[ Letter to the Editor Identifying Components of Variability in Breath Alcohol Analysis To The Editor:

The magnitude and source of measurement variability is an important consideration i any analytical system and needs to be understood and appreciated by those evaluating forensic breath alcohol systems. Variability in measurement data, of course, has different independent components hat usually combine additively and in different proportions. Two important sources of variation are biological and analytical, each of which combines to form total variability as measured by the total variance (SrZ). It is important in evaluating and approving breath alcohol instrumentation hatthese two sources of variability are understood and that their relative proportions be appropriate. It may not be adequate o simply state that if an instrument conforms to the National Highway Traffic Safety Administration standard of 0.0042 g/210 L standard eviation (1), then it has sufficient precision, because this ignores the biological contribution to total variability. This concept of variability component and magnitude has been widely studied and applied in other biological contexts (2-6). The present communication describes a brief and simple way to separate and measure components of variability in breath alcohol measurements. Two individuals (one male and one female) each provided 10 sequential breath samples into a BAC Verifier DataMaster infrared instrument (National Patent Analytical Systems, Inc.) subsequent toalcohol consumption and an appropriate waiting period to ensure no mouth alcohol bias. Following each breath sample, the instrument automatically sampled and measured an external simulator standard (Guth Model 34C, Guth Laboratories) fo