Posted by AIRC

Shell Refining (Australia) Pty Ltd, Clyde Refinery v Construction, Forestry, Mining and Energy Union (DR2008/1238)


Oil and gas industry


Implementation of random drug testing; use of oral fluids or urine as specimen for testing


[1] Shell Refining (Australia) Pty. Ltd. (Shell, the company) applied on 5 March 2008 to have a dispute resolution process conducted by the Commission under Division 3 of Part 13 of the Workplace Relations Act 1996 (the Act).

[2] The dispute arose as a result of concerns the Construction, Forestry, Mining and Energy Union (CFMEU, the union) representing refinery operators had with certain aspects of the revised drug and alcohol policy (the D&A Policy) that the company wished to introduce at its Clyde Refinery and Gore Bay Terminal.

[3] The refinery operators are covered by a preserved state agreement which incorporates the terms of the Refinery Operators Shell Refining Award 2006. The parties have entered into a Memorandum of Understanding (the MOU) in relation to dispute

[4] The MOU provides that the parties to disputes must attempt to resolve them through a process of discussion at the workplace level; however disputes that remain unsettled may be referred to the Commission. The MOU further provides as follows:

3. References to the Commission

3.1 Where a matter is referred to the Commission in accordance with sub-clause 1.5 or 2.4, it will be as an alternative dispute resolution process under the Model Dispute Resolution Process provided for in the Workplace Relations Act 1996.

3.2 The parties agree that, where necessary to resolve a grievance or other dispute, the Commission will initially conciliate in respect of the matter.

3.3 If the matter is not resolved by conciliation, the parties agree that the Commission may arbitrate, or otherwise determine the rights or obligations of a party to the dispute. In undertaking this role the Commission shall have regard to the matters set out in clause 4 below.

3.4 Where exercising arbitration powers under this agreement, the Commission shall issue its decision and reasons for decision in writing.

3.5 Where the Commission has issued a decision as provided for above, it shall be final and binding on the parties to this agreement, subject to a review process ….

4. Commission Procedure

4.1 The parties accept and will allow the Commission to inform itself in respect of a dispute before it in such manner as it considers appropriate.

4.2 The parties will do all things necessary to assist the Commission in the conduct of matters referred to it, including:

(a) Attending proceedings as required;

(b) Accepting any process and procedure for the conduct of proceedings considered appropriate by the Commission;

(c) Accepting that such process and procedure may involve the taking of evidence on oath, or affirmation. However, the Commission will not be bound by any rules of evidence;

(d) Producing relevant documents (subject to exclusions for legal professional privilege, commercial-in-confidence documents, or other reasons accepted as appropriate by the Commission); and

(e) Making persons available whose presence the Commission regards would assist help in the resolution of the matter.

4.3 In any process to resolve disputes, the parties expect the Commission to:

(a) Recognise that the Company has the responsibility to manage and operate the Refinery in a safe, reliable and profitable manner;

(b) To the extent permitted by law, have regard to the principles and objectives set out in Part 2 of the Award;

(c) Act according to equity, good conscience and the substantial merits of the case, without regard for technicalities and legal forms; and

(d) Where relevant and circumstances warrant, consider previous decisions of the Commission or the Industrial Relations Commission of New South Wales.

4.4 The matters set out in this clause 4 are applicable to all proceedings before the Commission initiated in accordance with this Memorandum, including any review process….

[5] Conciliation conferences were held on 28 March, 2 and 9 April 2008. While these conferences assisted in clarifying a number of issues they did not result in the resolution of the dispute, and the parties agreed that the matter should be arbitrated by the Commission.

[6] The parties signed an agreed on a statement of issues to be determined by arbitration on 28 May 2008. This read as follows:
Shell Refining (Australia) Pty Ltd (Shell) and the Construction, Forestry, Mining and Energy Union, Mining and Energy Division (the Union) consent to the Australian Industrial Relations Commission arbitrating and determining the questions set out below in an alternative dispute resolution process conducted under Division 3 of Part 13 of the Act and in accordance with the Memorandum of Understanding concerning Dispute Resolution Process entered into between Shell and the Union and attached to the Application in this matter dated 5 March 2008.

1. Should Shell be required to only apply random drug and alcohol testing under its Drug and Alcohol Policy to operators at Clyde Refinery and Gore Bay Terminal if such testing is applied to all Shell employees at Clyde Refinery and Gore Bay Terminal?

2. Should Shell be required to only apply random drug and alcohol testing under its Drug and Alcohol Policy to operators at Clyde Refinery and Gore Bay Terminal if all employees of contractors engaged at Clyde Refinery and Gore Bay Terminal are subject to such testing?

3. Should Shell be required to restrict initial random drug testing of operators to oral fluid testing?

[7] The dispute was heard in Sydney on 26 and 27 June, and 14 July 2008. Shell was represented by Mr Adrian Morris (Partner, Blake Dawson). The CFMEU was represented by Ms. Judy Gray (National Legal Officer).

[8] Section 702 of the Act states that alternative dispute resolution processes under Division 3 of Part 13 must be conducted by the Commission in private, and the Commission must not disclose or use any information or document that is given to the Commission in the course of conducting the alternative dispute resolution process to any person, except in certain circumstances, including where the parties to the process have consented to the disclosure or use. In this case, both parties agreed to waive confidentiality and gave their consent to this decision being made public1.


[9] The following witnesses gave evidence on behalf of the company:

  • Mr Neil McKenzie, Country Health Safety Security and Environment (HSSE) Manager, Shell;
  • Dr John Vine, Laboratory Director, Racing Analytical Services Ltd (RASL); and
  • Dr Teri Lillington, Manager, Shell Health, Oceania.

[10] The following witnesses gave evidence on behalf of the CFMEU:

  • Professor Olaf Drummer, Head of Forensic & Scientific Services, Victorian Instituteof Forensic Medicine; and
  • Mr Mark McGrath, Relief Senior Operator, Shell Clyde Refinery.

[11] According to Mr McKenzie, the D&A Policy currently being implemented forms an important element of Shell’s HSSE Policy in discharge of Shell’s occupational health and safety (OH&S) and environmental responsibilities and obligations across all its businesses in Australia2.

[12] Mr McKenzie gave evidence that the global Shell Group’s expectation is that all Shell companies have in place policies and processes to manage the risks of drugs and alcohol in the workplace3.

[13] Mr McKenzie gave evidence about the process that led to the development of the D&A Policy. He stated that …

given the multiple subjective responses to issues associated with drugs and alcohol and the difficulty in understanding and applying a policy in such circumstances, it was also necessary for the D&A Policy to incorporate generally accepted and applied principles such as:

(a) The management of objectively assessed risk to health and safety in the work place consistent with the risk management regimes under State and Territory occupational health and safety regulatory regimes;

(b) Reference to and application of external indicators and sources such as Australian Standards and accepted and understood blood alcohol levels;

(c) Use of testing agents, analysts and reviewers independent of management;

(d) Any random testing structured primarily as an effective deterrent, rather than as a point of detection for enforcing disciplinary action;

(e) The need for an holistic approach to the management and use of drugs and alcohol and the interaction of their usage with the workplace environment in which employees and other stakeholders could place their confidence, including:

(i) providing a range of preventative, educational and rehabilitative measures to overcome problems that could impair an
individual’s safety and performance at work;

(ii) providing that people who are unfit for work as a result of alcohol or drug use are managed in a fair and constructive manner;

(iii) ensuring that there is consistency across all Shell Australia businesses in all jurisdictions and clarity around expected
behaviours, management systems and consequences; and

(f) The need for policy to be informed by community, employee and regulatory workplace expectations, the experience of other Australian corporations in the implementation of drug and alcohol policies and international experience and views.4

[14] Mr McKenzie noted that:

The Policy has built into it an ongoing formal review process [see D&A Policy: section 11]. This recognises that management of the threats and risks associated with drugs and alcohol in the workplace is an evolving area that needs to be formally reviewed on a periodic basis. The review periods provide for a review after 12 months of full operation of the D&A Policy across all Shell business units and every 2 years thereafter, or earlier if required.5

[15] The objectives of the D&A Policy are described in the policy document (which was attached to Mr McKenzie’s statement) as follows:

Shell is committed to providing a safe work system and a safe and productive workplace by eliminating conditions and work practices that could lead to personal injury and equipment and other damage.

People whose fitness for work may be impaired by alcohol or the use of drugs (such as cannabis, amphetamines, cocaine, opiates, and other narcotics) the use of prescription drugs (such as benzodiazepines, barbiturates or methadone) or the
misuse of non-prescription drugs (such as codeine) can pose a risk to themselves and other people. Shell’s objectives are that employees recognise the threat that alcohol and other drug consumption can present and that the associated risks to people, property and Shell’s business are eliminated.

In support of Shell’s commitment, high priority is given to:

  • Creating a safe working environment by the elimination of the hazards associated with inappropriate alcohol and drug consumption.
  • Providing a range of preventative, educational and rehabilitative measures to overcome problems that could impair an individual’s safety and performance at work.
  • Providing that people who are deemed unfit for work as a result of alcohol or drug use are managed in a fair constructive manner.
  • Meeting our obligations to employees, contractors and the general public to carry out our operations safely.6

[16] The key elements of the policy are described as including:

  • Assessment and management of risk posed by the impacts of drugs and alcohol in the workplace.
  • Provision by Shell of training and education that promotes an understanding and awareness of the impact of drugs and alcohol on safety, health and performance.
  • Encouragement and support for individuals to take responsibility for managing personal alcohol and drug problems by making provision for counselling services and leave where appropriate.
  • Line Management responsibility for policy operation.
  • Testing as a condition of employment or appointment to Designated Safety & Environmentally Sensitive positions and Designated Management positions.
  • Random testing for specified levels of nominated drugs and alcohol for employees in Designated Safety & Environmentally Sensitive positions.
  • Testing in circumstances where there is a reasonable belief that a person’s fitness for work may be adversely affected by alcohol or drugs.
  • Testing methods and protocols consistent with Australian Standards.
  • Application of Shell’s disciplinary procedures in support of this policy.
  • Integration with other Shell policies and procedures that apply in Australia.

[17] The policy is described as applying to all employees and contractors of Shell (and all wholly owned subsidiary companies and joint ventures under Shell’s operational control).

Contractors are expected to adopt practices consistent with the objectives of the policy, including demonstrating how they manage to and support the objectives of the policy.

In the case of Contractors who perform high and medium risk activities as assessed under the Shell Contractor Management Accreditation System, contractor employees may be subjected to the same conditions for testing as Shell employees.7

[18] The policy encourages employees voluntarily to disclose to Shell that their fitness for work may be adversely affected through alcohol or drug usage. The policy indicates that such disclosures would not place employment with Shell in jeopardy. Support is provided for treatment and rehabilitation.8

[19] The testing program is described in the policy document as intended to:

  • Assist in identifying people who may either be unfit for work or at risk of being unfit for work.
  • Discourage people from coming to work when they may be impaired by alcohol or other drugs.
  • Raise awareness about the effects of alcohol and other drugs in the workplace.9

[20] Tests are conducted by authorised personnel in accordance with the relevant Australian standard. Drug testing is to be done by urine analysis by an accredited laboratory.

While the policy screens for substances at cut-off levels in the relevant Australian standard it states:

These cut-off levels are not intended as being a scientific indication that a person’s ability to carry out their work tasks is impaired. Rather, they indicate a positive presence of alcohol or drug that is a clear signpost for the need for management
follow up.10

[21] The policy also states:

Both Shell employees and contractors and contractor employees may be subject to testing.

[22] Mr McKenzie stated that:

The D&A Policy proceeds on the basis that contractors working on Shell premises or under Shell operational control are expected to adopt practices consistent with the objectives of the Policy, including being able to demonstrate how they manage to and support the objectives of the Policy. The D&A Policy further provides that in the case of contractors who perform high and medium risk activities as assessed under the Shell Contractor Management Accreditation System, contractor employees may be subjected to the same conditions for testing as Shell employees [See D&A Policy; section 4.4].11

[23] In relation to Designated Safety & Environmentally Sensitive or Designated Management positions, the policy provides for pre-offer of employment testing for external applicants, pre-offer of appointment testing for internal applicants and random testing. There is also non random testing where management has formed the considered opinion that alcohol or drugs may have contributed to an incident or a person’s fitness for work is diminished by alcohol or drugs. There may also be specific testing as a result of specific business or legislative requirements or as part of a return to work assessment or for rehabilitation plan monitoring.12

[24] The policy contains a graduated series of steps in response to positive tests, with the main focus on counselling designed to achieve a change in behaviour for the first or second confirmed positive result. The consequences of a third confirmed positive result are however potentially more severe. The policy states:

Formal disciplinary procedures commenced will allow the employee to understand the nature of the concern about his or her fitness for work, the nature of the breach of the policy and provide an explanation about the circumstances or challenge the
processes or any test results. Three confirmed positive tests under the terms of this policy will lead to the possibility of the employee being dismissed. Action to be taken will be determined on a case by case basis.13

[25] Appendix 1 to the policy document defines fit for work as meaning:

a state (mental, physical or emotional) which enables the employee to perform assigned tasks competently in a manner which does not compromise or threaten the safety or health of themselves or of others.14

[26] Appendix 2 to the policy document states that:

Random testing is intended as a credible deterrent to drug and alcohol use that could adversely impact an individual’s ability to safely and responsibly undertake their work.15

[27] Appendix 4 to the policy document concerns Designated Safety and Environmentally Sensitive Positions and Designated Management Positions. It states:

The impact of errors or misjudgement, due to a lack of fitness for work (for example as a result of alcohol or drugs) is not equal in every job. For example, the risk of immediate endangerment of personnel safety or the environment is greater in first line operational positions than it is in support staff positions. To reflect the reality of disproportionate impact, certain positions within Shell are classified as Designated Safety & Environmentally Sensitive positions or Designated Management

[28] The policy document notes that:

Subsequent to appointment, employees in Designated Safety & Environmentally Sensitive positions or Designated Management positions should anticipate random testing to occur at least once every two years. The random nature of the testing process may result in more frequent testing17.

[29] The policy document states:

Designated Safety & Environmentally Sensitive positions are positions in which action of the incumbent, or failure to act, can be a significant factor in events causing or leading to unsafe acts, environmental damage or material losses.
Subject to the exception below, nomination of Designated Safety & Environmentally Sensitive positions is made following assessment of whether:

1. The position exists as part of a plant, process or business operation where the potential exists for an incident the consequences of which are assessed as having a Severity rating 4 or 5 under the Shell Risk Assessment Matrix (the Incident) and

2. There is evidence in Industry*:

a. within Australia or
b. overseas of such a failure occurring, within the plant, process or business operation resulting in an Incident (the Event) and

3. The person in the position, through their action, inaction or failure of competency, has the ability to override, disregard or render ineffective controls that would otherwise prevent the Event.

Where the assessment is that criteria 1 to 3 apply the role is a Designated Safety & Environmentally Sensitive position. It is the obligation of each line of business to assess and document those roles within its structure that are Designated Safety &
Environmentally Sensitive positions.

This requirement is ongoing and includes periodic reviews and updating.

The exceptions to this assessment process are roles that are subject to specific legislated standards or requirements relating to alcohol and drugs, such as drivers of dangerous goods vehicles. These roles are designated automatically18.

[30] The policy document then deals with Designated Management Positions.

Nomination of Designated Management positions is made following assessment of whether:

1. The position is a management role where a person in that role

a. has authority to make or participate in making decisions or issue instructions in relation to or
b. has the capacity to materially influence operational or business processes, or decisions relating to Safety & Environmentally Sensitive positions where the exercise of that authority or capacity (including the failure to do so) could reasonably be assessed as likely to result in an incident with a Severity rating of 3, 4 or 5 under the Shell Risk Assessment Matrix (the Incident) and

2. There is evidence in Industry*:

a. within Australia or
b. overseas of the exercise of authority or capacity (including the failure to do so) by the person in a similar management or leadership position, causing or materially contributing to such an Incident.

Where the assessment is that criteria 1 and 2 apply, the role is a Designated Management position. It is the obligation of each line of business to assess and document those roles within its structure that are Designated Management positions.
This requirement is ongoing and includes periodic reviews and updating.

*For the purposes of this policy, when making an assessment the term “Industry” is used in a broader sense than the “oil industry”. This is to reflect that in undertaking its business, Shell covers many different “industries” e.g. the “transport” industry for vehicle operation, the “shipping” industry, for marine bunkering activities etc.

The expectation is that when making an assessment, real risks and their probabilities are considered, based on available knowledge of related industry groups and that their relevance is not overlooked because they did not fall by definition, into the “oil industry”.

In considering:

  • Experience from industry in Australia other than the oil and gas industry and
  • Overseas experience or knowledge

An evaluation is made as to the relevance, credibility and appropriateness of the comparison of that experience to Shell in Australia19.

[31] Mr McKenzie stated that:

Shell has determined that there is no value to be gained by allocating resources to management of risk where objectively applying risk management techniques it is assessed that none exists. To do so is contrary to risk management principles
generally. It also undermines Shell’s policy approach of setting the expectation that individuals act and behave responsibly i.e. take ownership of management of risks that arise from personal behaviours. Asking people whose roles do not pose a
credible risk to subject themselves to testing, whether on the basis of “fairness” or other “policy” views, lacks credibility and is contrary to Shell’s values and business principles..

The review of the D&A Policy also took into account the consideration that drug and alcohol testing imposes on privacy and there must be a proper risk management based justification for doing so. The need for such justification has been supported
by privacy and workplace safety agencies.20

[32] Appendix 9 of the policy document provides guidance for line managers when responding to a positive test. It includes the following:

  • The primary concern of Shell is for workplace safety and the wellbeing of the individual.
  • Each return of a positive test result is treated on its merits from the perspective of managing any risk that is posed in the workplace to the individual, fellow workers, the environment and property.
  • The return of a positive test does not necessarily mean that the individual is “impaired”; it merely means that the individual has evidence of a substance that may create an elevated level of risk that needs to be managed.
  • Consistent with Shell’s Values and Business Principles, individuals are entitled to be treated with respect and dignity.
  • What an employee does outside the workplace is not in itself of any concern to Shell. However, it may become relevant in the context of adverse effects, or the potential for adverse effects, on performance at work…
  • A positive test is a breach of policy and the Line Manager will need to act according to the circumstances. This may involve the initiation of stand down procedures and the management of a subsequent return to work, in accordance with Shell’s Counselling & Disciplinary Procedures, outlined in the PPM and the Rehabilitation Procedures applicable at the location21.

[33] Mr McKenzie stated that:

One of the principles that Shell has applied in revising its D&A Policy has been that any testing methods needed to be accurate, verifiable, reliable, repeatable, and consistent in application, with a minimum of false negatives and false positives. A positive result of a test is not a scientific indication that a person’s ability to carry out their work tasks in impaired. Rather, they indicate a positive presence of alcohol or drugs that is a clear signpost for the need for management intervention…

…the advice that Shell has received through Dr Teri Lillington, Manager of Shell Health, Oceania, has lead it to conclude that currently oral fluid testing is in the early stages of development and acceptability and is not a good fit with its Policy.

Shell anticipates that oral fluid testing will continue to develop, as will others (for example hair testing which is used in the United States but for which there is no standard in Australia) and will continue to monitor these and other initiatives.

Shell has concluded based on advice received through Dr Lillington that testing for drugs via urine is the best fit against the requirement that any test method be accurate, verifiable, reliable, repeatable and consistent in application, with a minimum of false negatives and false positives….

In particular, Shell has been advised through Dr Lillington that saliva testing delivers a relatively high level of false negative and false positive test results. This has serious unacceptable implications for the management of the risks of drugs and
alcohol in Shell’s workplaces. There is the risk that a substance may not be detected, thus preventing management of the associated risk. Equally, a false positive gives rise to the risk that action is taken that is unnecessary and wrongly directed. Both these outcomes have the potential to undermine the effectiveness of the testing as a deterrent and the credibility of the Policy in the eyes of all Shell employees and other stakeholders.

In addition to this, Shell has been advised through Dr Lillington that the suite of drugs and metabolites tested for with oral fluids is more limited than that able to be tested for with urine. This gives rise to the need for a greater number of tests to
cover the same range of drugs, with additional costs and complexity added to the process and even then the range of drugs tested for is limited. Again a credibility issue arises.

Nonetheless, the periodic review process that has been built in to the D&A Policy will facilitate reviews of testing methodologies in a structured way….

Shell has provided advice, in discussions with the refinery operators group at Clyde (the ROG) that it would consider a trial of saliva testing at Clyde for use in addition to urine testing, to better assess its overall merits. Shell has advised the ROG that it has concerns about the reliability of saliva testing and the potential adverse impacts from false positive results. Any decision to proceed with a trial will be dependent on the trial being able to be conducted against parameters designed to maintain the integrity of the Policy and protect both individuals and Shell.22

[34] Dr Vine is an expert on the analysis of drugs and other biologically significant substances. He is the Laboratory Director of RASL. RASL operates as a not-for-profit company. It provides drug testing services for racing. In addition, since 1990, RASL has
provided workplace drug testing services to Australia’s major petrochemical companies (including Shell) as well as to a number of smaller industrial companies.

[35] According to Dr Vine’s written statement RASL is accredited to AS/NZS 4308 for urine drug testing in humans and will shortly be seeking accreditation to Australian Standard AS 4760 for oral fluid testing.23

[36] In his statement, Dr Vine stated:

Urine drug testing covers and is validated for a wider range of drugs than oral fluid testing. AS/NZS 4308:200124 lists 18 safety related drugs and metabolites and the recently released AS/NZS 4308:200825 lists 22 drugs and metabolites which are covered by the procedures. In contrast, the Australian standard for oral fluid testing, AS4760-200626 lists only 11 drugs and metabolites and does not cover the benzodiazepines (sedating drugs known to cause impairment) at all.27

[37] According to Dr Vine:

There are a few laboratories in Australia which claim that they can provide oral fluid drug tests and one or two of them claim they can comply with AS4760-2006.

However, at the present time, no laboratories have yet been accredited by NATA as complying with the standard. Two laboratories, RASL being one of them, are currently working on achieving accreditation in the near future.

Nothing more clearly illustrates the developing nature of oral fluid testing. If no laboratories are yet accredited to AS4760-2006 it is difficult to see how properly validated testing, as required by any reputable workplace drug testing programme,
can be carried out at this stage.”28

[38] Dr Vine, in his written statement, referred to the ROSITA-2 project which was carried out initially in Europe in 2003-05 and as a parallel study in the USA which concluded in 2006. According to Dr Vine:

The study found that no on-site oral fluid testing device was reliable enough to be recommended for roadside screening of drivers as no device adequately covered the required range of drugs. The study also concluded that drivers would quickly realise that they often produce a negative test after having used drugs because of the limited sensitivity of the test.29

[39] Dr Vine continued:

As stated in the Rosita study referred to earlier the limited sensitivity of oral fluid tests will lead to the rapid realisation by those who have used drugs that their chances of being detected are relatively low. Urine drug testing is much more likely
to uncover patterns of drug use by individuals which may lead to levels of impairment and concomitant safety concerns.

This is highlighted particularly by the failure of oral fluid tests to detect the use of cannabis unless used very recently or in very large quantities. A study by Huestis and Cone30 showed that the detection time window in oral fluid after use of cannabis was no more than a few hours. While cannabis can be detected very easily in oral fluid after ingestion by smoking for an hour or two, its concentration then drops very rapidly and in many cases becomes undetectable using the on-site detection devices after a few hours. It can be detected for longer using laboratory-based tests.

The possibility of detecting recent cannabis use is often promoted as a vindication of the use of oral fluid testing. However, in practice, cannabis use becomes undetectable well before the impairing effects of cannabis wear off. Well controlled studies carried out on pilots in flight simulators have shown that significant impairment can be demonstrated up to 24 hours after smoking cannabis.”31

[40] During his examination-in-chief Dr Vine indicated that he was referring to a review published by the Australian Safety Bureau.32

[41] Dr Vine also referred to a study he had participated in evaluating oral fluid testing in track riders and jockeys.

Comparison of the positive rate of the oral fluid laboratory tests with the positive rate for urine samples collected from the same population of individuals showed that oral fluid testing gave approximately one tenth the number of positive samples. The conclusion reached from this study was that oral fluid testing, even when conducted using a sophisticated laboratory based testing procedure would produce few positives compared with urine testing and would, in many cases, fail to detect the presence of drugs even when used recently.33

[42] During his oral evidence, Dr Vine agreed that

On any reasonable level the urine drug testing is not correlated with impairment and urine drug levels don’t necessarily indicate any level of impairment34.

[43] Dr Vine was also asked whether the oral fluid Australian standard was more or less correlated with impairment in terms of its cut off or target levels.

There’s still no clear correlation with impairment. Detection of a drug in oral fluid usually would indicate more recent use than perhaps would be the case in urine, but that doesn’t necessarily correlate with impairment on any accurate basis. One can envisage that obviously if a drug has been taken recently there is more chance that that that drug is having an impairing effect, but the correlation is not all clear cut, yes35.

[44] Dr Vine also agreed that benzodiazepines are detectable in oral fluid.36

[45] During Dr Vine’s examination-in-chief Shell’s representative tendered a table prepared by the witness.37 In effect this reproduced a table contained in Professor Drummer’s statement which compared detection times in urine and oral fluid. However it added a fourth column which Dr Vine stated set out the detection time in oral fluid when tested to the levels
specified in AS 4760.

[46] Dr Vine referred in his oral evidence to amphetamine type drugs which he stated have a hang over effect:

…when those drugs are consumed their initial effect is as central nervous system stimulants and they cause euphoria, they cause heightened awareness, heightened alertness, possibly a whole range of physiological effects that make people feel they’re capable of doing more than they’re really capable of… However, the downside of amphetamine abuse is that that’s then followed by effectively a hangover period in the same way you get with alcohol. So when people have taken amphetamines they will then suffer afterwards a period where they’ll become exceedingly fatigued, they will suffer from lack of ability to concentrate, they will have – they’ll be inattentive, they’ll be possibly quite lethargic in the way that they do things and those effects may well be present and constitute a risk to safety after the time that the drug is detectable in oral fluid. However it is very likely that during that time when those effects are present the drug would still be detectable in urine.38

[47] Dr Vine also stated that one of the difficulties with oral fluid testing is:

That a regular user of cannabis may well not have the drug detectable in oral fluid samples beyond 12 or possibly 24 hours but that person may, by the very nature of their lifestyle, if they are regular users of cannabis, may well constitute a risk in the work place.39

[48] Dr Vine expanded on this during his cross examination:

…if you have evidence from urine drug testing that an individual is a regular cannabis user, then clearly I think as an employer you would have some concerns about potential impact of that employee in terms of workplace risks and workplace

And why? ---Because cannabis is an impairing drug, and if they’re using it sometimes, they may be using it at times when they are doing safety critical procedures and you can’t test everybody all the time. So if you have knowledge that you have employees within your organisation who are clearly regular users of impairing substances, you would naturally have some concern about their contribution to workplace safety.40

[49] Dr Vine agreed that urine testing does not necessarily provide proof of recent use, whereas oral fluid does.

[50] Dr Teri Lillington has practiced as a medical specialist in the field of occupational medicine for 22 years, the last 16 of which have been spent in the oil and gas industry. Dr Lillington was involved in the development of Shell’s D&A Policy.

[51] In her written statement41, Dr Lillington gave a number of reasons for supporting the use of urine testing. She indicated that laboratory oral fluid testing does not cover the range of drugs that can be tested for in urine. She specifically mentioned Benzodiazepines as being poorly excreted in oral fluid and therefore unable to be tested effectively using that method.42

[52] Dr Lillington also stated that there is evidence that the accuracy of testing for cannabis via on-site oral fluid testing is lower than that found with urine testing, leading to both false negative and false positive tests, with a range of negative consequences.43
[53] Dr Lillington’s statement concluded:

Of the current available methodologies for workplaces, urine testing provides the most accurate and broadest spread of drug and metabolite analysis and in the workplace where appropriate facilities are readily available, it is entirely feasible.
Oral fluid testing is acknowledged to be generally easy in terms of collection (although some difficulties can also exist if a person has a dry mouth – dry mouth can be found as a result of dehydration and nervousness and is a common side effect of methylamphetamine use) but the accuracy for many drugs is not as good as urine testing, and oral fluid cannot be tested for as broad a range of drugs as urine. I refer here to drugs of concern in a workplace setting.44

[54] Dr Lillington did however indicate that Shell Australia would continue to monitor the development of oral fluid testing. When accuracy and range of testing is equivalent to that of urine testing, then Shell may consider the use of this methodology as part of the application of the D&A Policy45.

[55] Attached to Dr Lillington’s statement was a paper prepared by her in April 2008 on oral fluid and urine drug testing. This included the following:

Due to the difference in detection times between oral fluid and urine, it is generally accepted that oral fluid is a more accurate representation of recent use than urine. It is on this basis that the suggestion is sometimes made that this may therefore be able to be correlated more closely with impairment. However, due to the lack of research, it is well recognised that currently this correlation does not exist.46

[56] The paper also refers to a review of the results of the first year of the Victorian police roadside-testing programme (published by Professor Drummer and others) and commented:

The data from the study shows that false positives and false negatives for cannabis in particular are high. This supports the general literature view that oral fluid testing for cannabis shows poor accuracy.47

[57] The paper concludes:

This evidence suggests that whilst THC has a significant effect on capacity to drive safely, it is the substance that is most poorly detected using oral fluid testing. It also supports the general understanding that saliva testing as a tool in management of impairment is ill-founded.48

[58] During her examination-in-chief, Dr Lillington made certain corrections to her statement. In particular, in the light of Professor Drummer’s evidence, she conceded that using appropriately sensitive laboratory equipment it is possible to detect the presence of benzodiazepines in oral fluid.49 Dr Lillington also clarified that the comparisons she made in her April 2008 paper between oral fluid and urine testing were based on on-site rather than laboratory-based testing of oral fluids.50

[59] Under cross examination she indicated that she did not know whether all of the drugs that are tested for in urine in the laboratory can be tested for in oral fluid in the laboratory, nor did she know if it was possible under the Australian standard to have additional drugs added to those already listed.51 She said there was nothing in Professor Drummer’s statement that she would take issue with52.

[60] Professor Olaf Drummer is Head of Forensic & Scientific Services, Victorian Institute of Forensic Medicine. He is also the Adjunct Professor in Forensic Medicine at Monash University. He has a background as a biochemical pharmacologist and toxicologist.

He has been the Convenor of the Oral Fluid Standards Committee of Standards Australia since 2004.

[61] In his written statement53 he outlines what he considers to be the relative advantages and disadvantages of testing in urine and oral fluid, in the context of the purposes of drug testing for workers.

[62] According to Professor Drummer, the presence of drugs in oral fluid when compared
to urine is much more similar to the concentration in blood.54

[63] Professor Drummer stated that for many drugs their presence in urine persists for one to three days after consumption. He continued :

In general drugs can be detected in urine well beyond the time the drug is having a significant biological effect. For example, cannabis, when smoked, the behavioural and physiological effects are generally no longer detectable after 4 hours. These 
pharmacological effects are caused primarily by THC. This substance is detectable in blood for several hours post dose by common analytical techniques. Beyond this time only trace amounts are detected. The rapid decline in blood concentrations is determined by its rapid uptake in tissues. Metabolism of this substance proceeds slowly allowing the detection of the acid metabolite (carboxy-THC) in urine for some days post consumption. In heavy users urine concentrations of carboxy-THC
can persist for weeks following the last consumption.55

[64] Professor Drummer presented a table summarising the average detection time in oral fluid for common drugs of abuse and showed how these compare to urine.

[65] For example, the table showed that the detection time for cannabinoids in urine is 3 to 28 days, whereas for oral fluid it is up to 12 hours. For cocaine metabolites, the figures are 1 to 3 days compared to up to 12 hours and for methamphetamine 2 to 5 days compared to up to 24 hours.56

[66] In his oral evidence, Professor Drummer took issue with some of the evidence given by Dr Vine. In particular he referred to a range of recently published research in rejecting Dr Vine’s claims concerning the detection times in oral fluid tested to AS 4760. For example, while Dr Vine claimed a detection time of 2-4 hours for cannabinoids, research cited by Professor Drummer pointed to 6 or more hours. Her considered Dr Vine’s figures for detection times as generally on the low side.57

[67] Professor Drummer acknowledged that his table

…wasn’t designed necessarily to be definitive, because it would vary a lot on individuals on doses and so and so forth [sic], but indicate that the detection time are hours, or several hours by and large, not two or three or more days. And that distinguishes that oral fluid is more likely to indicate recent use rather than past use.”58

[68] Professor Drummer’s evidence was that the choice of specimen (e.g. urine or oral fluid) should be dictated by the reason for the testing59.

If the employers of the workplace seek to establish who are using illegal substances then urine testing is quite appropriate. Indeed hair testing may be even more appropriate since the desire is to look as far back into their recent history to
ascertain use of any prohibited substance.

Urine testing is well established for this purpose (as is hair). It is relatively easy to collect and the testing is also relatively cheap. It will allow employers to see if any of the banned substances have been used in the (usually) days prior to collection of urine…

Urine testing will not indicate when a drug was used, or how much. Nor will it be able to indicate if the worker was impaired at the time the collection occurred.

If drug testing is used to indicate recent use of drug at a time when impairment is most likely, then urine is not an appropriate specimen. This testing could be dictated by an accident at the workplace or some other incident where drug testing may (quite properly) be applied to exclude (or include) this possible causative factor in the investigation.

In this situation drug tests should best be conducted in blood. This is often not practicable or convenient. Oral fluid is the next best specimen and as indicated earlier there is a reasonable relationship between a blood and oral fluid concentration whereas there is no relationship between blood and urine concentrations.60

[69] In response to a question from the Commission about Dr Vine’s evidence concerning how long impairment lasts after consumption, including any “hangover effects”, Professor Drummer said I mean there’s no easy answer to this, your Honour.

Clinical – if you look at a clinical – if you look at an individual clinically, not that I’m a doctor, but assessing clinically in terms of their functioning, for the most part these drugs, the effects on impairment, whether they’re to do with cognitive function, gross cognitive function or motor skills, by and large last for hours. If you put them into a flight simulator as this…here has for that study, you can see fine changes in some individuals and some times. Unfortunately we’re all different so you can’t do all – test all of us in this courtroom and assess whether we’re normal, because we’re all different in terms of our function, our background, our cognitive skills we’ve developed over a lifetime…So there’s a lot of variation on the norm…

…All I can say is on the one hand urine will detect somebody using cannabis yesterday, although they won’t be impaired today, and that might or might not be a good thing, depending on the reason for doing the testing…

…But for the most part clinical impairment for drugs of interest that we’re talking about here, cannabis, cocaine, heroin, for example, amphetamines, that normally last for hours and as a gross impairment, not days. But again it depends. You know,
if the concern is that anyone using a drug is a type of concern or possible concern to the industry or to the workforce, that – then urine testing would be better. If we’re looking at recent use and possible impairment, then if you get a choice between
those two specimens and that was the only choice you have, well oral fluid would give you a better indication of that.61

[70] Professor Drummer submitted that the analysis of biological specimens is best conducted in a laboratory accredited to perform the testing. An Australian standard (AS4308) provides guidance on the collection and testing of urine for selected drugs of abuse, and a number of laboratories are accredited to this standard by the National Association of Testing Authorities (NATA). In the case of oral fluid drug testing there is also an Australian standard (AS4760).62

[71] Professor Drummer indicated that some manufacturers provide kits for on-site testing of drugs (both using urine and oral fluids). However for maximum reliability testing should occur in the laboratory.63

[72] Professor Drummer indicated that while a larger number of laboratories can test for drugs in urine compared to oral fluid an increasing number of laboratories have either developed or are developing the technology for testing drugs in oral fluid. He stated that:

The weakness in the oral fluid drug testing is at the collection point. Companies will need to develop procedures for the collection, transportation, and if necessary onsite testing as they would for urine testing. These procedures can be accredited
under AS4760.64

[73] Professor Drummer’s written statement concluded that:

The choice of urine as the specimen to determine the presence of certain (banned) substances is widespread and is a perfectly adequate specimen to determine past use of drug, but it does not provide any proof of recent use; use that might render the worker unsafe to perform their duties.

Oral fluid does provide a valid alternative specimen to detect drugs used much more recently at a time when drug use might render a person unsafe to perform their duties….65

[74] In his oral evidence, Professor Drummer indicated that he disagreed with Dr Lillington’s evidence that certain drugs such as benzodiazepines could not be tested for using oral fluids. He rejected her statement that benzodiazepines were poorly excreted. In fact he stated they are excreted quite easily.

Indeed he was not aware of any drug that was not present in oral fluid. He indicated there were a number of reasons why benzodiazepines had not been included in the oral fluids Australian standard; however he disagreed that oral fluids could not be used to test for their presence.66

[75] Professor Drummer indicated that there was nothing to stop laboratories adding benzodiazepines (or other drugs of interest) to their testing program. They would just have to choose a concentration above which they would report to their client. This would still be regarded as conforming to AS 4760.67

[76] During cross examination, Professor Drummer indicated that he was only aware of some laboratories seeking accreditation under AS 4760; rather than any having already been accredited.68 During re-examination he also emphasised that accreditation related not only to the laboratory but also the collection agency. He estimated that it would take months to arrange for a laboratory to be accredited.69

[77] Mr Mark McGrath has been employed at Shell Clyde Refinery for around 20 years, first as an Operator, and for the last seven years as a Relief Senior Operator. He has held various positions in the Refinery Operators Group (ROG) of the CFMEU, as well as having been Chairman of the OH&S Committee. In his written statement he indicated he had numerous concerns about the validity of the risk assessment process that Shell used to designate which positions would be subject to drug and alcohol testing.

[78] Mr McGrath’s statement referred to a number of non designated positions which he considered should be subject to testing because of their implications for safety or the environment. He stated that in his view

Almost every role on site has a very real potential for health and safety risk if the incumbents had substance abuse problems and attended work impaired by drug or alcohol to perform such roles.70

[79] Following the preparation of Mr McGrath’s statement, Mr McKenzie prepared a supplementary statement71. In this second statement Mr McKenzie indicated that there had been a “challenge session” in relation to the designation of positions at Clyde Refinery, Gore Bay Terminal and Geelong Refinery. As a result of this session a number of additional positions were designated. These included many – though not all – of the positions highlighted by Mr McGrath in his statement.

[80] During his examination-in-chief Mr McGrath expressed concern that anyone who could walk around the site unescorted could be a safety risk. For example, with regard to the incumbent of one particular position he commented:

So not only that safety side of it there, he also, the incumbent as per that, walks around the site unescorted, frequently, looking at things on plant, looking at this.

He’ll just say, “I’m going out on a plant”. I mean, if he was to walk through a plant affected by drugs or alcohol, stagger, hit a valve, major gas leak, finds a source of ignition, bang, it’s a major explosion. I mean these things could happen.72

Submissions on behalf of Shell

[81] Mr Morris, on behalf of the company made four initial points.

[82] First, the Shell companies have onerous non delegable duties to ensure the health and safety of employees in all of Shell’s workplaces, flowing from State and territory occupational health and safety legislation.73

[83] Secondly, Shell has developed a drug and alcohol policy and procedures to assist in fulfilling these onerous health and safety obligations74.

[84] Thirdly, Shell’s D&A Policy is detailed, comprehensive and holistic. In particular, while the objective of the policy is to eliminate the risks to health and safety resulting from the use of alcohol and drugs75 it seeks to do so by changing employee behaviour by a combination of education, information, assistance, advice, and rehabilitation76.

[85] Random testing should be seen in the context of the policy as a whole. Critically, a positive test indicates a positive presence of alcohol or drugs that is a clear signpost for the need for management intervention.77

[86] Fourthly, the Commission should apply its general principles about intervening in the way management manages and regulates its business.

This is a case where the Commission should apply its well established consistently applied principles, namely, that it will not intervene in the way… the management of a business or enterprise regulates its business, unless it’s satisfied that the employer is acting unjustly or unreasonably, or that it’s imposing on its employees requirements that are unjust or unreasonable.78

[87] Mr Morris referred to the decision of the Australian Conciliation and Arbitration Commission in the XPT case79. At page 191 the Commission stated:

The principles which the Commission should apply in circumstances such as those before us have been the subject of a number of submissions to us and reference to a number of cases.

The main case relied upon by the State Rail Authority is the decision of Coldham J in the Airline Hostesses’ Case. In that decision Coldham J applied the test whether or not the work asked to be done was “. . . unjust . . . unreasonable, harsh or oppressive”. In adopting this test his Honour referred to a decision of Wright J in an appeal under the Public Service Arbitration Act.

In that case Wright J said “. . . this Commission, and the Arbitration Court before it, have throughout their existence acknowledged the right of an employer to manage and regulate his own business, subject to the protection of his employees from injustice or unreasonable demands”. In that case not only did Wright J use that expression but Williams and Franki JJ in their separate decision referred to “. . . the right of an employer to manage and regulate his own business, unless in doing so he imposes unjust or unreasonable demands upon his employees” and said:

“This approach has been accepted by the Commission and the Arbitration Court since the Conciliation and Arbitration Act became operative and has been reiterated from time to time since then.” It is not clear why Coldham J added the words “harsh” and “oppressive”. It seems to us that the proper test to be applied and which has been applied for many years by the Commission is for the Commission to examine all the facts and not to interfere with the right of an employer to manage his own business unless he is seeking from the employees something which is unjust or unreasonable.

The test of injustice or unreasonableness would embrace matters of safety and health because a requirement by an employer for an employee to perform work which was unsafe or might damage the health of the employee would be both unjust and unreasonable. The ACTU submitted to us that we should apply the test as to whether the demand of the employer was just and equitable having regard to all the circumstances. It is our view that under any given set of facts the test suggested by the ACTU would not lead to a different decision from the test which the Commission has applied over time. Accordingly in reaching our decision we have approached the matter from the point of view of making a judgment whether the request of the SRA that the XPT be manned by one man is unjust or unreasonable80.

[88] Mr Morris submitted that this is the test to be applied on each of the specific questions the Commission is asked to arbitrate in this matter.81 He then took the Commission to the decision in the BHP Iron Ore Case by the Western Australian Industrial Relations Commission in Court Session82 to demonstrate the application of this test to a case dealing with drug and alcohol policies in a workplace environment.

[89] At page 164, the Commission stated:

Clearly, it is not for the Commission to manage the affairs of the Company as if it was an alternate or surrogate manager. As counsel for both parties correctly suggest, the Commission is required to make an objective assessment of the Programme. As a consequence, it is not the function of the Commission, on this occasion, to determine what is the most ideal drug and alcohol programme in the circumstances, but to determine whether this Programme satisfies acceptable industrial standards. Further, as counsel for the Union suggests, that assessment must be made having regard to industrial relations considerations.83

[90] In that case, the Commission concluded:

In our view, the Programme cannot be said to be unreasonable, harsh or unfair. On the contrary, we consider it to be both fair and reasonable.84

[91] Mr Morris submitted that, as provided in the D&A Policy, drug and alcohol testing should only apply to designated positions, established on the basis of a sound risk management process. The intrusion on privacy inherent in drug and alcohol testing is only justified where there is a proper risk management rationale. The methodology for designating positions at Shell is rigorous and allows for periodic reviews.85

[92] In relation to contractors, Mr Morris submitted that there is no basis for saying it is unjust or unreasonable to test Shell employees until all employees of all contractors are subject to the same regime.86 The intention is that contractors are required to demonstrate that they have policies that are consistent with Shell’s D&A Policy. The policy is being applied progressively to contractors.87 While it would be unreal of Shell to say to every contractor that on a particular date they must have a policy “that looks like ours”88 contractors representing between 85 and 90 per cent of the personnel working for contractors

…within short order have in application and implementation policies and procedures that are consistent with Shell’s and help achieve Shell’s ultimate objective, which is to ensure the health and safety of all its people and all people
working on its premises.89

[93] In relation to the issue of the method of testing, Mr Morris agreed with the proposition put forward by Professor Drummer, that the starting point for dealing with this issue is the purpose and the role of drug testing, as the kind of medium to be used for testing needs to suit the objectives of the policy as a whole.90

[94] Testing under the policy, according to Mr Morris, is intended as a credible deterrent to use drugs that may impair performance and may cause a risk to health and safety. It is intended as a sign post to employees for whom management intervention is appropriate (not necessarily a sign post that the employee is impaired).91

[95] In response to a question from the Commission, Mr Morris said that “acute impairment” following or in close proximity to ingestion or consumption of a substance that has an acute and immediate effect should not be the only matter of concern to the employer. There is also the “hangover” issue, as described by Dr Vine in his evidence.

The third, if you like, category of impairment or potential impairment is that associated with chronic use and the medical evidence is abundant about the long term consequences of using some substances, whether it’s alcohol or other drugs.

Now, what should a responsible employer do in that case where it’s looking and professing a holistic policy? Should it simply say we’re only interested if you’re acutely intoxicated? From a risk management point of view it’s a very superficial approach. In my submission, and this is consistent with the policy and the evidence that Mr McKenzie and the evidence that Dr Lillington gave, is that we’re actually interested more in influencing the behaviour of people around the use of drugs and alcohol to eliminate the risk, whether it’s acute impairment, it’s a lingering ongoing impairment or whether it’s a chronic effect. So a responsible employer professing a policy like Shell’s must be interested in more than whether X comes to work intoxicated.92

[96] Mr Morris submitted that urine testing is in extensive use around Australia, it is a well established and reliable methodology supported by an Australian and New Zealand standard of some 20 years of development. It can be administered or applied in accredited laboratories, applies to a wide range of substances, and is well suited to Shell’s “holistic policy approach”.93

[97] Mr Morris submitted that oral fluid testing, by contrast, is less well developed as a technique, is not widely used in Australia, the relevant standard has target levels for a smaller range of substances, and there are as yet no laboratories accredited to undertake testing in accordance with the standard. Because it only detects more recent use it could deprive the company and the individual of the opportunity for intervention, counselling, advice, medical assistance and so on.94

[98] In conclusion, Mr Morris submitted that there was no case to say that it would be unjust or unreasonable to require urine testing.

Submissions on behalf of the Construction, Forestry, Mining and Energy Union

[99] Ms Gray, on behalf of the CFMEU, agreed that the test, as outlined by Mr Morris, was the appropriate one for the Commission to apply in the current case.

“Yes, and we believe that in respect to each of the questions the subject of this hearing, your Honour, that the company’s position is unfair and unjust to the employees95.

[100] Ms Gray presented a written submission on the evidence. She drew attention to the acknowledgement by Mr McKenzie that Mr McGrath’s witness statement caused him concern in respect to the positions raised and so he challenged these positions with the result that they became designated.

[101] Ms Gray also noted that while the company had indicated that 85 per cent of contractor employees would be subject to an equivalent policy almost straight away they had been unable to give an expected implementation date for the remaining 15 per cent.

[102] Ms Gray submitted that:

The review of positions to determine those which would remain designated and subject to testing was conducted in March and resulted in numerous positions previously non-designated becoming designated. The further challenge following the
company receiving Mr McGrath’s evidence also led to numerous positions which had previously been non-designated becoming designated. In other words the company, in applying the same test and criteria to the same positions delivered a different result in March and June. Positions previously regarded as not posing a credible risk or where the objective application of risk management techniques assessed that no risk existed, became positions objectively assessed as having risk….

Such high level of error in either the risk assessment or its application to the positions at Clyde Refinery or Gore Bay is of grave concern to the Union and its members.96

[103] Ms Gray referred to a paper produced by the Privacy Committee of New South Wales in October 1992 concerning drug testing in the workplace97. Ms Gray acknowledged that the introduction of random drug testing involved a balance between privacy and workplace safety and reiterated that the union was not opposed to random testing if it was done in a fair manner.98

[104] In particular, Ms Gray submitted that using urine samples detects drug use rather than recent use which is associated more with impairment – and is an unnecessary and unfair incursion into employees’ private lives.

[105] Ms Gray noted that the Civil Aviation Safety Authority (CASA) was introducing random drug testing using oral fluids and submitted that the use of oral fluid testing would be consistent with the rest of Shell’s Drug & Alcohol Policy.


[106] I accept that the test for the Commission in this case is as outlined by Mr Morris and agreed to by Ms Gray. This means that the role of the Commission is to consider whether it would be unjust or unreasonable for Shell to:

  • Implement random drug testing of operators under its D & A Policy using urine as opposed to oral fluid testing;
  • Apply random drug and alcohol testing under its D & A Policy to operators at Clyde Refinery and Gore Bay Terminal unless such testing is applied to all Shell employees at those two locations;
  • Apply random drug and alcohol testing under its D & A Policy to operators at Clyde Refinery and Gore Bay Terminal unless such all employees of contractors engaged at those two locations are subject to such testing.

[107] The Commission was presented with considerable expert evidence, particularly in relation to the first of these issues. When it comes to technical or scientific matters, I have generally accepted the evidence of Professor Drummer, and have preferred it to that of the other expert witnesses where there is a conflict. It is quite clear from Professor Drummer’s curriculum vitae that he is a leading expert in the fields of drug testing and the effects of drug use on the human body and human behaviour. While he is not a medical practitioner he has a PhD in medicine from the University of Melbourne and has published a large number of articles in the biochemical and medical literature. His expertise on testing for drugs using oral fluids is obvious, especially given that he is the convenor of the Oral Fluids Standards Committee of Standards Australia. His expertise is considerably greater than that of Dr Lillington, who was indeed forced to amend part of her evidence in the light of that of Professor Drummer. While Dr Vine also has genuine expertise and is widely published in the field of the analysis of drugs and other biologically significant substances his expertise on the effect of drugs on humans is inferior to that of Professor Drummer. While he said during his examination-in-chief that he had “taken an interest in the physiological effect of drugs”99 he conceded that he lacked medical qualifications. While I have no doubt his professional interest in the issue is genuine, it is not of the same order as that of Professor Drummer.

[108] It is clear, based on Professor Drummer’s evidence, that at least some of the reasons that the Shell witnesses put forward for preferring testing using urine as opposed to oral fluids are not well founded – at least when considering the use of laboratory (as opposed to onsite) testing of oral fluids, the approach proposed by the union.

[109] Mr McKenzie, based on his own evidence, played a key role in the development of the revised D & A Policy. In his statement100he said that Shell had concluded “based on advice from Dr Lillington that testing for drugs via saliva does not yet meet the requirement that any test method be accurate, verifiable, reliable, repeatable and consistent in application, with a minimum of false negatives and false positives.”

[110] It is clear from the evidence that this conclusion was mainly based on Dr Lillington’s examination of the literature in relation to on site testing of oral fluids and is not correct in relation to laboratory oral fluid testing. There appear to be no grounds for Shell’s view that laboratory-based oral fluid testing is not accurate, verifiable, reliable, repeatable and consistent in application. The concern about false negatives and false positives does not arise with laboratory testing.

[111] Another reason given by Mr McKenzie for preferring urine to oral fluid testing is that

Shell has been advised through Dr Lillington that the suite of drugs and metabolites tested for with oral fluids is more limited than that able to be tested for with urine. This gives rise to the need for a greater number of tests to cover the same range of drugs, with additional costs and complexity added to the process and even then the range of drugs tested for is limited. Again a credibility issue arises101.

[112] These conclusions are not consistent with the expert evidence given to the Commission by Professor Drummer. Dr Lillington specifically conceded that her evidence about the inability to test oral fluids for the presence of benzodiazepines was incorrect.102 It is clear from Professor Drummer’s evidence that while a decision was taken by the Oral Fluids Standards Committee not to include target concentrations for certain drugs, these drugs can nevertheless be tested for using oral fluids. Moreover it would be consistent with the standards to test for these drugs using oral fluid. This is an issue to which I return below.

[113] Mr Morris, on behalf of Shell, made no attempt to challenge Professor Drummer’s evidence on the accuracy or reliability of oral fluid testing and did not defend most of the reasons given by Mr McKenzie for Shell preferring urine over oral fluid testing.
[114] It is clear from the evidence that the key difference between oral fluid and urine testing is that the former will detect drug use in the previous few hours, whereas the latter will detect drug use over the previous few days.

[115] Dr. Vine claimed that in some cases the “window of detection” of oral fluid testing being so short it could fail to detect some employees who might still be impaired – particularly as a result of a “hangover effect”.

[116] I put this issue to Professor Drummer. My conclusion – based on his and Dr Vine’s evidence – is that, while conceivably there might be some very limited effect over a longer period, significant impairment only occurs for a few hours after the ingestion of drugs. Thus while both oral fluid and urine testing will usually pick up anyone who has taken drugs in the previous few hours (and thus may well be impaired), urine testing will also pick up employees who have ingested drugs over the previous few days and are thus highly unlikely to be impaired at the time of the testing. Moreover urine testing will not indicate when the
drugs were taken.

[117] Neither party in this dispute sought to argue that random testing for drugs (or alcohol) was unjust or unreasonable. However both parties also recognise that random testing is an intrusion on the privacy of the individual which can only be justified on health and safety grounds. The employer has a legitimate right (and indeed obligation) to try and eliminate the risk that employees might come to work impaired by drugs or alcohol such that they could pose a risk to health or safety. Beyond that the employer has no right to dictate what drugs or alcohol its employees take in their own time. Indeed, it would be unjust and unreasonable to do so.

[118] Mr Morris emphasized, with some justification, what he described as the “holistic” nature of the D&A Policy. The policy strongly emphasises education and rehabilitation. According to the policy document, testing is designed to assist in identifying people who may either be unfit for work or at risk of being unfit for work; discourage people from coming to work when they may be impaired by alcohol or other drugs; and raise awareness about the effects of alcohol and other drugs in the workplace.103

[119] Mr Morris acknowledged that a positive test did not necessarily indicate impairment – but it did provide a “signpost for management action”. The question is whether management has a legitimate right to take action against an employee based purely on what he or she may have done during their own time, if there is no impairment.

[120] While a positive drug test under the policy does not in isolation lead to dismissal, it does constitute a breach of the D&A Policy, and repeated breaches can lead to dismissal. As a consequence positive tests caused by taking drugs outside working hours, at a time when they are very unlikely to indicate significant impairment at work, could form the basis of the termination of an employee’s employment.

[121] I note that the Western Australian Industrial Relations Commission in Court Session in the BHP Iron Ore Case104 specifically found that a random testing programme using urine samples was justified on safety grounds – and indeed was both fair and reasonable. However that case was decided ten years ago. Since then oral fluid testing has become available and an Australian standard for oral fluid testing has been developed. The question now is whether it would be unjust or unreasonable for the company to implement a urine based random testing regime with its wide “window of detection”, with all that implies for
interfering with the private lives of employees, when a much more focussed method is available, where a positive test is far more likely to indicate actual impairment, and is far less likely to detect the use of drugs at a time that would have no consequential effect on the employee’s performance at work.

[122] My conclusion is that the implementation of a urine based random drug testing regime in these circumstances would be unjust and unreasonable. However I have two qualifications.

[123] First, the evidence is that, at this stage, no laboratories have as yet been accredited under the relevant Australian standard. I note that Dr Vine’s laboratory has applied for accreditation and the lack of accredited laboratories is likely to be resolved in the relatively near future. Nevertheless the company cannot reasonably be expected to implement a random drug testing system based on oral fluids until laboratories such as RASL have been accredited.

[124] Secondly, there are drugs that the company may wish to test for (such as benzodiazepines) for which AS 4760 does not contain target concentration levels. The company should not be expected to implement an oral fluids based regime until it has the agreement of the union and the laboratory it wishes to use on what other drugs it wishes to
test for and what would be an appropriate target concentration level.

[125] Once these two issues are satisfactorily resolved, any random drug testing should be conducted using oral fluids. Until then it would not be unreasonable for the company to implement a urine based testing regime on an interim basis.

[126] I now turn to the issue of whether it would be unjust or unreasonable for the company to implement random drug testing to any employees at Clyde or Gore Bay unless all employees are included.

[127] As noted above, random drug testing involves a trade off between the privacy of the individual and workplace safety. Employees should not have to suffer the intrusion into their privacy involved in random testing unless there is a good safety reason for doing so. Accordingly there is no basis to conclude that it would be unjust or unreasonable for some
employees to be subject to testing and not others, if there is an objective reason for this distinction.

[128] The adoption of a risk assessment process to identify which employees should be subject to random testing is entirely appropriate. Moreover the methodology contained in the D&A Policy does not seem unreasonable. Clearly there is scope for positions to be reviewed to ensure they are designated appropriately. I would urge the members of ROG to provide
any information they have about potential for safety risks to the company.

[129] Finally, I turn to the issue of contractors. I note that the D&A Policy already provides that contractors must demonstrate that they have their own consistent policies. Shell has indicated that this principle is being implemented progressively. This is not
unreasonable. However, it would seem appropriate to put some time line around this process. Shell should be able to advise the union within six months of the date of this decision that all contractors have adopted consistent D&A Policies. If there are any who have not done so they should be identified and a sound reason given for their failure to comply with the target date.


Mr Adrian Morris for Shell Refining (Australia) Pty Ltd, with Ms H Fairhall and Mr H Powell.
Ms Judy Gray for the Construction, Forestry, Mining and Energy Union, with Mr W Deveccis
Hearing details: Sydney 26, 27 June 2008, 14 July 2008

1 PN783-785
2 Exhibit Shell 1, paragraph 13
3 Exhibit Shell 1, paragraph 15
4 Exhibit Shell 1, paragraph 24
5 Exhibit Shell 1, paragraph 31
6 Exhibit Shell 1, Attachment NM-2, page 26
7 Exhibit Shell 1, Attachment NM-2, pages 27-28
8 Exhibit Shell 1, Attachment NM-2, page 28
9 Exhibit Shell 1, Attachment NM-2, page 30
10 Ibid.
11 Exhibit Shell 1, paragraph 54
12 Ibid.
13 Exhibit Shell 1, Attachment NM-2, page 33
14 Exhibit Shell 1, Attachment N-M2, page 36
15 Exhibit Shell 1, Attachment NM-2, page 37
16 Exhibit Shell 1, Attachment NM-2, page 40
17 Ibid.
18 Exhibit Shell 1, Attachment N-M2, pages 40-41
19 Exhibit Shell 1, Attachment N-M2, page 41
20 Exhibit Shell 1, paragraphs 50-51
21 Exhibit Shell 1, Attachment N-M2, pages 50-51
22 Exhibit Shell 1, paragraphs 65-75
23 Exhibit Shell 3, paragraph 10
24 The Australian and New Zealand standard for urine testing released in 2001
25 The revised Australian and New Zealand standard for urine testing released in 2008
26 The Australian standard for oral fluid testing
27 Exhibit Shell 3, paragraph 20
28 Exhibit Shell 3, paragraphs 22-23
29 Exhibit Shell 3, paragraph 26
30 Huestis MA and Cone CJ, Relationship of delta-9-tetrahydrocannabinol concentrations in oral fluid and plasma after controlled administration of smoked cannabis, 2004, J. Anal. Toxicol., 28(6), pp 394-9
31 Exhibit Shell 3, paragraphs 27-29
32 Newman DG, Cannabis and its Effects on Pilot Performance and Flight Safety: a Review, Australian Transport Safety Bureau Research Report March 2004
33 Exhibit Shell 3, paragraph 37
34 PN 509
35 PN510
36 PN517
37 Exhibit Shell 4
38 PN546
39 PN548
40 PN 601-602
41 Exhibit Shell 5
42 Exhibit Shell 5, paragraph 24
43 Exhibit Shell 5, paragraphs 25 -32
44 Exhibit Shell 5, paragraph 34
45 Exhibit Shell 5, paragraph 35
46 Exhibit Shell 5, page 16
47 Exhibit Shell 5, page 17
48 Exhibit Shell 5, page 18
49 For example, PN695
50 PN716-721
51 PN727-728
52 PN750
53 Exhibit CFMEU 4
54 Exhibit CFMEU 4, paragraphs 15-17
55 Exhibit CFMEU 4, paragraph 19
56 Exhibit CFMEU 4, paragraph 22
57 PN805-811
58 PN811
59 Exhibit CFMEU 4, paragraph 28
60 Exhibit CFMEU 4, paragraphs 29-33
61 PN 814-818
62 Exhibit CFMEU 4, paragraphs 34-36
63 Exhibit CFMEU 4, paragraphs 34-41
64 Exhibit CFMEU 4, paragraphs 42-45
65 Exhibit CFMEU 4, paragraphs 46-47
66 PN801-802
67 PN901-902
68 PN 860
69 PN 903-904
70 Exhibit CFMEU 6, paragraph 31
71 Exhibit Shell 2
72 PN949
73 PN1147-1155
74 PN1155-1165
75 PN1247
76 PN1166
77 PN1183
78 PN1252
79 (1984) 295 CAR at pages 188 - 193
80 References deleted
81 PN1265
82 (1998) 82 IR at 162 - 171
83 Ibid. at 164-165
84 Ibid. at 171
85 PN1318-1327
86 PN1391
87 PN1357
88 PN1362
89 PN1375
90 PN1395
91 PN1416
92 PN1426-1427
93 PN1489
94 PN 1491
95 PN1583
96 CFMEU submission on evidence, pages 9-10
97 Exhibit CFMEU 8
98 PN1611
99 PN545
100 Exhibit Shell 1
101 Exhibit Shell 1, paragraph 72
102 PN695
103 Exhibit Shell 1, Attachment NM-2, page 30
104 (1998) 82 IR at 162 - 171

« Back to Legal Rulings