Continued from Page 1...
The results indicated that the rate of decline in the blood alcohol curve was constant for an individual from day to day over time. The rate of decline among individuals though varied considerably from one to another. Widmark essentially calibrated has individual test subjects in order to estimate how much one would have to drink in order to reach the observed blood alcohol level. This method of calculation seemed reliable as long as the same approach was used which required the consumption of alcohol all at one time on an empty stomach. During the course of his research Widmark discovered what has been labeled as Widmark's r factor.
Widmark's r factor was the product of a test subject's calibration curve. Widmark would take a line and extend it straight back from the straight line portion of the alcohol curve until it intersected the time drinking began. This value would be an estimate of the theoretical blood alcohol level an individual could have if all the alcohol had been instantaneously absorbed and distributed throughout the body at the moment the drink was swallowed. This number would then be divided by the actual blood alcohol value and the result would be the r factor. While the r factor was quite constant and reproducible with regard to any given individual, it varied considerably from subject to subject. While this particular calculation seemed simple, its subsequent use in generalized approaches ignored the fundamental concept of its reliability which was that the true value of "r" for an individual must be known accurately to obtain a valid result. The only way to know the value accurately is to measure it.
Since it is rare for any individual defendant's r value to be known, experts have tended to use an assumed average value of 0.68 for Widmark's r value. This reportedly reflects the average value for the male subjects in Widmark's 1932 study. One should note that 0.68 is an arithmetic average that was obtained from a very small group of subjects. The entire subject base in Widmark's 1932 study was 20 men and 10 women.
Even though the arithmetic average for men in Widniark's study was 0.68, the range of values for the 20 subjects was from 0.52 to 0.86. The range for women was 0.47 to 0.64 with an average r value for women of 0.55.
Another researcher, Osterlind, has confirmed that the range of r values for men and women is substantial. It is not scientifically sound nor is it borne out from the research for an expert to state that an "average person" would have an average r value. In the studies conducted there is not even a close grouping of observed results around the average r value associated for men and women. In Widmark's study, of the 30 subjects, only three had r values between 0.68 and 0.70. Nineteen were below 0.65 and eight were above 0.70.
Therefore, an arithmetic average r value cannot, with any reliability, be assigned to any individual. There was no correlation of r value by size, sex, age, or general physical condition.
Perhaps most fundamentally, there is no research to establish that the r values observed by Widmark under the conditions of simultaneous consumption on any empty stomach apply to the more complex drinking patterns of the real world. Finally, the danger of using the average r can be demonstrated by measuring values obtained with r at the extremes of its ranges. Calculating the amount of alcohol consumed based on a measured blood alcohol level using the average r of 0.68 could get results which are as much as 31°k too high or 21% too low for a man and as much as 45% too high for a woman if calculated using the high and low ends of the r value ranges stated earlier.
Return to Chemical Evidence