Urinalysis to Determine Blood Alcohol in DUI Investigations

Urinalysis is the chemical analysis of a subject's urine. It has been used by police agencies in DUI investigations as a means of determining a driver's blood-alcohol level since long before breath testing devices came on the scene. It is still used in a few jurisdictions, usually as an alternative to breath or blood analysis, but it is today probably the least commonly encountered of the DUI test procedures. This is true for two reasons. First, taking a urine sample from a drunk driving arrestee is time consuming and, frankly, a distasteful job for the police officer. In jurisdictions requiring the sample to be taken at a hospital or laboratory rather than by the officer at the station, it involves considerably more time and trouble than a quick, on-the-spot breath test.

The second reason for the relative unpopularity of urinalysis is, quite simply, that it is the least accurate and reliable of the three methods of blood-alcohol analysis used in DUI cases. As researchers at Duquesne University have concluded, "the unreliability of using a urine ethanol concentration to predict a blood ethanol concentration cannot be questioned." Winek, Murphy, and Winek, The Unreliability of Using a Urine Ethanol Concentration to Predict a Blood Ethanol Concentration, 25 Forensic Science International 277, 280 (1984).

Dr. Kurt Duhowski, probably the most recognized blood alcohol expert in the country, prepared a document for the U.S. Department of Transportation entitled Manual for Analysis of Ethanol and Biological Liquids. That document concluded that:

"The only urine specimen which could, in theory, be confidently employed for alcohol analysis for purposes of estimating the co-existing BAC is ureteral [i.e., from the urethra] urine clearly an impractical specimen for law enforcement purposes. In summary, in the living human subject, only blood, saliva and urine are potentially usable specimens in law enforcement practice; of these only blood is sufficiently free of forensicallydisabling disadvantages and contradictions to be a practical specimen material for routine applications."

Similarly, the National Safety Council's Committee on Alcohol and Drugs (Ad Hoc Committee on Testing and Training) has recommended in DUI investigations that:

"Because of various problems in the interpretation of the results of the analysis of urine specimens for alcohol which cannot be readily overcome in law enforcement practice, urinalysis of urine for the purpose of determining blood alcohol is to be discouraged except under the strictly controlled conditions employed in determining renal [i.e., kidney] solute clearances."

Most forensic criminalists called by the prosecution in a DUI trial will be familiar with the studies and recommendations of the Department of Transportation and the National Safety Council, as well as the works and expertise of Dr. Dubowski. They will also acknowledge that many states have now repealed legislation authorizing the use of urinalysis in drunk driving cases.

The theory underlying urinalysis in a DUI case is simple. Alcohol is secreted by the subject's kidney by diffusion into the urine. The concentration of alcohol in the urine at the time of secretion in the average person is about 1.33 times the concentration of alcohol in the blood at the same time. Therefore, measurement of the alcohol in the recently secreted urine, followed by a simple computation, should theoretically give the subject's blood-alcohol ratio.

The urine sample is obtained necessarily from urine voided by the DUI arrestee. This represents urine that has been secreted by the kidney and passed into the bladder, where it sits until the subject urinates. Therefore, the urine that is being analyzed consists of a mixture of urine that has been secreted by the kidney over a period of time, at least since the last time the subject urinated. If the DUI suspect has not voided in 12 hours, for example, the specimen will contain a mixture of urine secreted by the kidney over that 12-hour period. In other words, the urine sample will not be indicative of the blood-alcohol level of the DUI suspect while he was driving within the past half hour or so, but rather probably will represent a chemical average of the level over the past 12 hours. Thus a man who had been drinking 8 or 10 hours earlier but who had "sobered" to a blood-alcohol level of zero by the time he was arrested for drunk driving, will produce a sample that clearly indicates - falsely - the presence of alcohol in his blood.