From the 12th to the 14th of June this year, we’ll be hosting the Scientific Meeting of the Society of Hair Testing (SoHT) in Cardiff. As a preview, we’ve brought together the keynote speakers from the conference for a series of exclusive interviews. This month, we’re pleased to welcome Michelle Taylor.
Michelle is a senior research associate in epidemiology MRC Integrative Epidemiology Unit at the University of Bristol, with a background in forensic science and genetic epidemiology. Her research focuses on the causes and consequences of drug use, addictive behaviours and mental health.
Cansford Labs (CL): Michelle, tell us a little about your background and your interest in the Society of Hair Testing.
Michelle Taylor (MT): First, I wouldn’t say I’m a hair researcher as such. I’m an addiction researcher who has tried to find out if this method is something we can use in the kind of research we do.
My interest in hair drug testing comes from wanting to find a reliable biomarker of light, heavy and long-term drug use - something that can be used in general population samples. If we’re able to find a biomarker like that, we could better determine why people use drugs and what the consequences are: we could tell how much people have used drugs, how long they’ve used them and how often.
One of the problems we have is relying on self-reported data. If we want to know the age that someone started using a drug, or how much they’ve used it, we often find that people are embarrassed and under-report what they’re using - or they want to big themselves up in some way and report more. There are ways to get around that, but it would be great if we could take a bit of their hair, test that, and get a really good marker of what they’ve used. I’m more interested in using hair testing in the research sense, to improve data collection.
CL: Having read some of your papers, in the past you’ve seemed unsure about the efficacy of hair testing.
MT: I want to be quite careful here… I think there are potentially some situations where it will be really useful - in forensics, for example, in court, where you might have a really heavy user. The thing for me is that those aren’t the kind of samples I do my research on. I work with a broad spectrum of drug use, from people who maybe smoke one joint a month to those who are having five a day. I need to try and encompass all of that, and have a really reliable measure of it, which is where hair testing methodology currently falls down for me.
That’s what my research has shown. I don’t want to say I’m critical of the whole method!
CL: We’re with you now. Because you’re trying to find links between mental health and drug use, you have to monitor lower levels of the drug than hair testing allows?
MT: Exactly. We’re not only interested in whether, say, really heavy users of cannabis go on to have this health outcome or that. We also want to answer questions like “does very occasional use of cannabis have an impact on your later health?” Hair testing falls down with those sorts of samples. We can reliably pick up heavy users of cannabis, but when we take it to the lighter users, we very regularly get negative tests, and we can’t be sure if those negative results actually reflect abstinence or if the individual’s level of use is just so low it can’t be detected.
CL: Where does your interest in addiction come from?
MT: I generally find it fascinating. I think it’s the decision-making process. A lot of epidemiologists are interested in cancer or cardiovascular disease, but there isn’t a decision-making process in those. What makes people actually go out and use the drug? That’s the extra dimension that I find really interesting about addiction.
CL: And your talk at the Scientific Meeting - that’s all around this research into cannabis use?
MT: I’m going to talk about two studies that link together quite well. The first was a small pilot study - we recruited 136 subjects who were self-reported heavy, light and non-users of cannabis, and we looked to see if we could use hair as a reliable marker of cannabis exposure, and if there was a correlation between the amount of cannabinoids in the hair and the self-reports.
We found that only 40% of the light users had THC detected in their hair, compared to nearly 80% of the heavy users - the people using at least five days a week. This study also showed that 90% of the positive values for the positive hair tests were correctly identified, but only 60% of the negative hair tests were correct. You can see that’s where the uncertainty comes in; when you have a negative hair test, what does it actually mean?
This fed into the second study, which used participants from the Avon Longitudinal Study of Parents and Children. The children from that study are now around 25, and we’ve had so much data from them. When they were 18, we collected hair samples from 3,500 of them: it’s a much larger sample size than any of the previous work on hair testing has looked at. Of the people who reported using cannabis during the three months before having their hair tested, we only detected cannabis in 20%. When we compared the hair testing to the self-reporting data, we found more false positives than false negatives.
CL: What are the ramifications of that?
MT: Well, for us in epidemiology it means that we can't use current hair testing methods to generate a biomarker of cannabis use in general population samples. It’s way too unreliable - it doesn’t have the authenticity that we need. We can get more information from self-reports, even if they may be biased.
In terms of implications for the wider world… I don’t want to comment on the legal implications, because the nature of my research is population and public health, and I don’t think I’ve the expertise to give an informed answer on court cases and such.
CL: Have you been to the Scientific Meetings before?
MT: This is my first time, but I think it will be good. I’m looking forward to something that I’m also very wary of: there’ll be a lot of scientists from different disciplines, particularly chemists, and I’ll have to portray epidemiology to all of them. In my unit we work with people from economics and neuroscience, but I don’t think we’ve worked with anyone from chemistry yet, so maybe I’ll be the first on that.
Interdisciplinary research is a challenge, but it’s also really exciting, and it’s becoming more and more necessary to accelerate to new levels. We’re starting to find that a lot of funding bodies want to see cross-discipline collaboration, too. It’ll be exciting to learn how other people are progressing, how we can learn from each other. You never know, I might spark a new collaboration that can help us progress hair testing in general populations.
We’d like to thank Michelle for answering our questions - and like her, we’re looking forward to the Scientific Meeting in June. If you’re as intrigued as we are, download a registration form for the event here.