Effects of Eating and Smoking
on DUI Breath Alcohol Tests

The delaying effects of food on alcohol absorption - and thus on any attempts to use retrograde extrapolation in guessing earlier blood alcohol levels in a drunk driving case - have been documented. In A. W. Jones, "Concentration-Time Profiles of Ethanol in Capillary Blood After Ingestion of Beer", 31 Journal of the Forensic Science Society 429 (1991), groups of subjects were given both light and regular beer on an empty stomach, during a normal lunch, and one to two hours after a meal.

The results of blood-alcohol testing showed lower peak blood-alcohol concentration (BAC) with light beer on an empty stomach than with regular beer; the same was true for beer consumed after a meal. Peak concentrations for the two types of beer taken during a meal were similar, but the regular beer produced a slower rate of elimination.

Peak BACs for both kinds of beer were lower with a full stomach, and those peaks were reached earlier. This confirmed earlier studies indicating that some of the ingested alcohol is trapped by food in the stomach, forming a so-called bound pool which is not absorbed immediately. The untrapped "free pool" of alcohol, on the other hand, is absorbed as if in an empty stomach, causing a peak BAC to be reached quickly. The bound pool of alcohol continues to be slowly absorbed, creating a slowly dropping level of blood-alcohol concentration.

See also Jones and Neri, "Evaluation of Blood-Ethanol Profiles After Consumption of Alcohol Together with a Large Meal", 24(3) Canadian Society of Forensic Science Journal 165 (1991), for a study with similar results. After extensive experimentation involving ingestion of alcohol by 16 DUI police officers, the researchers concluded that "[t]he results presented in this paper emphasize the need for caution when engaging in retrograde extrapolation."

Studies of the effects of eating on DUI breath alcohol tests have also been conducted by scientists at the Southern California Research Institute. Conducted under a contract with the Highway Safety Research Institute, the results are contained in a final report entitled "Methods for Estimating Expected Blood Alcohol Concentration", DOT HS-805-563 (U.S. Department of Transportation, National Highway Traffic Safety Administration, Wash., D.C. 20590). Among the findings:

The limited literature on alcohol and food effects provides considerable evidence that food delays gastric emptying, and that it is the delay which slows the rate of alcohol absorption. In turn, it is the slower absorption process which underlies the lower BACs associated with food intake.

If a drinker wishes to hold his or her BAC at a low to moderate level, what are the guidelines for food consumption? What is the best time to eat in relation to alcohol intake? What type of food is best? How much food is it necessary to eat to insure the desired effect? The data from this study indicate that the following are appropriate responses to these questions.

In brief summary of the study findings, when food is eaten in close proximity to drinking alcohol, it can substantially reduce the BAC from the level that would be reached if the alcohol were consumed on an empty stomach. Protein has the largest effect, and carbohydrates also are very effective.

Fatty foods are relatively ineffective in any quantity, and for the purposes of limiting BAC, it would be best to restrict the fat content of meals to a minimum, choosing instead foods which are largely protein and carbohydrates. Finally, a large amount of these latter types of food will be more effective than a small amount.

The findings regarding time of food intake show an inverse relationship; the greatest reduction in BAG was found at one-half hour, the shortest interval between eating and drinking which was examined. At an interval as long as two hours, the laboratory data show that the food exerted an effect. At four hours there was no effect of the food (as would be expected inasmuch as that period of time allows the food to pass through the stomach). Note that the findings regarding time of food intake are from pilot data only. This variable was not examined in the larger experiment in which the single interval of one-half hour between food intake and alcohol consumption was used.

The findings concerning the amount of food are fairly straight forward. A full meal will offer more protection than a snack, but even small amounts of food will have some effects on the BAC. In the laboratory 1000 calorie meals, which were either high in protein content or high in carbohydrate content, were followed by a 40% reduction in BAG. Certainly a change of that magnitude has practical significance to the individual.

Type of food is a more complex issue than the other variables of interest, but the data provide the basis for some recommendations. The person who expects to be drinking and would prefer to offset the alcohol to some extent by eating can be advised to consume a typical, full meal. For example, a meal which contains bread, meat, potatoes, and vegetables would be ideal. Such foods are largely protein and carbohydrates with a smaller proportion of the food being fats, and that kind of basic menu can be recommended. There is no evidence in these data to suggest that any specific, special foods should be eaten.

The pilot study and experiment findings have defined the food conditions which will make the maximum difference in BAG. It is important to emphasize, however, that these are not all or nothing variables. Rather, almost any food, eaten during the time interval of two hours or less prior to drinking, can be expected to produce some reduction in BAG compared to drinking in a fasted state. Although a small amount of food may have relatively minor effects, it would be unfortunate if the public believed that only a full meal would suffice. Snack foods, light meals, and typical party foods which accompany drinking all can be eaten with some benefit expected for the individual who wishes to minimize BAC.

There are some significant limits to these findings which should be noted. The alcohol beverage (80-proof vodka and orange juice mixed in a 1:1.5 ratio) was consumed over a half-hour period. No examination was made of other types of alcohol or of other alcohol concentrations, nor was it possible within the scope of this study to examine the condition in which alcohol is consumed as multiple drinks over a longer time period. The individual who may use the findings from this study as a basis for estimating BAG is likely not to duplicate the laboratory conditions and will need to be advised of the limits of the data.

Clearly a high calorie meal is more effective than a low calorie meal of equal weight. However, to determine whether weight of food is a critical variable, an additional test is needed. Specifically, the question requires comparisons of meals with equal calories but varying weight. The tests did not include this condition, and thus the issue is not resolved by these data.

Also, the findings are based on a study of male subjects only. It is expected that the effects of food on BAG will be closely similar for women, but no data have been obtained with female subjects.

Finally, the subjects were young men, ages 2-30 years, and "age" as a variable was not included in the stud. It is possible that findings for older individuals will differ from those reported. Physiological changes, including changes in body composition, metabolic processes, and organ function which may occur with aging, presumably may affect the reported food-alcohol relationship.

It can reasonably be assumed that food will have some effect under these various conditions, but the extent of that effect has not been established. It is recommended that further study include a wider age range and include women as subjects. Additionally, if food is to find optimal use as an alcohol countermeasure, it will be necessary to examine the effects in circumstances more typical of social drinking. For example, it will be important to the alcohol user to know what may be expected if he/she has consumed an average-size, typical evening meal at 6 or 7 P.M. and then goes to a party or bar and drinks steadily for the rest of the evening without further food intake. Further study also will be needed to determine food effects on BAG when alcohol is consumed simultaneously with a meal.

It is recommended that the following findings be incorporated into a pamphlet for drivers:

  1. Eat a meal in close proximity to the time of drinking.
  2. Eat a full meal, if possible, with a variety of foods which are largely protein and/or carbohydrates.
  3. When it is not possible to have a meal before drinking, take advantage of whatever food is available. Almost any food will be better than no food at all.
  4. Avoid drinking when no food has been eaten for more than two hours.
Back to Top