Age groups under 21 demonstrated dramatic drops in non-alcohol crashes while there was virtually no change in alcohol related deaths.  Between 2007 and 2008 there were 65 fewer non-alcohol deaths for the 0- 15 age group, a 34% decline.  Because of small numbers, the significance of this change was only P<.10).   For the 16-20 age group the number of non-alcohol deaths dropped by 64 or a 22% drop (P< .05).  Notice for non-alcohol related deaths age 16-20 the difference between 2006 and 2008 is 121 or 34% drop and that there has been a consistent drop for both groups from the high in 2003.

There was no significant difference by gender when comparing 2007 and 2008 by alcohol use.

Given the pattern of on-alcohol related crashes for passengers and the younger age groups, respectively, it is no surprise that one observes a significant difference when these groups are examined in combination.  For the passenger 0-15 year old age group there was a drop of an astounding 52% in one year, from 128 deaths in 2007 to 62 in 2008.  For deaths in the non-alcohol related passenger deaths for the age group 16 - 20, the drop what less but still impressive 42%.  The pattern for passenger alcohol crashes displayed virtually no change between 2007 and 2008 for either age group.

Hence, out of the total reduction of 566 deaths between 2007 and 2008, 116, or 29%, came from non-alcohol related passenger deaths for people under 21.  This is a dramatic change for such a sub-group and difficult to hypothesize what the explanation might be.  Overall, there was a dramatic increase in gasoline princes between 2008 compared to 2007 in California with a corresponding drop in the total Vehicle Miles Traveled (VMT).  Further the economy went down and unemployment went up both of which might have also reduced VMT.   Given these changes one might expect the dramatic drop in overall deaths on California highways, but one would expect the change to be relatively the same for drivers, passengers and pedestrians.  (See role graphs.).   In fact,  there was a reduction in the number of deaths between 2007 and 2008 for all six combinations of the two alcohol use categories and drivers, passengers, and pedestrians, respectively. (See Table of Numbers).  However the change for non-alcohol passengers between these two years was 33%, with the next largest reduction being in non-alcohol drivers that had a 18% reduction, close to the overall average of change.  The non-alcohol passenger reduction in deaths still stands out, and most dramatically for non-alcohol deaths amongst passengers under 21.

One might also of expected for the decrease in alcohol related deaths to be the same or greater for non-alcohol related deaths for these age groups.  There has been a prolonged push to discourage young people from riding with drivers under the influence, but this data shows no indication of an effect from these efforts.  One might also think that whatever is the explanation for fewer deaths among young passengers to be the same for both non-alcohol and alcohol related deaths.  That is there is a reduction in young passengers for some reason and this effect is a cross the board.   Similarly if there was a pattern in the change of seat belt use by young passengers one might expect to see the effect across alcohol related and non-alcohol related deaths, although there is reason to consider the possibility that passengers are less likely to use safe belts when they are under the influence, or less likely to use a safety belt when the driver is under the influence.

Why the reduction of passengers deaths for non-alcohol related collisions, particularly among young adults, remains a mystery.   Comments welcome:   mailto:cacrash@cacrash.org!

*Alcohol involvement is determined by the police.  They may use a blood test, breathalyzer, road side sobriety check (eg. can the individual walk straight and stand one one foot), smell or only observation.  Generally fatality crashes are comparatively well investigated and documented.