NOTE:  NHTSA has since answered the specific concerns that I voiced in this letter.  See:  NHTSA admits Misinterpretation of Alcohol-Related Traffic Fatality Statistics by some Data Users

An Open Letter to NHTSA

April 12, 2004

Jeffery W. Runge MD

NHTSA Administrator

400 7^th St. SW

Washington, DC 20590

Dear Mr. Runge,

I am deeply troubled by the use and abuse of the term alcohol-related.  I understand that NHTSA counts a fatality as alcohol related if it occurs in a crash where any of the drivers, pedestrians or pedal cyclists involved has a blood alcohol concentration (BAC) or 0.01 of more.  I understand that a BAC of 0.01 is one-eighth the legal limit in most states.  It also concerns me that NHTSA determines a fatality to be alcohol related based on known blood alcohol concentrations in the FARS as well as imputed BAC when alcohol test results are not reported.  According to information obtained on the NHTSA web site, in more than 50 percent of the cases, imputation is applied when alcohol involvement is not known.

The problem with this sort of computation methodology is that alcohol related intentionally, and perhaps unintentionally gets translated into deaths caused by drunk drivers.  Organizations such as MADD frequently refer to NHTSA alcohol related statistics as evidence of a drunk driving epidemic.  This exaggeration and manipulation of statistics has shaped our current DUI laws.  As a direct result, approximately 1.6 million people are arrested each year for DUI. 

When NHTSA reports that there were 17,419 alcohol-related traffic fatalities in 2002, other organizations and the media will inevitably allow the public to believe that these same 17,419 fatalities are in fact victims of drunk drivers.  (See enclosed Reuters article)

Despite significant concerns regarding the aforementioned, I’d like to stay focused on the term alcohol related, while discussing your administration’s culpability. As I am certain you are aware, Section 515 of the Fiscal Year (FY) 2001 Treasury and General Government Appropriations Act (Pub. L. 106-554, sec. 515, codified at 44 U.S.C. 3516 historical and statutory note), commonly referred to as the Data Quality Act, directed OMB to establish government-wide standards in the form of guidelines designed to maximize the "quality," "objectivity," " utility," and "integrity" of information that federal agencies disseminate to the public.

Based on all my research, I do not see how the NHTSA definition of alcohol-related fulfills the requirements set forth by the Data Quality Act, particularly when the associated statistics are deliberately employed by some organizations to mislead the public.  Related by definition, calls for a connection.  Yet, the NHTSA alcohol related statistics do not consider cause of accident.  If a drunk driver is struck and killed by a sober driver, and the sober driver was at fault, that fatality will be categorized as alcohol-related by NHTSA even though alcohol was unrelated to the cause of the accident. 

Is NHTSA misleading the public?  Why would NHTSA decide to use the word related when no effort is made to establish a connection and cause of an accident?  What is the benefit recognized by NHTSA to allow outside organizations to distort the alcohol related statistics?  Does NHTSA grant monies to MADD?  What is the NHTSA / MADD connection?  What legal action should a concerned citizen take to bring about a meaningful restructuring process in regard to NHTSA terminology? 

In 2002, how many people did drivers, who were tested and found to have a BAC beyond the legal limit, kill?  No imputation here, just the hard facts please.  Anecdotal evidence suggests that 55 percent of the fatalities are single vehicle, meaning the driver is killed with no other victim.  Based on my research, if the single vehicle fatalities are removed from the equation, the actual number of innocent victims of drunk driving will be approximately 5,000 people each year, which is significantly different from the 18,000 fatalities being reported by the media as deaths caused by drunk drivers. 

What are my goals?  The NHTSA needs to set the record straight - publicly.  The NHTSA should immediately cease using the term alcohol related until such time when the connection between alcohol detection and cause of accident is firmly established.  Finally, the NHTSA should create and maintain an easily assessable report that reflects the actual, substantiated traffic fatalities that included a driver that tested beyond the legal limit.  No imputation, just the hard facts.  This is the data that should then be used to shape future laws. 

I take this issue very seriously.  I have made numerous calls to your agency, only to get bounced around to different departments, none of which could answer my questions.  If we are going to continue enacting tougher DUI laws with stiffer penalties, then our elected officials should be able to rely on accurate information from the NHTSA.  No tricks, no bells, no whistles, just the data presented with "quality," "objectivity," " utility," and "integrity", absent from political agendas. 

Please advise of actions taken.

Sincerely,

Ed Haas

Cc:      Senator Lindsey O. Graham, Senator Ernest F. Hollings, Rep. Henry E. Brown