I recently traveled out of state for a DWI seminar where they discussed some interesting breakthroughs in recent studies regarding breath testing.
Recent studies have suggested that breath-alcohol test results can be off as much as .14! The problem is that our bodies do not process alcohol in the same manner, and the breath testing machines have no way of adjusting for the problem.
The problem is with what's called a partition ratio. Alcohol is absorbed into the system and gets in the alveolar sacks in the lungs with oxygen. The alcohol leaves the membrane and travels through the lungs at a different rate for each person. The partition ratio that the breath testing machine uses on each person's breath-alcohol test is 2100/1, even though it is possible for someone's partition ration to be anywhere from 1050/1 to 4200/1. That would mean that someone testing at a .08 breath-alcohol content could actually register as high as a .16 or as low as a .04!
More troubling is that preliminary testing has shown that partition ratios are, on average, higher in women (especially during the menstrual cycle), and African Americans. This research could give rise to serious due process concerns.
The problem is compounded when administration errors and other machine errors are considered. In turn, I always advise my clients to consider taking an independent test.