We are scientists

In the words of the great Gil Grissom: 'I tend not to believe people. People lie. The evidence never lies.' The fear that the American TV crime series CSI may have turned us all into keen but amateur forensic scientists has spawned serious academic research into the effects of such programmes on jurors in the US. At least some of that research suggests that CSI fans are likely to approach forensic evidence in a more questioning way. Committed fans of Grissom will recognise the importance of another thought of his: 'I'm wrong all the time. It's how I get to 'right'.'

I recently had the opportunity to look at the interface between forensic test results and the family justice system when I was instructed by Trimega Labs as intervenor in a case before Moylan J reported as LB Richmond v B [2010] EWCA 2903 Fam. Hair tests for alcohol are relatively new to the legal arena in comparison to fingerprints, DNA, blood tests, etc. but they have a respectable ten-year history and a sound science base.

The alcohol hair testing industry is not regulated but self-regulated. It governs itself through the Society of Hair Testing whose board is made up of an international group of experts. All the main testers in the UK are members of the society. In 2009, the society met and agreed that certain cut-off levels should be adopted to show whether or not a person has been drinking to excessive levels. The level where a person's drinking is deemed to be excessive is taken from World Health Organisation research and literature as 60gm of alcohol a day (about 7.5 units, nearly a bottle of wine or 2.5 pints of strong lager).

Hair comes the science bit

Blood or urine tests can directly show the presence of ethanol but only for a period of hours after its ingestion. Blood can provide information covering a longer time frame than urine '“ four to six weeks '“ when liver damage if present may be detected, but not everyone who drinks will have liver damage and not everyone with liver damage drinks.

Hair tests for alcohol do not directly test for ethanol but for two types of alcohol marker: Ethyl Glucoronide (EtG) and Fatty Acid Ethyl Esters (FAEEs). A number of labs in the UK test for EtG '“ only Trimega Labs currently have accreditation to test for FAEEs.

EtG is water soluble, incorporated into the hair through sweat. Over time it may be washed out of hair so it has a relatively limited shelf life. FAEEs are fat loving and are much less sensitive to hair treatments and washing. Both types of marker may remain in the hair for a period of many months '“ hair growing at approximately 1cm a month.

Hair test results are a more reliable indicator that ethanol is present in the body than blood tests for any time frame above a couple of days.

The SOHT recommends that hair samples of 3cm are taken and that in respect of FAEEs the cut off recommended for a 3cm sample is 0.5 nanograms per milligram (ng/mg).

For EtG cut-off levels of 30 picograms per milligram (pg/mg) are applied. One nanogram is 1000 picograms, so for EtG the cut off in nanograms is 0.03ng/mg (and for FAEEs 500pg is the equivalent of 0.5ng).

Testing times

In the case, test results for EtG from another lab were reported as showing alcohol use by a mother who was claiming to be abstinent and in recovery. Not only that but it was reported that her drinking in a given one-month period was at a specific level (between 20 and 50 units a week).

The mother's denial of alcohol use in the face of this caused the usual undermining of her relationship with the professional network (addicts and former addicts are understandably always treated with a high degree of scepticism at the best of times) and caused the addiction specialist to doubt her ability to maintain abstinence.

The subsequent FAEE test done by Trimega Labs covering the same period was reported as being negative and showing abstention or virtual abstention (albeit it was not a zero reading but well below cut-off levels).

In fairness, the standard literature which accompanied their original test results did suggest that the FAEE markers would only be detected when a subject had consumed alcohol. That literature was quickly revised to make it plain that 'zero results may not be returned even by teetotallers because ethanol is present in all hair'¦ these traces are the product of the environment, the metabolism of certain foods and cosmetics'.

When the labs were originally instructed to carry out the tests they were simply asked to do so by using the standard forms supplied by both of them. Further questions were then put to the labs to seek clarification and in due course two experts met to discuss the issues and deal with a number of questions put by the lawyers.

Further written reports were commissioned and eventually full oral evidence was heard by Moylan J. At paragraph 22 he summarises the upshot of that evidence:

• Hair tests should only be used as part of the evidential picture '“ the higher the level above cut off the more significant they are likely to be. However, they should never be viewed in isolation.
• Hair tests should always be commissioned for both FAEE and EtG.
• Hair tests should not generally be claimed to demonstrate anything more than consistent with excessive alcohol use or not consistent with alcohol use. Below the cut off for excessive alcohol use they are consistent with abstinence or social drinking and cannot be used to prove abstinence or drinking.
• Cutting the hair into 1cm segments does not yield reliable results (you cannot just take the cut off for 3cm and divide by three because the markers are not evenly distributed in the hair and there is not yet a society approved cut off for 1cm segments).

In other words, the absence of evidence is not evidence of abstinence and the evidence of abstinence may not be what it seems!

For the moment, the tests are good to show that someone has been drinking an average of 60gm a day in a three-month period (it makes no difference whether the subject drinks daily or in binges).

There are already respectable opinions within the industry that lower levels than the society cut off may not be consistent with abstinence. Until these figures and the accompanying research have been subject to greater scrutiny they are probably not yet good enough on a standalone basis for a court of law '“ but this is not to say that they lack scientific validity.