This systematically reviewed the evidence (as of Jan 2013) concerning the effectiveness of workplace drug testing (WDT) as a workplace safety strategy.
Authors were interested in answering two questions:
1. Does WDT reduce occupational accident injury rates?
2. Does WDT deter employee drug use?
Of the 285 studies identified, only 23 met inclusion requirements: 6 reported on the effectiveness of WDT for deterrence & 17 reported WDT on injury rates. All data was from the US.
Results:
Overall, although most studies indicated a benefit of WDT the majority of studies contained methodological weaknesses – with only one study assessed as having a strong methodology.
The authors note that despite the growth of workplace drug testing & generally insufficient quality research: “the evidence base for the effectiveness of testing in deterring employee drug use or improving workplace safety appears at best tenuous, and at worst, seriously flawed” (p161-162).
No randomized controlled trials were found. Further, lack of randomization, low response rates, small sample sizes, failure to control for confounding variables and other issues were endemic through the research. One study which found random drug testing could result in heavy truck drivers being 14.5% less likely to be involved in an alcohol related fatal crash only reported the relative risk reduction (as above), rather than the absolute – which was found to be a total reduction of only 1.7% in absolute terms.
All six of the deterrence studies (assessed as methodologically weak) indicated that WDT was associated with lower levels of employee drug use. However, “methodological issues limit definitive conclusions that can be drawn from these findings” (p162).
5 studies assessed as having moderate methodological quality found that random alcohol & drug testing & post-accident testing programs was associated with reductions in injury rates. Further, one study
For the one study with a strong methodology found that random alcohol testing reduced alcohol-involved road fatalities among heavy truck drivers. Some other studies with moderate methodologies found similar reductions in transport accidents.
Reducing deterrence, authors suggest that, perhaps, WDT may motivate employees to use drugs that are perceived to be less readily detectable than other drugs. However, if this is even an effect, it’s likely to be small. Some evidence supported this suggestion, suggesting that it’s feasible that WDT may not deter users from drugs so much as deter current and past drug users from applying for jobs where testing is done.
Further, authors suggest that lack of evidence supporting may be because WDT is based on 3 core assumptions:
1) testing can identify most employees who use drugs,
2) a positive test identifies employees who are likely to be impaired, &
3) a causal relationship exists b.t. drug use and safety.
However, evidence suggests that:
1) WDT is only likely to identify *frequent* users and most users of drugs do so infrequently,
2) positive drug tests merely indicate past drug use, not impairment, and
3), “the relationship between drug use, cognitive and psychomotor performance, and safety outcomes is not strong and more complex than typically acknowledged” (p163).
Author: Ken Pidd, Ann M.Roche, 2014, Accident Analysis & Prevention
SafetyInsights.org 