Drugs in the workplace 'widespread'

A “million workers have Class A drugs in their system”, The Guardian has wrongly reported. The Guardian is not alone in making a hash of its reporting of an apparent 43% rise in the use of drugs at work over the past five years.

These findings come from the analysis of more than 1.6 million workplace drug tests in the UK from 2007 to 2011. The analysis was carried out by a drug testing company, which has produced a report of its findings. There is a clear conflict of interest because the company has a commercial interest in promoting the use of drug testing.

The findings from 2011 show that around 3.23% of tests on workers found positive results for either illicit drugs, or prescription or over-the-counter drugs that employees had failed to report as part of the testing programme. The report also showed an apparent increase in the use of drugs among workers since 2007, as well as showing some evidence about the different drugs people use at different ages, and which drugs are favoured by men and women. However, the reports in most of the media lack any independent scrutiny of these statistics.

Overall, this report serves to stimulate discussion on several issues, such as:

whether the reported levels of positive drug tests are acceptable
to what extent they could potentially lead to any injury or harm
what can be done about reducing the levels of drug use for employers who consider them to be too high

The bottom line is that this research has important drawbacks and a clear conflict of interest. The results should not be accepted at face value until further research has confirmed or contested them.

Who carried out this drug testing research?

The research was carried out by Concateno, which describes itself as a “leading European drug and alcohol testing provider”. There is a clear conflict of interest because the company has a corporate interest in promoting the use of drug testing in the workplace.

What did the drug testing research involve?

The research analysed the results of more than 1,668,330 drug tests conducted by Concateno’s laboratories on behalf of 856 UK employers over a five-year period (2007 to 2011) to measure the prevalence of drug use among employees.

The drug tests that were analysed were conducted as part of random or pre-employment programmes, rather than following an incident at work or where there was a suspicion of drug use. All tests were conducted in organisations that had an established drug testing policy in place. This means that employees knew they could be tested as part of their employment.

The report offers very little detail on the people who were tested. Type of job, seniority in a company, where the participants lived, ethnicity and other important factors, would typically give a better picture of the results. With such a lack of detail on the people who were tested, we cannot be sure whether the results from this group are representative of the UK workforce in general. The effect of the drugs on workers’ performance, safety, sickness or absence was not analysed, so claims about the problems drug use has caused in the workplace can’t be verified.

Test results were from urine and “oral fluid” analysis, but most of the workplace drug testing used urine samples. The research stated that if a person tested positive for a drug, they would have “most likely” consumed the drug in “recent days”. Hence, a positive test did not necessarily mean that employees had taken drugs during working hours. If they had taken drugs at the weekend, for instance, traces of the drug may still have been present on a Monday – and this could have led to a positive test result.

At the time the samples were collected, employees were asked to provide information about what prescribed and over-the-counter medicines they were taking. The positive drug tests in the research relate to those that were not accounted for by medication declared at the time of the sample. In some cases, positive tests could be prescription or over-the-counter drugs that were not declared at the time of the sample, or they could be illicit drug use.

What did the drug testing research find?

The brief report of the findings did not appear in a peer-reviewed journal. The peer review process is designed to ensure that the methods and results of a study are scrutinised by other experts in the field to test their rigour and reliability. As this did not happen, the results below remain unconfirmed and unchallenged. Discussion about the strengths or weaknesses of test results was noticeably absent from the report.

The report highlights the following results:

During 2011, 3.23% of all those analysed tested positive for drugs from samples taken in the workplace. It is not clear how many people were tested in 2011 because only the total number of tests for 2007 to 2011 (1,668,330) was reported.
The report’s authors estimated, by extrapolating their figure to the 29.23 million people employed in the UK, that 940,000 people would have drugs in their system “right now”. This equated to approximately one in 30 UK employees having drugs in their system at any point in time.
The 2011 positive test rate represented a 43% increase in UK employees testing positive for drug use from 2007 to 2011 (from 2.26% to 3.23%). However, it was not clear whether this rise compared like with like because no information was provided about who and how many people were tested in the 2007 and 2011 groups.
The most prevalent drugs used by UK employees were cannabis (1.93%), over-the-counter opiates, such as codeine (1.87%), and cocaine (0.59%).
Benzodiazepines (0.2%), amphetamines (0.11%), methadone (0.06%) and heroin (0.02%) were also detected.
Under 25s were most likely to test positive for drugs and, in this age group, cannabis was the most commonly detected drug.
Cannabis use decreased markedly as people got older. Over-the-counter opiate use increased with age.
The age group most likely to test positive for class A drugs was 25-34-year-olds. The figures for drug use by age group excluded tests from the transport industry.

The survey also includes a case study from a British manufacturing company reportedly showing how implementing drug testing in the workplace reduced their positive drug test results from “exceeding 25%” to 6% in 12 months. However, because the report provides no further details, it is not possible to conclude that implementing drug testing itself contributed to this reduction in positive drug test results. To examine whether drug testing did cause a drop in the number of workers using drugs would require a randomised controlled trial.

Why do so many people test positive for drugs at work?

The report did not address the key issue of why so many workers tested positive for drugs. Importantly, it also did not report the diagnostic accuracy of its tests. For example, we have no idea how many of the positive test results were falsely positive.

The report noted that the type of drug used changed with age and this was likely to be due to changes in disposable income. It suggested class A drug use (such as cocaine use) peaks a few years after starting working life rather than earlier because these drugs are more expensive and only become affordable later in life.

Similarly, opiate use also increased with age. The report suggested this may be due to increased periodic use of over-the-counter painkillers or may reflect long-term dependency on opiate-based drugs.

How have the media reported these findings?

Many media outlets picked up on the main finding of the Concateno report, that “nearly a million employees in the UK ‘have drugs in their system at work’”, as the Daily Mail reported. However, there are numerous errors in the universally uncritical reporting of these findings. For example:

Potentially flawed extrapolation: all media sources have uncritically stated the “million” figure without reporting that it is extrapolated from a rate of 3.23% among a largely unknown sample of workers, with no reference to how representative the sample was of the wider workforce.
Incorrectly reporting all of the drugs found to be class A: The Guardian’s reporters have apparently failed to read the research. The Guardian’s report claims that a “million workers have class A drugs in their system”. In fact, a vast proportion of the drugs examined are likely to be prescription opiates such as codeine, or prescription benzodiazepines – in some cases workers may have inadvertently forgotten they were taking them, but other cases may suggest an addiction. Further, cannabis, the drug most commonly found by testing, is currently a class B drug.
Over-inflated figures: the Daily Star claims that the number of people found to have drugs in their system “had shot up by nearly 50% since 2007”. However, the true figure is only 43% and this is a relative figure. The absolute increase in positive tests was only 0.97% over five years (from 2.26% in 2007 to 3.23% in 2011). The absolute difference can often give readers a much better sense of the size of any increase in real terms. However, this is less headline grabbing so it was not a figure reported in any of the news headlines or reports.
*Lack of differing opinions: *almost all the papers included quotes from the director of the drug testing company, who claimed that the findings were “conservative” estimates. Only the Metro has offered any analysis from anyone not connected with the company involved.
Unsubstantiated claims: the Daily Star reports that “Narcotics experts found one in 30 workers tested positive for cannabis, cocaine or heroin while on duty, with others popping banned pills in the workplace”. The report states that “where an employee tests positive for a drug, they will have most likely consumed the drug in recent days” – so it cannot tell whether the workers tested used drugs at work or during their own time.

What is being done about drug use in the workplace?

Concateno’s report includes advice for employers, in which it suggests that 171 workers were killed at work in 2010/11. In the same year there were 200,000 reportable injuries and 26.4 million working days lost due to workplace injuries or work-related illness, it says, while the cost to the UK was estimated to be £14 billion for 2009/10. However, these figures were for all injuries. In no way do they suggest or prove that drug use was a contributing factor. The report goes on to highlight a report that suggests cannabis use can increase the risk of “collision resulting in serious injury or death”.

The report’s advice also highlighted the Health and Safety at Work Act 1974, which emphasises employers’ duty of care to provide a safe work environment, and an obligation to ensure that employees are not misusing substances that can impair their ability to carry out tasks safely.

The final word on the subject from Concateno’s report was that “drug testing programmes have been proven to reduce the level of substance misuse in the workplace”. This statement was not backed up by the evidence cited.