Vitreous alcohol: the author's reply

Abstract

We are pleased to find that our paper has generated some interest amongst our colleagues and we thank them for their constructive criticisms. Our paper did not contain reference to the more recent publication by Chao and his colleagues [l] because of overlapping publication times; our paper was already in press when that paper was published. We addressed the problem of predicting the blood alcohol concentration (BAC) from the vitreous humor alcohol concentration (VHAC) in an individual case in which the associated circumstances of alcohol consumption and death were unknown and no assumptions were made. The reference data which we used were appropriate to this issue. However, Kraut is correct in pointing out that if information on circumstances are available or if assumptions are made then it may be possible to narrow the range of the predicted BAC. Unfortunately, for many cases this information is lacking or incomplete, and this was so in 47 of the 117 cases studied by Kraut [2]. Although not included in our publication, we did look at the possibility of improving the predicted BAC by developing regression equations for different ranges of VHAC, but no substantial improvement was achieved. We believe that any estimate of BAC based upon VHAC in an individual case fatality must, of necessity, provide both a predicted range and the degree of certainty with which the true value falls within this range. Kraut uses a range of VHAC/BAC ratios of 1.1-1.5. We are unclear from his publications [2,3] why these specific cutoff points were taken. However, using this approach, a VHAC of 90 mg% predicts a BAC of 60-82 mg% (degree of certainty unknown). Our own formula would predict a BAC of 2913 1 mg% (with 95% certainty). In individual forensic case work the prediction interval for an individual case will be necessarily broader than the prediction interval for the mean of the population as a whole. In other words, if the