Workplace Testing

Policy on Alcohol and Other Drug Testing in the Workplace

A policy and strategy to address unsafe work practices related to alcohol or other drug use has been developed and supported by the Committee. If an employee is observed to be working or behaving in an unsafe manner, it is the responsibility of the safety committee to remove that employee from the workplace, in accordance with the policy of The Building Trades Drug and Alcohol Safety and Rehabilitation Program.

It is the policy of the Committee to actively oppose drug testing in the workplace. The method is costly, can undermine and divert resources away from other strategies, invades employee privacy and demands that the employee demonstrate innocence of drug use, not impairment. Breath testing for alcohol reliably indicates blood alcohol levels and raised blood alcohol levels have been demonstrated to be associated with impairments in a number of performance domains. Nevertheless, the assumption that breath testing for alcohol reduces workplace hazard has not been unequivocally supported, and no legislated levels of blood alcohol content for the workplace exist. Drug testing, in the form of urine testing, does not detect impairment, has not been demonstrated to reliably predict or reduce drug-related hazard or to reduce accident-risk, has not been demonstrated to be cost effective and, unless carefully conducted, can be error prone.


The Building Trades Group Drug and Alcohol Committee is concerned to prevent and reduce alcohol and other drug related harm in the workplace. Drug related impairment may contribute to workplace hazard. It is the proper duty of employers and employees to prevent and reduce such hazards. The Committee therefore supports the development and implementation of responses to hazardous and harmful alcohol and other drug use, based on the principles of workplace consultation, demonstrated efficacy and consistent with the concerns of privacy and Natural Justice.

On this basis, the Committee has developed policies; strategies and resources for responding to alcohol and other drug related harm in the workplace. The same considerations lead the Committee to conclude that it does not support the introduction or use of drug testing in the workplace

Rationale for Policy on Alcohol and other Drug Testing

Preventing and responding to impairment is the duty of employers and employees. Whether drug-testing employees is an appropriate and effective means to this goal is questionable.

Scientific consensus is that breath testing for alcohol can provide a reliable indication of level of alcohol intoxication. This conclusion is, of course, based on the assumption of accurate testing apparatus, regular assessment of the functioning of the apparatus, correct application of the technology and interpretation of results. Scientific consensus also supports the conclusion that raised blood alcohol levels (BAL) (e.g. 0.05mg% BAL and above for driving) are associated with impairments in functioning in realms such as judgment, reaction time and so on. However, it is also important to note that there are variations in this relationship. These variations are dependent on individual attributes (e.g. those with low alcohol tolerance will be impaired at lower BAL’s, and vice versa) and the nature of the task (e.g. even at higher blood alcohol levels routine, repetitive tasks can remain relatively unaffected, while responding to novel information may be distinctly impaired at lower levels). In addition, breath analysis will not identify impaired performance that can arise from “hangover” effects, even at 0.00mg% BAL.

The usual methods of urine testing for drugs, unlike breath testing for alcohol, do not measure the level of a drug in a person’s system. Urine testing generally involves the identification and measurement of drug metabolites. This creates a number of threats to the validity of the method.

First, the data on levels of drug use and the effects of drugs, other than alcohol, on work performance are limited and do not allow a clear conclusion about impairment. Second, there are no data reliably linking the detection of drug metabolites with performance. The appropriate conclusion of a positive test, from even the most robust methods is, at best, that the person was exposed to a drug in the recent past. “Recent past” can be defined as within broad parameters of several hours or even days, dependent on drug type. Third, for some drugs, such as cannabis, there is considerable individual variation in metabolism, influenced by factors such as body fat and frequency and level of drug use.

Scientific consensus is that urine testing does not provide any information on patterns or level of drug use, degree of dependence or mental or physical impairment. In short, the rationale for drug testing is to measure impairment. Drug testing, in the form of urine testing, does not do this, and so the rationale for the strategy is not supported.

Nevertheless, some authorities have argued that drug testing may identify potential and current employees at risk, deter drug use and reduce workplace hazard.

There are some reports that a positive drug test at initial employment predicts future performance. However, the more carefully conducted and robust studies have indicated that the measure is at best a poor indicator of future performance, especially of accident risk, and that the advantages have been exaggerated. The method has not demonstrated cost effectiveness in preventing and reducing drug related harm in the workplace.

A number of reports have suggested that the method may dissuade current employees from using drugs and consequently reduce drug-related hazard. A variety of testing methods have been proposed for current employees (e.g., testing for cause, random testing etc.). However, none of these methods have been demonstrated in scientifically controlled studies to reduce hazard or accident rates in the workplace. Thus, the efficacy of the method is unproven.

One unintended outcome of drug testing may be to shift patterns of drug use, with a subsequent increase in hazard. Although this is an untested hypothesis, some individuals may change from the use of drugs with a long biological half life (e.g. cannabis) to using drugs with a shorter half life (e.g. amphetamines) which are less easily detected but may be associated with greater hazard.

Finally, drug testing is costly and the cost-effectiveness of the method has not been demonstrated. This strategy can divert resources away from other safety initiatives, which may be much more cost-effective.

In summary, drug testing does not provide any indication of level of impairment and testing prospective and current employees has not been demonstrated in any scientifically controlled study to reduce drug use or drug related hazard in the workplace.

Drug testing has a number of other potential flaws. Anything less than 100% sensitivity and specificity of a test can result in a higher level of false positives. That is, there remains the possibility that some employees will be identified as drug positive when in fact drug use has not occurred. It is acknowledged that such errors are more likely with an initial screening test (e.g. immunoassay) and less likely when a positive immunoassay result is subject to a confirmatory test, such as gas liquid chromatography.

In addition, there have been reports that some laboratories have made errors in handling and interpreting results. Again, adherence to strict guidelines on urine handling and analysis have reduced the likelihood of such problems. Nevertheless, it is important to note that errors have occurred.

Drug testing is invasive and intrusive, no matter how carefully and courteously conducted. As already stated, the prevention and identification of hazard is the proper concern of employers and employees. An individual’s private behaviour is less clearly the concern of employers, especially when it cannot be demonstrated that urine testing allows identification of the existence or level of impairment.

Employees can be placed in the position of having private behaviour scrutinised with no demonstrable benefit. Providing body fluids for examination, whether under close observation or not, is embarrassing and invasive for many people, a problem compounded by the lack of demonstrable advantage to employees and employers.

Drug testing may undermine other methods to prevent and reduce drug-related hazard in the workplace. For example, drug testing can contribute to a culture of cover up and collusion and create an atmosphere of distrust. The Committee has developed strategies, which are based on changing workplace culture and drug use, using peer concern and action by workplace safety committees. It is considered that drug testing will undermine these methods.

The problems associated with interpretation of drug testing, including urine and blood analysis, can be illustrated in the case of testing for cannabis use and intoxication. The initial difficulty is that there is a substantial time delay between the subjective experience of intoxication and levels of blood THC (the psychoactive compound in cannabis). There are also substantial variations in the subjective experience of intoxication, for different individuals at the same blood levels of THC. Due to substantial variability in individuals, no “realistic limit of cannabinoid levels in blood has been set which can be related to an undesirable level of intoxication… To date there is no consistently demonstrated correlation between blood levels of THC and its effect on human mind and performance”. (Hall, Solowij and Lemon, 1994, p 36)

It is important to note that while the detection of THC in blood over 10-15ng/ml is suggestive of recent exposure to cannabis, it is not possible to define narrow parameters of “recent use” which have utility in determining level of intoxication or impairment. It is acknowledged, however, that if a blood sample is taken, and that blood sample indicates similar concentrations of THC and 9-carboxy-THC (a metabolite of THC) it could be indicative of use within the last 40 minutes and may be associated with intoxication. If the levels of 9-carboxy-THC are much higher than THC, consumption may be estimated to have occurred more than 30 minutes previously (Hawks, 1982; Perez-Reyes et al, 1982). However, even this interpretation is limited to naive users who have resting levels of zero. If one cannot make this assumption, estimation of time of use is almost impossible. “It is very difficult to determine the time of administration from blood concentrations of THC and its metabolites, even if the smoking habits of the individual and the exact dose consumed are known.”(Hall et al, 1994, p35).


In conclusion, the Committee does not support drug testing in the workplace. The method has a number of fatal weaknesses. It invades employee privacy and demands that the employee demonstrate innocence of drug use, not impairment. Drug testing does not detect impairment and has not been demonstrated to reliably predict or reduce drug-related hazard or to reduce accident risk. Importantly, it has not been demonstrated to be cost effective and, unless carefully conducted, can be error prone.

Key References

Allsop, S and Phillips, M (1996) Drug Testing in the Workplace: An Unfortunate Marriage. National Centre for Education and Training on Addiction, Flinders University of South Australia, Adelaide.

Hall, W, Solowij, N and Lemon, J. (1994) The Health and Psychological Consequences of Cannabis Use. National Drug Strategy Monograph Series No. 25. Australian Government Publishing Service, Canberra.

Hawks, R (1982) The constituents of cannabis and the disposition and metabolism of cannabinoids. In, R Hawks (ed) The Analysis of Cannabinoids in Biological Fluids. National Institute on Drug Abuse Research Monograph No 42. Rockville, MD. US Department of Health and Human Services.

McDaniel, M. (1987) Does pre-employment drug use predict on-the-job suitability?, in S.W. Gust & J.M. Walsh. Drugs in the Workplace: Research and Evaluation Data, National Institute of Drug Abuse Research Monograph 91, Rockville, MD. US Department of Health and Human Services.

Perez-Reyes, M, Guiseppi, S, Davis, K, Schindler, V and Cook, C. (1982) Comparison of effects of marihuana cigarettes of three different potencies. Clinical Pharmacology and Therapeutics, 31, 617-624.

Ryan, J, Zwerling, C and Jones, M. (1992) The effectiveness of preemployment drug screening in the prediction of employment outcome. Journal of Medicine, 34, 1057-1061.