Interpretation and presentation of alcohol results
The measurement process is not complete until the test results have been interpreted appropriately and presented clearly to those responsible for making a decision based on them. Too often, most effort is directed towards ensuring the highest integrity of the measurement process with less consideration given to the clear communication of the results and their possible limitations. Appropriate interpretation of blood-, urine-, oral fluid- or breath-alcohol results begins with a thorough understanding of the purpose of the measurement and the process that produced them. The results of a roadside screening test, for example, are interpreted differently from an evidential test result. The roadside breath test is generally undertaken to establish alcohol use and provide probable cause for an arrest, to be followed by a more highly controlled evidential test. In many jurisdictions, the roadside screening test is not considered acceptable for prosecution purposes and only the result of an evidential test will be accepted by the courts. As with any forensic evidence, alcohol results must be presented to the court along with a brief explanation of the measurement process. This allows the court to put the measurement results in their appropriate context and assign the relevant weight. Pre-trial meetings with the legal representatives are always useful to discuss the testimony and any unique issues in the case. It is very important that scientists present themselves as objective and impartial expert witnesses, and they should be an advocate for the evidence, acknowledging its strengths and weaknesses but giving an unbiased and objective opinion about the most likely interpretation. The expert witness should be prepared to present the basic principles of the analytical method and describe the instrument in terms that can be understood by a lay person. The legal basis for the approval of the instrumentation and method (e.g. administrative rules, statutory law or case law) must also be presented to establish the legal foundation for admissibility. When new technology is first introduced into a juris diction, a legal judgment is typically made to establish whether the technology has gained general acceptance within the relevant scientific community. In the US this is referred to as a Frye (Frye vs. United States 1923) or Daubert (Daubert vs. Merrell Dow Pharmaceuticals 1993) standard. To present the technical features of modern instruments in simple terms can be a challenge for the forensic scientist. For example, in a breath-test case the use of some very basic visual aids to illustrate the principles of IR absorption and the basic components of the instrument can be very helpful. Alcohol results should be presented and interpreted according to the legislative definitions and framework to which a particular country or state adheres. For instance, in the UK, the blood alcohol limit is quoted as 80 mg/100 mL and the breath-alcohol limit 35 lg/100 mL. In the USA, corresponding values are 0.08 g/100 mL and 0.08 g/210 L, respectively. It is also worth bearing in mind that conversion factors used, for example, to translate breath-alcohol levels into blood-alcohol levels and vice versa may also differ slightly from country to country. The expert witness must be prepared to acknowledge the uncertainty inherent in measurement results. All measurements possess uncertainty that arises from both systematic and random sources. Estimates of uncertainty become particularly important when the results are near the critical prohibited levels. Obviously the more replicate determinations that can be carried out on a sample the better, because the mean value is likely to be closer to the true value (assuming the procedures are accurate) compared with the situation where a single replicate is analysed. However, it is not practical to analyse large numbers of replicates for reasons of time and cost, and also because there may be a limited volume of sample available for analysis. Hence the norm is for duplicate subsamples to be analysed or, in the case of breath alcohol determinations, for the alcohol concentration in two separate exhalations of breath to be determined. The mean of the values is then reported. However, as noted above, there will be an uncertainty associated with the result reported. Ideally, the results presented to the court are accompanied by an estimate of their uncertainty. This provides rele vant information that allows the court to weigh the evidence in view of the required fitness-for purpose and also to make a decision regarding the offence as defined in the statute; for example, whether the level of alcohol in the breath of the defendant was higher or lower than the legislative limit. Even when the esti mate of uncertainty does not accompany the results presented in court, the forensic scientist should be prepared to determine and present these estimates. Although several methods exist to determine and present measurement uncer tainty (Ellison et al. 2000), one useful and intu itive method is to provide a confidence interval for the individual’s mean breath-alcohol concentration (Gullberg 2003). A similar approach can be taken for BAC results.
