AISC 2004 Homepage for team 002 (Matthew Woller, Levi Blackstone)

Predicting blood alcohol content (BAC) and the elimination rate of alcohol could be useful in a variety of circumstances. Individuals wishing to estimate their blood alcohol content after a certain time could benefit from such a prediction as well as professionals such as bartenders, police officers, and lawyers. Perhaps a more important application would be to scientists investigating the effects of alcohol on human health by providing a guide for effective dosing methods.

Several general models for BAC predictions are currently used by the Traffic Department that use an average blood alcohol curve to estimate ethanol elimination. These models do not provide an accurate prediction in a majority of circumstances because of variations in a number of factors for each individual. Most general models merely account for the body weight and number of drinks consumed by the individual, ignoring individual height, age, gender, drinking history, amount of food in the stomach and a host of other variables affecting the accuracy of the prediction. These models also use a general elimination curve to calculate alcohol elimination instead of modeling the process of absorption and elimination. Using the overall body mass in predictions creates especially inaccurate results because alcohol is only distributed in the lean body mass. People with a higher percentage of body fat reach higher blood alcohol contents after consuming the same dose of alcohol than people with a lower percentage of fat.

Two recently developed computer models are used to predict BAC (cBAC; Addiction Research Foundation, London, Ontario, Canada, 1991; and BACest; National Highway Traffic Safety Administration, Washington, DC, 1994). The cBAC model uses height, weight, gender, and (for men only) age to estimate total body water (TBW), and the BACest program uses only body weight and gender as variables, using a separate percentage of body weight for men and women to determine TBW. In a recent study, Davies and Bowen (2000) tested these models to determine the accuracy of the predicted peak BAC’s compared to actual experimental data. They found that each model seriously underestimated the peak BAC’s for their group of test subjects.

We believe that the process of ethanol metabolism can be modeled to provide more accurate estimates of blood alcohol content over time, and may overcome some of the problems posed by general models.


				Introduction
			Predicting blood alcohol content (BAC) and the elimination rate of alcohol could be useful in a variety of circumstances. Individuals wishing to estimate their blood alcohol content after a certain time could benefit from such a prediction as well as professionals such as bartenders, police officers, and lawyers. Perhaps a more important application would be to scientists investigating the effects of alcohol on human health by providing a guide for effective dosing methods. Several general models for BAC predictions are currently used by the Traffic Department that use an average blood alcohol curve to estimate ethanol elimination. These models do not provide an accurate prediction in a majority of circumstances because of variations in a number of factors for each individual. Most general models merely account for the body weight and number of drinks consumed by the individual, ignoring individual height, age, gender, drinking history, amount of food in the stomach and a host of other variables affecting the accuracy of the prediction. These models also use a general elimination curve to calculate alcohol elimination instead of modeling the process of absorption and elimination. Using the overall body mass in predictions creates especially inaccurate results because alcohol is only distributed in the lean body mass. People with a higher percentage of body fat reach higher blood alcohol contents after consuming the same dose of alcohol than people with a lower percentage of fat. Two recently developed computer models are used to predict BAC (cBAC; Addiction Research Foundation, London, Ontario, Canada, 1991; and BACest; National Highway Traffic Safety Administration, Washington, DC, 1994). The cBAC model uses height, weight, gender, and (for men only) age to estimate total body water (TBW), and the BACest program uses only body weight and gender as variables, using a separate percentage of body weight for men and women to determine TBW. In a recent study, Davies and Bowen (2000) tested these models to determine the accuracy of the predicted peak BAC’s compared to actual experimental data. They found that each model seriously underestimated the peak BAC’s for their group of test subjects. We believe that the process of ethanol metabolism can be modeled to provide more accurate estimates of blood alcohol content over time, and may overcome some of the problems posed by general models.

Introduction