How reliable are test results for drink driving?

When a prosecution commences for driving with excess alcohol, the prosecution relies upon either a breath alcohol, blood alcohol, or urine alcohol reading. These readings form the basis of one of the criminal elements of the offence and therefore are the backbone of the prosecution’s case. Whilst each of these tests are designed to provide a reliable indication of the level of alcohol in a person’s body, how reliable are they?

Roadside Tests

When a motorist is stopped and is suspected to have consumed alcohol the first step the police usually take is to test them on a roadside preliminary breath test device. The reading generated by this device often forms the basis for an arrest. With that being said, these devices are not considered reliable enough to provide an evidential reading and therefore cannot be used to form the basis of the reading the prosecution rely upon when charging you for the offence of drink driving.

It is important to note that whilst these readings cannot be used as the evidential reading for drink driving, they can be relied upon as evidence of unfitness/impairment through alcohol in driving whilst unfit cases.

Evidential Breath Test Machines

There are three type approved evidential breath testing machines in the UK. These are the Camic Datamaster, Intoximeter EC/IR, and Lion Intoxilyser 6000UK. Whilst there are often considerable challenges to these devices, the starting position as far as the court is concerned is that these machines are reliable. This is because there is a presumption that the breath testing devices are working correctly unless evidence is introduced to establish the contrary.

No machine is infallible (as outlined in the House of Lords case of Cracknell v Willis [1988] AC 450. This means that whilst there is a presumption the machines are working correctly, that presumption is rebuttable by means of any admissible evidence. Such admissible evidence can include a defendant giving evidence about what they have consumed supported by expert evidence about why this consumption would not account for the reading produced. This is often referred to as a “Cracknell v Willis” defence.

Therefore, whilst there are numerous technical challenges that can be raised in relation to the evidential breath testing machines, they will often be considered reliable unless there is strong evidence to rebut the presumption of reliability.

Blood and Urine Specimens  

Unlike the breath testing machines, specimens of blood and urine have to be analysed by an authorised analyst. Furthermore, there is no presumption of reliability in relation to specimens of blood and urine. This means that the starting position is not that the reading is reliable, and the prosecution does need to establish that the reading is one on which the court can safely rely. Quite often this will be done by calling the analyst from the laboratory to give evidence in relation to the analysis. 

Despite the assertion that an analysis is reliable, this does not automatically make it so. Historically there have been significant issues when it comes to the analysis of evidential samples of blood and urine at labs. By way of example, in 2017 a significant number of motorists had their criminal convictions overturned due to suspected data manipulation at Randox Testing Services. More recently in 2023 1778 samples reported as above the prescribed limit of drugs reported by Synlab were rescinded due to quality failures in the lab.

In both these examples there are many cases where labs produced results that were not accurate, the prosecution relied upon these results, and motorists were convicted despite the results not being reliable.

Given the nature of forensic analysis, and the possibility of contamination, human error, and quality control error (among other things), an analysis of blood or urine whilst typically reliable should not be taken solely on face value, especially when there is a dispute in relation to the level of alcohol recorded.

Factors Impacting Reliability

When considering the reliability of evidential samples, there are a multitude of factors which can cast doubt over the reliability of an evidential reading. This includes, but are not limited to the following:

Machine Errors

During the course of obtaining an evidential sample of breath, there are a number of errors which can occur which call into question the reliability of the sample. Despite the machine reporting such errors, the police still sometimes proceed to charge. The police also sometimes mistake the machine aborting the procedure as a failure to provide, despite this not being the fault of the defendant.

These errors include:

• Mouth Alcohol: The evidential machines are designed to abort the procedure if the presence of alcohol in the upper respiratory tract due to conditions such as GORD are found. In the case of the EC/IR, the machines do not always reliably detect mouth alcohol, which can result in a falsely high reading.
• Interfering Substances: The evidential machines are designed to identify the presence of substances which are not ethanol and abort the procedure if such a substance is detected. If this does occur, the machine will record an error message of “interfering substances” on the printout.
• Incorrect Simulator Checks: During the breath test procedure, the evidential machine will check it is measuring alcohol within the tolerated limits by testing itself against a mixture of ethanol and air. If the check is outside of the acceptable range a simulator check error will occur.
• Breath Difference: The evidential machines are designed to abort the procedure when there is too great a difference between the readings, namely more than 15%. In such circumstances the machine will endorse the printout with the error message “breath difference”.
• Ambient Fail: If the evidential machines detect the presence of alcohol in the air at the end of their purge the test will abort and endorse the printout with the error message “ambient fail”.
• High Blank: During the evidential procedure the machines are tasked with checking themselves to ensure they are reading zero. If any one of the checks are not zero the machine will endorse the printout with an error message “high blank”.

Depending on the circumstance and the error that has occurred, it may be possible to argue that the reading is one which is not safe to rely, or that the officer should have moved on to require a specimen of blood or urine.

Contamination

Contamination can occur when there is an introduction of extraneous material into a sample. This is something which can occur before, or after a sample is received at a laboratory.

A sample could be contaminated during the extraction process or prior to arriving at a lab. In the case of R v Bolton Justices ex parte Scally [1991] QB 537 a conviction for driving with excess alcohol was quashed despite an earlier guilty plea when it came to light that the blood tests on which the prosecution relied on had been taken using “alcohol swabs”.

By way of example of how contamination could occur at a lab, the Forensic Science Regulator Code of Practice stipulates that environmental monitoring must occur when conducting an analysis for drug driving offences. This is because it is possible for samples to be contaminated during analysis from external sources which would directly call into question the reliability of the analysis. The Code of Practice states:

“The appearance of a drug in any sample or matrix blank where that drug should not have been present will also be investigated and monitored (including the presence of cocaine without its metabolite BZE). The presence of a drug in a solvent blank where that drug was present in the case sample analysed immediately before the solvent blank shall be investigated to attempt to determine the source and the potential effect on the result.”

Biotransformation

Biotransformation is the alteration of a substance (typically a drug) in the body. What is not often realised is that this can be very relevant to cases involving Cocaine.

Cocaine over time will naturally metabolise into Benzoylecgonine. This does not pose an issue if a sample is analysed quickly, however, if an analysis occurs after an extended period of time, regardless of how it has been stored, it is possible for the Cocaine to continue to metabolise into Benzoylecgonine and therefore artificially increase the actual level of Benzoylecgonine.

This can mean that when a sample is taken from a motorist it may be under the limit for Benzoylecgonine, but upon analysis it could be over the limit. If expert evidence was to establish that biotransformation could be an issue, this would directly call into question the reliability of the sample and whether it should be excluded from evidence.

Microbial Action

In the case of urine samples, it is possible that if microorganisms in the urine multiply, they can alter alcohol levels. An example of when this can occur is when there is sugar in the urine. If a sample is analysed quickly, the numbers of microorganisms will be too low to affect alcohol levels, but if the sample is not analysed quickly, the microbes could produce alcohol from sugars present in the urine. This would call into question the reliability of the result and whether it is a true reflection of the alcohol in the motorist’s system at the time of the alleged offence.

Back-Calculations

A back-calculation will likely occur in circumstances where an alleged offence has occurred a number of hours earlier and a motorist only provides a sample under the limit several hours later.

A back-calculation is used to determine the alcohol concentration of a motorist at a previous time and will normally be considered “safe”. With that being said, there are circumstances where such a calculation is relied upon, and the evidential reading reported may not be safe.

One such situation is when a back-calculation is relied upon when alcohol has been consumed within one hour of the alleged offence. The United Kingdom and Ireland Association of Forensic Toxicologists guidance on back-calculations stipulates that “standard back-calculations should not normally be performed where the individual has consumed alcohol within one hour of the incident”.

The House of Lords considered back-calculations in the case of Gumbley v Cunningham [1989] R.T.R. 49 highlighting that the prosecution should not seek to rely on evidence of back-calculations save where the evidence is both easily understood and clearly establishes the presence of excess alcohol at the time when a defendant was driving.

Therefore, if the prosecution does rely upon a back-calculation, and there is any doubt regarding the calculation, it is possible to challenge the reliability of the reading relied upon.

How can Olliers help if I believe my test results are not reliable?

Whilst test results for breath, blood, and urine may be reliable, this is not always the case.

Therefore, it is important to obtain expert legal advice if there is any doubt about the reading in your case. Our expert motoring solicitors are here to help, offering confidential and professional legal support at every stage of your case.

Contact our specialist motoring team by emailing info@olliers.com or by telephoning 01618341515 (Manchester) or 02038836790 (London).