Measurement uncertainty of blood alcohol concentration (BAC) by headspace gas chromatography: comparison of different strategies

Analysis of blood alcohol concentration (BAC) is a routine analysis performed in many forensic laboratories. As BAC results are usually contested in court, measurement uncertainty (MU) becomes a critical topic. This contribution reports the results of an investigation of the major sources of uncertainty affecting BAC determinations and compares two different approaches to the quantification of the corresponding measurement uncertainty in routine BAC determinations. First, a bottom-up approach with method validation data was used to evaluate and estimate the MU of the analytical method. The interplay between the different sources of uncertainty was characterized using a cause-and-effect diagram, their contributions were evaluated, and they were combined using standard methods for uncertainty propagation to derive the overall uncertainty of the analytical method. Second, a top-down approach is presented where MU is estimated from long-term results obtained in proficiency testing. Comparison and validation of the results of the two approaches suggests that the top-down approach yielded a reasonable evaluation of MU while also being much simpler and more cost-effective than the bottom-up approach.