Drug testing: How it works and what it can, and cannot, achieve

October 2017
Associate Professor Ken Pidd, Deputy Director (Research), NCETA

The recent announcement that the Federal Government plans to introduce a trial of compulsory drug tests for welfare recipients has created substantial media attention.  This media attention has led to public debate and controversy about the utility and effectiveness of drug testing.  Central to this debate and controversy are underlying assumptions about how drug testing works and what it can, or cannot, achieve.

Types of drug tests

The most common drug tests currently used in Australia are urinalysis and oral fluid (saliva) tests.  Australian standards exist for both of these tests.  These standards outline the procedures for specimen collection, and the detection and quantitation of drugs in either urine (AS/NZS 4308:2008) or oral fluid (AS 4760-2006).  


Urinalysis is one of the most researched drug test technologies.  It is also the least expensive of all drug test types. For most drug types, it can detect use that has occurred up to three days prior to the test.  One exception to this is for cannabis use, in which case occasional use that has occurred up to 6-8 days prior to the test can be detected, while for regular use the window of detection can increase to several weeks.  The main disadvantages of urinalysis are that it may not detect very recent drug use that has occurred in the past 2-4 hours, and urine specimens can be adulterated or substituted relatively easily.

Oral fluid (saliva)

Oral fluid testing is a relatively new technology that is increasing in popularity as a less invasive form of testing compared to urinalysis.  Oral fluid testing has a much shorter window of detection compared to urinalysis, and for most drug types it detects use that has occurred in the previous 2-3 days. While generally more expensive than urinalysis, it is less invasive and oral fluid is less easily adulterated or substituted.  The main disadvantage of oral fluid is its limited ability to detect cannabis, however rapid advances in technology may be overcoming this issue.

Hair follicle

Hair follicle testing has been proposed for the testing of welfare recipients, however, it is not commonly used in Australian drug testing programs.  The main reasons for this are the lack of Australian standards, its long window of detection, and prohibitive cost.  While detection times vary according to hair growth rates, hair follicle analysis detects most drug types for up to 3 months after use. Currently hair follicle testing can only be undertaken in a laboratory and there are few Australian laboratories that can conduct hair follicle testing.  Moreover, hair follicle testing in Australia is expensive (in excess of $1,000/test).  In addition, hair follicle testing is easily evaded if the donor chooses to remove all body hair.

Drug testing programs

A number of drug testing programs are commonly used in Australia.  These include:

  • Random testing
  • For-cause (targeted) testing
  • Post-accident testing
  • Pre-employment testing.

The choice of which testing program to implement varies according to the testing context.  For example, for-cause testing is used in workplace and sporting contexts, where the target of the test is suspected of drug use, or in clinical settings where it is necessary to determine if drugs are being used by the client/patient.  Post-accident testing occurs in the context of workplace and road traffic accidents, while pre-employment testing is implemented solely in the workplace.  Random testing is applied in workplace, roadside, and sporting contexts, and has been proposed for the testing of welfare recipients.  The main objective of random testing is to deter use. It is an inefficient method of detecting use in a given population.  Research based on national US drug prevalence rates indicates that random testing 50% of a given population would result in only 40% of daily users and 8% of less frequent users being detected (Dupont et al., 1995). 

The drug testing process

Regardless of whether it occurs in the workplace, the sporting arena, the roadside, in clinical treatment, or in the welfare setting, drug testing is a two-step process.   The first step involves an initial on-site screen, using a point of collection test device. Most point of collection test devices use immunoassay techniques that are less reliable and accurate than laboratory analysis.  For this reason, the second stage involves laboratory analysis to confirm the accuracy of any initial on-site screen that detects the presence of a drug.  Laboratory analysis typically involves more reliable and accurate mass-spectrometry techniques.

Good practice dictates that the drug testing process must comply with the relevant Australian Standards. The purpose of these standards is to ensure that the initial on-site screen and confirmatory laboratory procedures meet the requirements for reliable detection and quantitation of drugs in biological specimens.  The standards cover procedures for specimen collection, storage, handling, dispatch/transport to a laboratory, chain of custody procedures, qualifications and training required for on-site specimen collectors and laboratory personal,  and quality control management of onsite and laboratory test devices/methods.

The effectiveness of drug testing

There is sufficient research evidence to indicate that urinalysis, oral fluid testing, and hair follicle testing are reliable methods for indicating past drug use.  However, none of these tests, whether conducted on-site or in the laboratory, can detect impairment or intoxication.  While Australian Standards for urinalysis and oral fluid testing outline cut off levels for detecting past drug use, these levels are set to minimise the likelihood of false positives rather than levels of impairment.  In addition, these drug tests cannot reliably indicate the time of use, the amount used, dose level, or the pattern of use. 

Evidence concerning the effectiveness of drug testing in deterring use or reducing risk of harm is limited.  Few quality evaluations of drug testing programs have been undertaken, and what evidence is available indicates that drug testing has little, if any, effect.  For example, reviews of research that have evaluated the effectiveness of workplace drug testing (e.g., Kraus 2001, Macdonald et al 2010, Frone 2013, Pidd & Roche, 2014), consistently conclude that the evidence base is weak.   Similarly, cost effectiveness studies of workplace (Halperin et al., 2008; Mehay and Webb, 2007) and welfare testing (Bloom, 2010; Klein, 2017) indicate testing programs have little economic value.

Why is testing so popular?

The lack of a sound evidence base for the effectiveness of testing is at odds with the apparent growth in the use of testing.  In the workplace, the growth in testing is likely due to genuine concerns over workplace safety and a misplaced belief that drug testing provides an objective and reliable marker of safety risk due to drug use.  Compliance with industrial (e.g., Schedule 4 of the Australian Building Industry Code and Civil Aviation Safety Regulations) or public (e.g., State Government Road Traffic Acts and the proposed testing of welfare recipients) legislation also plays a role.  However, in light of the poor evidence base for its effectiveness, there may be other explanations for the growth in testing. One alternative explanation may be successful marketing efforts by the drug testing industry. It has been estimated in the US and Canada that workplace testing is at least a $2 billion a year industry and increasing (Grantham, 2012). Previous research has also identified that workplace drug testing can perform a symbolic function, with the introduction of drug testing in reaction to normative pressure from the media and public opinion concerning the ‘drug problem’ (Knudsen et al., 2003).  As speculated elsewhere (Galea, 2013), health policy such as drug testing is based on a seemingly logical and compelling idea (i.e., testing can reduce drug related harm by deterring drug use or identifying those that use drugs).  Driven by wider social concerns regarding drug use, this seemingly logical and compelling idea is implemented and supported regardless of the evidence base for its effectiveness.

Good practice in drug testing

Any potential beneficial effect of drug testing is likely to be realised when it is combined with a more comprehensive response that provides better access to treatment, de-stigmatisation and improved support.  More comprehensive responses also need to be tailored to meet the specific needs and circumstances of the target group.  In addition, responses need to include strategies to minimise the risk of unexpected negative outcomes known to be associated with testing.  One unexpected negative outcome is that, rather than changing their behaviour to reduce drug use or related risk of harm, the target group may simply change their behaviour to avoid detection.  When this occurs, drug testing programs are likely to have counter-productive consequences.


Regardless of the poor evidence base for its effectiveness, drug testing is a growth industry that has implications for the AOD sector.  Most drug testing programs require people who test positive to access some form of assessment, counselling, and/or treatment.  This will mean increasing demand for AOD services.  A large proportion of this demand is likely to come from potential clients who are either ‘social’ users who are not experiencing serious drug related harm, and potential clients who are being coerced into counselling or treatment.  The AOD sector needs to be fully informed on how drug testing programs operate and consider the impact these programs may have on service delivery. 


Bloom, R (2012) Just as We Suspected: Florida Saved Nothing by Drug Testing Welfare Applicants. ACLU Blog of Rights 18 April 2012, https://www.aclu.org/blog/mass-incarceration/just-we-suspected-florida-saved-nothing-drug-testing-welfare-applicants accessed 4th October 2017

Dupont, R. L. Griffin, D. W. Siskin, B. R. Shiraki, S. Katze, E. (1995). Random Drug Tests at Work - The probability of identifying frequent and infrequent users of illicit drugs. Journal of Addictive Diseases, 14(3), 1-17.

Frone, M.R., 2013. Workplace interventions I: drug testing job applicants and employees. In: Alcohol and Illicit Drug Use in the Workforce and Workplace .American Psychological Association, Washington, DC, US, pp. 143–175.

Galea, S., 2013. Values, compelling ideas, the pace of science, and the implementation of evidence-based policy. Addiction 108 (5), 847–848.

Grantham, D., 2012. Third-party drug-test providers continue to grow nationwide. Behavioral Healthcare. 32 (5), 38–40.

Halperin, E. C., Andolsek, K. M., Jackson, G. W., & Weinerth, J. (2008). Pre-placement screening of resident physicians by substance abuse testing: Efficacy, cost, and physician opinions. Drugs: Education, Prevention & Policy, 15(1), 77-91.

Klein, A. (2017). Australia plans random drug tests for people receiving welfare, New Scientist, August 2017. https://www.newscientist.com/article/2145603-australia-plans-random-drug-tests-for-people-receiving-welfare/ accessed 3rd October 2017.

Knudsen, H.K., Roman, P.M., Johnson, J.A., 2003. Organizational compatibility and workplace drug testing: modeling the adoption of innovative social control practices. Sociol. Forum 18 (4), 621–640.

Kraus, J.F., 2001. The effects of certain drug-testing programs on injury reduction in the workplace: an evidence-based review. International Journal of Occupational and Environmental Health, 7 (2), 103–108.

Macdonald, S., Hall, W., Roman, P., Stockwell, T., Coghlan, M., Nesvaag, S., 2010. Testing for cannabis in the work-place: a review of the evidence. Addiction 105 (3), 408–416.

Mehay, S., & Webb, N. J. (2007). Workplace drug prevention programs: does zero tolerance work? Applied Economics, 39(21), 2743-2751.

Pidd, K. & Roche A.M. (2014). How effective is drug testing as a workplace safety strategy? A systematic review of the evidence. Accident Analysis and Prevention, 71, 154-165.