because these characterizations are infected with hindsight bias. In the case of drivers who in some of the crashes that were investigated were the sole surviving vehicle operators, interview information is already freighted with survivor bias.

In addition,although the final congressional report states early in the narrative that the LTCCS cannot actually provide causal explanations for truck crashes, the report nevertheless in its conclusions section states repeatedly that FMCSA intends to continue to use the data from the
study to seek the causes of truck crashes. However, this study, as was recognized by NAS, CDC, and Advocates, is thoroughly incapable of providing causal explanations for truck crashes that any scientific peer group would find acceptable. The study has no predictive or explanatory value whatever in understanding why truck crashes occur and how to prevent them. None of its findings, except for data about the mechanical condition of the vehicles involved in the investigated crashes, can be relied on for agency policy choices about safety interventions. The
LTCCS finding that the event immediately preceding each investigated crash involved some aspect of driver condition or performance rather than mechanical issues with the vehicles, is unhelpful and does not differ from the identical finding reached by countless other studies
conducted over many years. It is not useful in adopting agency safety policies. We are deeply concerned about the quality of the final congressional report on the LTCCS
because it appears to be a harbinger of what to expect as the product of many millions of taxpayer dollars being spent on the current Bus Crash Causation Study, Motorcycle Crash Causation Study, and Passenger Vehicle Crash Causation Study. The first of these is being conducted by
FMCSA, and the latter two are under the control of the National Highway Traffic Safety Administration (NHTSA).
The LTCCS, to date, has cost close to $20 million according to information provided to Advocates in conversations with FMCSA personnel. For the important safety goal of detecting why truck crashes occur and what can be done to reduce their frequency, the LTCCS has been
essentially a worthless use of public funds. On the basis of FMCSA’s and NHTSA’s rejection of all constructive criticisms of the major flaws in the LTCCS, there is every reason to expect that both agencies will ignore or evade similar criticisms of what appear to be the same research
design and data collection pitfalls being reproduced in these companion studies. If this is the case, these studies will spend many additional millions of dollars of scarce federal funds without any scientifically credible findings that can be used to reduce the horrific toll of
deaths and serious injuries suffered each year in the U.S. due to motor vehicle crashes. Unless Congress receives assurances from both agencies that major flaws in research design and data collection that have been repeatedly identified are being corrected, these studies should not be
funded.

Sincerely,
Judith Lee Stone
President