Summer drink drive campaign results

The Association of Chief Police Officers (APCO) has
released the results of this summer’s drink drive campaign. Compared
with 2008, the total number of breath tests administered rose by 32%
from 91,848 to 121,398 and the number of tests that were positive,
failed or refused dropped from 7.6% of tests to 5.8%.

During the campaign breath tests were given to drivers
involved in a collision, as well as motorists who were either committing
traffic offences or whose driving led to a suspicion of impairment
through alcohol or drugs. This is the usual procedure for police to
administer breath tests; the summer campaign calls for officers to place
particular emphasis on this part of their job.

Following collisions, 82% of drivers were tested for
breath alcohol levels. This is a similar figure to the 83% tested in the
2008 campaign, and considerably higher than the year-round figure of
56% of drivers involved in injury collisions (2007; more recent figures
unavailable). The proportion of these tests that were positive, failed
or refused increased from 7.2% in 2008 to 8.3% in 2009.

The number of drivers tested who had not been involved in a
collision increased by 40.1% relative to the 2008 campaign. APCO report
this increase as evidence of the commitment shown by the police to
tackling the problem of drink driving.

If the increased number of tests is an indication of
increased police commitment (and we have no reason to doubt this), this
implies a difference in the samples of drivers tested in each of the two
years. We can assume that police will give highest priority to testing
those most likely to be under the influence. So, if they increase their
efforts to test a larger number of drivers, this larger number will
include drivers who are judged to be somewhat less likely to be under
the influence. Therefore, increasing the number of drivers tested would
almost certainly reduce the percentage who test positive, which is
exactly what happened in this campaign.

It is likely that the reduction in the percentage of
drivers who tested positive is a result of the increased number of
drivers tested, rather than a genuine reduction in rates of
drink-driving. The number of breath tests following a collision changed
very little across the two years, so the increase in the percentage
testing positive almost certainly represents a genuine increase in the
number of drivers drinking and getting involved in a collision.