Calibrations and concentrations: Misconceptions of breath alcohol testing in the workplace

May 1st, 2023

Picture source: Gallo Images/Getty

Alcohol is a socially acceptable substance with well-known sensory, motor, and intellectual impairment abilities. Blood and breath alcohol concentrations are carefully regulated throughout South Africa (SA), in the public domain, on public roads, and in the labour environment where employees perform safety-sensitive tasks. This article focuses on breath alcohol testing in the workplace.

An interesting aspect regarding the adjudication of alcohol cases, comes to the fore when considering the two different standards of proof in each domain. The comparative nature of civil disputes with a balance of probabilities standard of proof, commonly regarded as a lower standard of proof, unfortunately, established the acceptance of alcohol test results of lesser quality in labour courts compared to the approach in the criminal courts with a beyond reasonable doubt standard of proof.

This approach severely impacts justice and its inherent notion of fairness. It is trite that science informs the law regardless of the standard of proof to enhance the capacity of the court to decide the matter fairly. For instance, if the test result is 51% accurate and the standard of proof is also on a balance of probability, it would result in a 49% likelihood that the matter is decided incorrectly. The accuracy of the test result should be non-negotiable to allow a presiding officer to attribute the weight of the test result, alongside other evidence, to decide the matter on a balance of probability correctly.

First and foremost, it is essential to remember that breath testing equipment (breathalysers) must not be viewed as a ‘calculator’, which can never give a wrong answer. These instruments have complex functioning mechanisms and should be employed with the same level of diligence as any other measuring instrument in an analytical chemistry laboratory, even though it is employed at the point of measurement. Test results must be interpreted and accepted with the same circumspection as required in any analytical laboratory where quality control measures are employed to ensure accurate and reliable results. The mere fact that the test is performed outside a laboratory environment does not imply that the veracity of the test results may be compromised since it severely impacts the individual’s livelihood and future. Therefore, it is also essential that the testing official understand the application of this piece of analytical equipment well before making a test result known, which may have dire consequences for the test subject.

Incorrect breath alcohol test results in workplaces are – in our experience – an actual reality in SA due to the incorrect testing protocols, misuse and incorrect application of instrumentation, and a general lack of understanding of the basic measurement science applicable to breath testing equipment.

The issues addressed in this discussion revolve around the –

  • interpretation of the concepts of ‘intoxication’ and ‘under the influence’;
  • concept of ‘zero-tolerance’ for inappropriate alcohol use in the workplace;
  • scientific correctness of lowering the breath alcohol threshold-cut-off concentration to ‘zero’;
  • misappropriation and misunderstandings of annual calibration certificates; and
  • scientifically correct analytical protocol for breath alcohol testing in the workplace.

Cognition and a basic understanding of the above issues will equip defence attorneys to assess the veracity of the alcohol tests performed on their clients and inform presiding officers of the pitfalls when making decisions that are critical to justice.

These matters become even more critical when considering that most labour disputes are adjudicated by the Commission for Conciliation, Mediation and Arbitration (CCMA), where most employees cannot afford proper legal representation and to use a forensic expert to testify on the scientific accuracy and reliability of the alcohol test results. Avoiding well-established scientific principles undermines the court system’s integrity and inequality since employers generally have deeper pockets than their opponents, allowing them to game the broken legal system.

Legislative framework

Regulation 2A of the General Safety Regulations of the Occupational Health and Safety Act 85 of 1993 (OHSA) addresses ‘intoxication’ and ‘admittance of persons’ in the workplace as follows:

‘Intoxication: (1) … an employer or a user … shall not permit any person who is or who appears to be under the influence of intoxicating liquor or drugs, to enter or remain at a workplace.

(2) … [N]o person at a workplace shall be under the influence of or have in his or her possession or partake of or offer any other person intoxicating liquor or drugs.’

The reader’s attention is drawn to the concepts of ‘intoxication’ and ‘under the influence’ in the regulation. Unfortunately, there are no formal regulations or guidelines with specified thresholds-cut-off alcohol concentrations exist for either breath alcohol concentration (BrAC) or blood alcohol concentration (BAC) in South African workplaces.

Due to the gravity of the situation, most industries in SA have a ‘zero-tolerance’ policy, which is often confused with a ‘zero-concentration’ threshold-cut off concentration value for BrAC. A threshold-cut off concentration of ‘0.00’ is, in principle, a scientific misnomer of the employers’ intention and may lead to a contractual impossibility – perhaps it is more accurate to say that it is simply unfair toward an employee.

It must be realised that the ‘0.00’ concentration level is dependent on the type of breath testing device (BTD) that is used – a more sensitive BTD will have a lower ‘0.00’ concentration level than a less sensitive BTD. The ‘0.00’ concentration level furthermore depends on the current operational state of the BTD. Employers strive to counter this by obtaining an annual calibration certificate on the recommendation of the suppliers of the BTDs.

We submit that the notion of ‘no-tolerance’ must be promoted rather than ‘zero-tolerance’ because the latter leads to the irrational conclusion by lay persons that the threshold-cut-off concentration must be ‘zero’. Zero-tolerance refers to a ‘stance against’ individuals with an alcohol concentration level above the required threshold-cut-off concentration to engage in safety-sensitive activities. Zero tolerance does not necessarily equate to a zero threshold-cut-off concentration.

The use of ‘intoxication’ and reference to ‘under the influence’ in the OHSA is problematic. We submit that a person with a low level of breath alcohol, such as 0,10 milligrams per 1000 millilitres of breath as required by s 65(5) of the National Road Traffic Act 93 of 1996, for instance, cannot be regarded as ‘intoxicated’ or under the influence. It must be kept in mind that finding an alcohol concentration at which an individual can be classified as intoxicated is rather challenging. Individuals present with various degrees of impairment with increasing amounts of alcohol in their bodies. The intention of the OHSA should be expressed by incorporating rational and clinically established BrAC threshold values in organisational policies or guidelines. The employee may then be disciplined for non-compliance with the prescribed threshold cut-off concentration. This approach places the burden of proof on the employer, who must ensure that the employee was tested according to due analytical procedures to guarantee the accuracy and veracity of the BrAC concentration test results. This is even more important if the BAC is the only evidence in the case.

In our opinion, to discipline an employee for intoxication at work if it is established that the employee has less than 0,1 milligrams alcohol per 1000 millilitres of breath in their system is problematic and irrational. The option for observational detection of alcohol use, does exist if employees are diagnosed as under the influence with sobriety testing if they are or appear to be under the influence or intoxicated in the workplace.

Medical-scientific framework
  • Calibration certificate

Calibration of measuring equipment in analytical chemistry, is an essential element of the process to ensure the accuracy of a test result. Analytical equipment must be calibrated, preferably before and after each measurement, with certified reference standards. If this is not possible, then a certified control specimen must be analysed together with the desired specimen to ensure that the instrument functions are within the calibration tolerance specified on the calibration certificate. The control specimen’s analytical results will corroborate the evidence that the BTD functioned adequately at the time of use.

The fuel cell, which is the Achilles heel of the BTD, may drift over time due to continued use or incorrect handling. Suppliers, however, recommend an annual ‘calibration’. In our opinion, this is a service and calibration combined. The instrument’s response tolerance requires continuous monitoring by the user and recalibration to ensure the instrument complies with the tolerance indicated on the calibration certificates. It is scientifically unacceptable to have it calibrated once a year and to assume that it remains calibrated for the rest of the year until its ‘next due date’. The most critical question in this instance is: What if the supplier finds that the calibration is out? Are the individuals who were tested then informed of the injustice?

Exclusion clauses on the calibration certificates typically state that the measurement results portrayed on the certificate were correct ‘at the time of calibration and that the subsequent accuracy depends on factors such as care and handling.’ Furthermore, recalibration is recommended by the suppliers ‘at intervals to ensure that the instrument remains within the desired limits.’

We believe that not calibrating the instrument or analysing control specimens in combination with every breath sample of a person and ‘assuming’ that the instrument responded within tolerance amounts to guesswork, which is irrational, despite all the assurances provided by suppliers. It is recommended that the legal representatives request the validated raw data from the employer and supplier that supports this claim rationally.

It is unfortunately commonly accepted that all that employers require to win a case, is to present a calibration certificate and a certificate of training for the operator.

It is in the interests of justice to inspect the calibration procedures and protocols of the supplier, in combination with the validation reports obtained in their facilities and not by their overseas parent companies, to exclude hearsay evidence.

  • Preliminary screening breath tests versus confirmation (evidentiary) breath tests

Due analytical process for threshold compliance testing in safety-sensitive workplaces, usually requires a preliminary screening test first and confirmation (evidentiary) tests for all non-negative screening results. Screening BTDs are used for detection above a safety alcohol threshold concentration in breath; however, the probability exists that a non-negative response can be triggered by substances other than alcohol. Therefore, screening BTD’s results, cannot be accepted as reliable evidence in any forum or court where decisive action is taken and important, far-reaching decisions are made. The confirmation test result has a legal effect, not the screening test results, which can be indicated as a non-negative result.

After a non-negative screening test result, the breath alcohol concentration must be confirmed with an evidentiary BTD at a forensically acceptable standard. A second screening test on another screening BTD does not qualify as a confirmation, even if the result is recorded on a printout.

An evidentiary test involves both a procedure and a breathalyser (EBT) of evidentiary quality. The correct protocol is as follows:

  • the test subject is observed for 20 – 30 minutes, during which they may not smoke or consume water or eat anything to oral cavity contamination by mouthwash for instance;
  • an initial air blank test is performed, which must register a ‘no alcohol’ response;
  • the first breath test is performed;
  • a second air blank analysis which must register a ‘no alcohol’ response;
  • the second breath test is performed;
  • a third and final air blank analysis is performed, which must again register a ‘no alcohol’ response; thereafter
  • the control specimen in the form of a certified alcohol dry gas reference standard is analysed (a gas mixture with a certified alcohol concentration traceable to the International System).

If the result of the control specimen is within the acceptable tolerance provided on the calibration certificate, then the average or the lowest of the two breath tests is recorded. The procedure must be repeated if the control specimen result does not comply with the tolerance requirement, and if the second test also does not comply, then the breathalyser must be decommissioned and serviced.

JB Laurens LLB LLM (Cum Laude) BEng (Chem) (UP) is a legal practitioner at Kisch IP in Johannesburg. Dr Johannes B Laurens BSc (Ed) BSc Hons MSc (Chem) MPhil (Med Law) PhD (Chem) PhD (Med Law) (UP) MSc (Appl Toxicology) (Surrey) is Director/Forensic Toxicologist at Expert Laboratory Services in Pretoria.

This article was first published in De Rebus in 2023 (May) DR 29.