Predictors of addiction: Why being able to hold your booze might not be a good thing
This weekend I’ve been thinking about addiction. Admittedly, I think a lot about addiction! I am in the process of making a series of podcasts about recreational drugs, and while at a neuropsychopharmacology conference last month, I interviewed Anne Lingford-Hughes about research in to alcohol. Some of her comments really got me thinking about the nature of addiction, and why some of us can have a few drinks while others find the need for alcohol can consume them.
Alcohol has an unusual position in the UK, as it’s the only completely socially acceptable drug. Indeed, alcohol is so accepted in society that there is pressure to use it on a regular basis, and potential for social ostracism if you don’t. Alcohol is almost synonymous with socialising, and this sets it apart, even from tobacco. It’s also a potentially very harmful drug, indeed according to David Nutt’s oft-cited Lancet paper discussing relative harms, if you consider harm to self and society combined, alcohol is the most harmful drug used in the UK. Of course, this is tied in with its ubiquity, but even so there is a problem.
It’s not the case that everyone who uses alcohol will abuse it, and despite binge drinking culture among young people, excessive alcohol consumption even for many months or years will not lead to addiction for all. Marc Schuckit has done some really interesting work on what predicts later alcohol addiction, including using the dataset I work with. His extensive report on alcohol use disorders is a great place for an in-depth discussion of alcohol, but I’m particularly interested in some work he did looking at people he refers to as ‘low responders’.
Alcohol is metabolised by proteins, which are coded for by genes. There is a lot of variation in these genes, which means some people make less of these proteins than others, so don’t metabolise alcohol or its byproducts as quickly. The first step in alcohol breakdown creates acetaldehyde, which is toxic in large quantities (it is believed this may be why we get hangovers). This is then broken down in to acetate which can be used in the body. If a person has less of the protein needed to break down either alcohol or acetaldehyde, drinking can be unpleasant and is often avoided. This makes these genes useful in Mendelian Randomisation studies as I’ve written about before (the paper last week about alcohol during pregnancy and child IQ is a great example of Mendelian Randomisation, and why the authors can be more confident of a casual association than is usual in observational epidemiology).
Schuckit is interested in those at the other end of the spectrum. Teenagers who, from the first time they drink, find they don’t get intoxicated easily, and therefore need more alcohol to get the same effect from it as their peers. These teenagers are more likely to have parents with problematic alcohol use, and are more likely themselves to end up with alcohol problems. There’s evidence that some, but not all, of the variability in response is explained by the genetic differences mentioned above. Also, alcohol and other substance use by friends, the use of alcohol as a coping mechanism, and positive expectancy of alcohol have all been linked to low response.
Level of alcohol response is measured by asking people to rate how many drinks they would need to reach different levels of intoxication, so it’s pretty straightforward to work out. After an explanation of how much of their chosen drink is equivalent to one unit, people are asked how much is needed before they:
- Feel any effect
- Start slurring speech
- Start stumbling
- Experience ‘unwanted falling asleep’
The questionnaire has been used in adults and teenagers, and is pretty good at predicting later alcohol problems.
Schuckit has a great idea about how this could have a therapeutic use. He has suggested that targeted interventions could be created, dependent on response to alcohol. He’s already run a pilot study where he took a group of freshmen at university in California, and gave 32 the currently used videos designed to advise against heavy drinking. The other 32 participants watched a video which showed evidence that those with low initial response to alcohol are more at risk of heavy drinking, and that this is also affected by the heavy drinking of their peers. The groups saw 4 videos across 4 sessions, and their drinking behaviour was assessed at the beginning of the experiment, throughout it, and a month after the final video session. He found that both interventions reduced alcohol use in the students. However, when he analysed by individual alcohol response, he found that the original videos were better at reducing alcohol use in high responders, but his new videos were better in low responders.
Although only a pilot study with a small sample size, it’s a great indicator that targeting interventions by alcohol response might be a great way to help people at higher risk. Those who have to drink a lot to feel an effect can be informed that they’re at increased risk of later problems, which could make them more likely to moderate their drinking. He ends his paper with a call for more research in this area, as it's a potential method to help people avoid alcohol addiction and problems before they begin; I hope it's happening.