Forcon: Forensic Consulting

Forensic Science





Historical Legal Issues

Current Issues




Email Forcon

Historical Legal Issues & Factors for Consideration

Appeal court ruling targets straddlers

The Ontario Court of Appeal decision in Regina vs Heideman, released September 12, 2002, addressed a specific application of Regina vs Carter, where BACs based on an indicated consumption pattern straddle the legal limit when the appropriate elimination rate range is applied.

In Heideman a toxicologist gave evidence that the drinking scenario presented at trial, coupled with the average elimination rate encountered in a given population, should have resulted in a reading of 71 mg/100 ml of blood rather than the 100 mg/100 ml indicated by the breath testing instrument. This expert also noted that this BAC could have ranged from 47-95 mg/100 ml when the forensically acceptable elimination rate range is applied. The expert noted that four million Canadians would be slow eliminators, whereas another four million would be fast eliminators.

As part of its decision, the court noted "the appellant seeks to say that he is an average person but cannot establish that fact. Absorption and elimination rates vary not only from person to person but also from time to time within each individual". It concluded that "moving from the average to a particular person is impermissible speculation in the context of the purpose and functioning of s.258 [the Criminal Code section dealing with breath testing]". The appeal was dismissed and the conviction upheld.

Certain comments must be made regarding the scientific issues dealt with in the decision.

  1. Absorption rates are certainly variable within individuals and among drinking subjects. The absorption rate may also vary from one drinking session to another. It is highly dependent on food eaten, the quantity and type of alcohol consumed, the rate of alcohol consumption and many other factors. However, the absorption rate has absolutely no relationship to the elimination rate. As a result, it has no impact on calculations of the type used in Heideman, except to lower the projected BAC, when the drinking scenario in question might suggest that absorption had not been completed at the relevant time.

  2. There is no doubt that elimination rates will vary among individuals. The current forensically acceptable elimination rate range is 10-20 mg/100 ml/hour. The average rate is considered to be 15 mg/100 ml/hour. When large populations are tested it is clear that the vast majority will have rates ranging from 13-18 mg/100 ml/hour.

  3. Insufficient evidence provided

  4. In Heideman, the expert indicated that the number of fast eliminators and slow eliminators are approximately equal. It does not appear sufficient evidence was provided regarding the distribution of elimination rates about the average. Anyone who has conducted hundreds of properly designed tests can confirm that the least likely rates are those at the extreme ends of the range, with the most likely being those close to, or equal to, the mean.

  5. Within the bounds of measurement uncertainty, the elimination rate is constant for a given individual since the destruction of alcohol is controlled by the number of enzymes responsible for this process in the liver. It is hard to understand how a given person would have more enzymes at one time of the day than another, or from one day to the next. Those who suggest variable rates tend to establish subject elimination rates on the basis of minimal data, or incorporate BACs in their determination which are still subject to absorption and distribution phenomena. The data used to determine the elimination rate must be truly elimination data.

  6. In addition, elimination rates determined by means of blood analysis will not necessarily be the same as those determined by breath analysis, since the measurement uncertainty for the differing biological matrices, the instruments used to gather the data, and the scientific criteria associated with each process, will vary. For this reason elimination rates determined by breath analysis would be slightly more accurate representations for given subjects when breath samples are involved. Likewise, elimination rates determined by blood analysis would be slightly more accurate when blood samples are at issue.

  7. Elimination rates will change over time when there is a significant change in alcohol use or misuse. The change in drinking habits must be marked and sustained. Excessive use of alcohol on infrequent occasions will not necessarily lead to significant changes in the elimination rate. As long as the drinking habits remain similar the elimination rate will remain constant.

  8. Use of averages a common practice

  9. The learned Justice in Heideman commented that moving from the average to a particular person is impermissible speculation. However, the application of averages and ranges in science is common practice. Such parameters are the cornerstone of the scientific method. In fact, the entire breath-testing protocol, as recommended by the Alcohol Test Committee of The Canadian Society of Forensic Science and adopted by the Federal Department of Justice, is based on specifications established by averages and ranges, some of which include the following:
  1. Design and operation of instruments. Instruments must meet certain specifications before the Alcohol Test Committee will consider them for approval. One such condition states: "When vapours of known alcohol concentration in the range corresponding to BACs from 50-350 mg/100 mL are analyzed, the mean result of thirty consecutive analyses at each concentration in the range shall be within ±5% of the target value and the precision shall be: - at concentrations of 100 mg/100 mL or less, the standard deviation shall not exceed 3 mg/100 mL; and - at concentrations greater than 100 mg/100 mL, the coefficient of variation shall not exceed 2.5%."

  2. Design and use of calibrators. The temperature of the simulator must be 34.0 ±0.2 degrees Celsius. The ideal temperature of operation is 34.0 degrees Celsius.

  3. Formulation and use of standard alcohol solutions. Alcohol standards for use with a simulator must comply with certain specifications. "A solution to produce a result of 100 mg/100 mL at 34.0 °C with an instrument calibrated according to a blood /breath ratio of 2100/1 shall be prepared to contain 1.21 ±0.03 milligrams of ethyl alcohol per millilitre of solution."

  4. Quality of breath samples. The breathing technique employed just prior to breath testing can cause variability in breath test results. Hyperventilation immediately prior to sampling has been shown to cause a decrease in the breath alcohol concentration of up to 20%. On the other hand, it has been reported that a subject who holds his or her breath for a short period before exhalation can increase the alcohol content in exhaled air by up to 15%.

  5. Breath temperatures. Body temperature is considered to be 37 °C. Because air loses its heat as it passes through and out of the respiratory system, end-expiratory breath temperatures can vary from 32.4-35.7 °C. Since all breath testing instruments assume a breath temperature of 34.0 °C, a lower body temperature would result in less alcohol per given volume of breath and a higher temperature would result in more alcohol for the same volume of breath. It has been shown that a breath-testing instrument will overestimate the true BAC by about 8.6% per °C increase in body temperature. In a separate study, it was demonstrated that a breath-testing device would underestimate by about 6.9% for each degree decrease in body temperature.

  6. Blood:breath ratio. This ratio describes the relationship between the alcohol content of breath and the alcohol content of blood at a given point in time. It has been accepted that the average blood to breath ratio is 2300:1 for subjects (range 1700-3000:1) in the post-absorptive phase. Since instruments are calibrated at 2100:1, BACs may be overestimated when the blood/breath ratio is less than 2100:1 and underestimate BACs when this ratio exceeds 2100:1.

  7. Time to achieve completion of the absorption phase. In some drinking scenarios, the BAC at the time of driving is lower than the readings obtained later. Since breath tests are conducted up to two hours after an alleged offence, a given subject's BAC will likely have changed in the interim. The BAC at the time of the alleged offence could have been higher or lower than the reading obtained later. It is generally accepted that the maximum BAC will be achieved in 30-60 minutes after last consumption. Certain drinking scenarios generate maximum BACs in shorter time intervals, whereas others may require far longer. The point is that opinions on this issue are dependent on several research studies and practical experience with drinking subjects. This part of the science is far from exact.

Each of the foregoing considerations (and many others) is dependent upon data collected in actual testing situations. Actual specifications are established by evaluating this data in conjunction with universally accepted statistical procedures to establish specific measurement uncertainty. The validity of given premises or results must be considered in terms of averages, ranges, the distribution of the data about the mean etc. Nothing in science is absolute. Every measurement has a built-in uncertainty with confidence limits.

Heideman seems to suggest that measurement uncertainty and confidence limits are absent in those considerations relating to the scientific aspects of criminal drinking and driving prosecutions. If it is impermissible speculation to consider the average elimination rate and the distribution of elimination rates about the mean as they apply to specific drinking scenarios, it may also be impermissible speculation to consider the averages for the above parameters without specific testing of a given subject.

If "variable elimination rates" are an issue, so too should be the variable performance of breath-testing instrumentation and auxiliary devices and calibration solutions, as well as all biological variability among given subjects.

Expert lured into oversimplification

These concerns are specifically valid for a case such as Heideman when one considers that the readings alleged were slightly beyond the legal limit. From a scientific point of view we need to ensure that the scientific principles are applied properly and fairly.

The argument that Parliament must have known these issues when this part of the Criminal Code was enacted and therefore do not apply to criminal prosecutions is results-oriented science. It must not be forgotten that the scientific principles of breath-testing were established by scientists and recommended to Parliament. Their adoption by Parliament does not change how they should be used.

The toxicologist in Heideman was asked in cross-examination by the Crown to speculate about the number of persons who might eliminate alcohol at the lower elimination rate. The cross-examination was skillfully designed to boost the probability of lower elimination rates.

The questions put to the expert elicited answers in the form of population totals that failed to differentiate between adults and children, drinkers and non-drinkers, or alcoholics and social drinkers. Heideman highlights the need for experts to avoid being trapped into over-simplification.

Click here
                          for a printable version
Copyright © 2004 G. Kupferschmidt Consulting Services Ltd.