Variability in the blood–breath ratio (BBR) of alcohol is important, because it relates a measurement of the blood-alcohol concentration (BAC) with the co-existing breath-alcohol concentration (BrAC). The BBR is also used to establish the statutory BrAC limit for driving from the existing statutory BAC limits in different countries. The in-vivo BBR depends on a host of analytical, sampling and physiological factors, including subject demographics, time after end of drinking (rising or falling BAC), the nature of the blood draw (whether venous or arterial) and the subject’s breathing pattern prior to exhalation into the breath analyzer. The results from a controlled drinking study involving healthy volunteers (85 men and 15 women) from three ethnic groups (Caucasians, Hispanics and African Americans) were used to evaluate various factors influencing the BBR. Ethanol in breath was determined with a quantitative infrared analyzer (Intoxilyzer 8000) and BAC was determined by headspace gas chromatography (HS-GC). The BAC and BrAC were highly correlated (r = 0.948) and the BBR in the post-absorptive state was 2 382 ± 119 (mean ± SD). The BBR did not depend on gender (female: 2 396 ± 101 and male: 2 380 ± 123, P > 0.05) nor on racial group (Caucasians 2 398 ± 124, African Americans 2 344 ± 119 and Hispanics 2 364 ± 104, P > 0.05). The BBR was lower in subjects with higher breath- and body-temperatures (P < 0.05) and it also decreased with longer exhalation times into the breath-analyzer (P < 0.001). In the post-absorptive state, none of the 100 subjects had a BBR of less than 2 100:1.
This study investigated the frequency, magnitude, and duration of the mouth alcohol effect (MAE) following the ingestion of non-alcoholic food or beverages on two approved screening devices (ASD): the Dräger Alcotest 6810 and Alco-Sensor FST. Preliminary testing revealed only 5% of 39 non-alcohol containing food and beverage items tested produced a breath alcohol concentration (BrAC) ≥10 mg/100 mL on the Dräger Alcotest 6810 whereas there were no positive responses on the Alco-Sensor FST. Subsequent testing on 52 subjects consuming Dempster's® White Bread, Wonder™ 100% Whole Wheat Bread, 7-Up® and/or Rockstar® energy drink showed the MAE to be small in both alcohol-free and alcohol-positive subjects. The maximum BrAC produced in an alcohol-free subject was 22 mg/100 mL which dissipated within 90 seconds following ingestion. In alcohol-positive subjects, statistically significant increases in the BrAC were observed following ingestion of select breads and beverages; however, the increase was of sufficiently low magnitude that analytical and sampling variability could account for the differences observed. The MAE following ingestion of non-alcohol-containing food and beverages is rare, transient, and of low magnitude. There is negligible risk that motorists will be negatively impacted by this effect.
More than 1600 paired-control tests data obtained on Alco-Sensor IV—RBTIV and Intoxilyzer® 5000 C have been reviewed. A quasi-normal distribution around the target value of 100 mg/100 mL was observed. Moreover, not only are instruments accurate, but they are also stable. In sequences of tests in real cases, paired data comparison of first (CTRL1) and second (CTRL2) control results showed a difference (CTRL1—CTRL2) of zero in about one-third (⅓) of assays. The majority of cases (98.5%) were found to have a cumulative absolute difference of 3 mg/100 mL. In just one case, the difference between CTRL1 and CTRL2 was 10 mg/100 mL, the maximum acceptable in Quebec. Thus, although calibration checks of 95 to 105 mg/100 mL are acceptable, measurement uncertainty of these instruments should be considered well below 10 mg/100 mL in the great majority of cases.
A one-year retrospective study was conducted of 2,759 duplicate Intoxilyzer® 5000C test results in the City of Toronto during 1995 with a statutory wait of “at least fifteen minutes” between tests. The time between tests ranged from 19 to 73 minutes (median = 22 minutes). The absolute difference between the first and second breath tests ranged from 0 to 0.042 g/210 L (median 0.007 g/210 L). The distribution of these differences was not normal (KS = 0.1566, skewness = −0.1143). The differences between the truncated first and second tests were not within the recommended 0.02 g/210 L in 7.5% of the paired data. The second test was ≥ 0.01 g/210 L less than the first breath test in 35% of the cases (n=981) but was ≥ 0.01 g/210 L greater than the first breath test in only 7% (n=203) using truncated results. The observed skewness in this distribution is likely due to the elimination of alcohol that occurred during the time between tests. Following the adjustment of the second test to account for the mean pharmacokinetic alcohol elimination rate in drinking drivers, the distribution represented by the difference between the two tests still represented a non-normal distribution (KS = 0.3787) but was less skewed (skewness = −0.0549). The differences between the first and second tests following this pharmacokinetic correction resulted in reduction in the number of tests outside the recommended 0.02 g/210 L difference to 1.6%. It is recommended that the statutory wait of “at least fifteen minutes” in the Criminal Code be changed to a time period of between at least two minutes to five minutes, to reduce the variability between the two test results.
Random samples from normal distributions are an important assumption for many statistical methods. The present study evaluates this assumption with regard to quantitative breath alcohol analyses. Eight individuals (six male and two female) consumed alcoholic beverages and subsequently provided replicate (n ranging from 22 to 69) breath samples to an infrared breath alcohol instrument within short time intervals. The serially collected data were treated with several descriptive and inferential methods. Descriptive results among the eight individuals included: mean 0.0420-0.1175 g/210 L, SD 0.0008-0.0045 g/210 L and CV: 1.9%-4.7%. Statistical tests for normality showed seven of the distributions to be reasonably normal (p > or = 0.25) and the other marginal (p = 0.051). A test for runs about the median showed random results (p > or = 0.10) for four individuals and non-random (p < or = 0.01) for the other four. The results suggest an individual's breath alcohol measurement, when appropriately collected and analysed, should be considered a random sample from a normal within-subject distribution. The existing variability in breath alcohol analysis, due largely to biological and sampling considerations, is acceptably minimized to warrant forensic application.
Gastroesophageal reflux disease (GERD) is widespread in the population among all age groups and in both sexes. The reliability of breath alcohol analysis in subjects suffering from GERD is unknown. We investigated the relationship between breath-alcohol concentration (BrAC) and blood-alcohol concentration (BAC) in 5 male and 5 female subjects all suffering from severe gastroesophageal reflux disease and scheduled for antireflux surgery. Each subject served in two experiments in random order about 1-2 weeks apart. Both times they drank the same dose of ethanol (approximately 0.3 g/kg) as either beer, white wine, or vodka mixed with orange juice before venous blood and end-expired breath samples were obtained at 5-10 min intervals for 4 h. An attempt was made to provoke gastroesophageal reflux in one of the drinking experiments by applying an abdominal compression belt. Blood-ethanol concentration was determined by headspace gas chromatography and breath-ethanol was measured with an electrochemical instrument (Alcolmeter SD-400) or a quantitative infrared analyzer (Data-Master). During the absorption of alcohol, which occurred during the first 90 min after the start of drinking, BrAC (mg/210 L) tended to be the same or higher than venous BAC (mg/dL). In the post-peak phase, the BAC always exceeded BrAC. Four of the 10 subjects definitely experienced gastric reflux during the study although this did not result in widely deviant BrAC readings compared with BAC when sampling occurred at 5-min intervals. We conclude that the risk of alcohol erupting from the stomach into the mouth owing to gastric reflux and falsely increasing the result of an evidential breath-alcohol test is highly improbable.
ResearchGate has not been able to resolve any references for this publication.