No, those aren’t the odds that I’ll finish first in the NCAA tournament pool that I’ll neither confirm nor deny entering. Well, actually, they might be. I thought picking Texas to go to the finals might be a stroke of genius. Not so much.
Instead, 2100 to 1 is the average ratio that the concentration of alcohol in an individual’s blood bears to that in the person’s breath. In North Carolina, a person’s alcohol concentration for purposes of an impaired driving offense may be stated either as grams of alcohol per 100 milliliters of blood or grams of alcohol per 210 liters of breath. G.S. 20-4.01(1b). It is immaterial that this calculation is based only on an average blood to breath ratio and that it thus may overstate (in the case of an individual with a lower blood to breath ratio) or understate (in the case of an individual with a higher ratio) the person’s blood alcohol concentration. See State v. Cothran, 120 N.C. App. 633, 635, 463 S.E.2d 423, 424 (1995). In Cothran, the defendant sought to introduce testimony from a chemist that the defendant’s blood to breath ratio was 1722 to 1, resulting in a breath test result 18 percent higher than his alcohol concentration. The appellate court upheld the trial court’s exclusion of this testimony, explaining that the legislature adopted a breath alcohol concentration per se offense as an alternative method of committing the offense of impaired driving. Thus, the relationship of a defendant’s breath alcohol concentration to his or her blood alcohol concentration is irrelevant.
The legislature’s endorsement of the breath testing standard unquestionably is based on its acceptance of the scientific validity of breath testing, which is designed to measure the concentration of alcohol in deep lung breath. Michael Hlastala, a professor at the University of Washington, recently published this article questioning the accuracy of breath testing on the basis that alcohol’s exchange with airway tissue makes it impossible for a person to deliver air containing deep lung alcohol concentration to his or her mouth. The article is new, but Hlastala developed the theory years ago. Both Arizona and Vermont have rejected challenges based on Hlastala’s theory to breath test results. See State v. Esser, 70 P.3d 449 (Ariz. 2003); State v. Wells, 779 A.2d 680 (Vt. 2001). And though Hlastala states that exhaled alcohol concentration never reaches the level deep lung concentration (thereby resulting in a lower breath alcohol concentration reading than would be obtained from a sample of deep lung breath), he concludes that the interaction of alcohol with the airways results in a breath test that is fairer for some subjects than others. He reports that variations in breath alcohol concentration result from the volume of air delivered to the instrument as well as changes in breath patterns, noting variations that decrease breath alcohol concentration by as much as 11 percent (hyperventilation) and increase it by as much as 16 percent (a 30-second breath hold) . He recommends reconsideration of current breath testing protocols to recognize the limitations of accuracy of breath testing and further proposes decreasing the importance of threshold levels for penalties. The latter proposal strikes me as a pretty far swing on the BAC pendulum and pretty unlikely to happen (though, as noted above, my wagering skills are suspect). As to the protocol suggestion, some, though not all, of Hlastala’s concerns already are addressed by the requirement in G.S. 20-139.1(b3) that breath test results not differ by more than 0.02, and that only the lower of the two alcohol concentrations can be used to prove a particular alcohol concentration. In addition, North Carolina’s approval only of certain breath testing instruments listed on the National Highway Traffic Safety Administration’s Conforming Products List ensures that the devices have been evaluated for performance and accuracy. Moreover, approved instruments in North Carolina are subjected to further evaluation by the Forensic Tests for Alcohol Branch of the Department of Health and Human Services.