Preview Mode Links will not work in preview mode

Jul 3, 2019

In Episode 6 of the DIAL podcast, Christian Zünd from the University of Zurich discusses his research looking at the interplay between our genes and what we drink, local availability of alcohol and the role of licensing laws. The research is part of the NORFACE-funded project, Gene-Environment Interplay in the Generation of Health and Education Inequalities (GEIGHEI), which is looking at how Genes and the Environment (GxE) interact to generate inequalities in education and health over the life course.

Further information:

Christian Zünd was discussing research presented at the DIAL Mid-Term Conference in June 2019.


In Episode 6 of the DIAL podcast, Christian Zünd from the University of Zurich discusses his research looking at the interplay between our genes and what we drink, local availability of alcohol and the role of licensing laws. The research is part of the NORFACE-funded project, Gene-Environment Interplay in the Generation of Health and Education Inequalities (GEIGHEI), which is looking at how Genes and the Environment (GxE) interact to generate inequalities in education and health over the life course.

Christine Garrington  0:00  

Welcome to DIAL a podcast where we tune into evidence on inequality over the lifecourse. In today's episode we're discussing genes, pubs and drinks and asking what our genes can tell us about our alcohol consumption? How they interact with our environment? And what all that might mean for public health policy? Our guest is Christian Zünd from the University of Zurich, he's part of the DIAL funded project Gene-Environment Interplay in the Generation of Health and Education Inequalities.

Christian Zünd  0:28  

Basically there's about 60 different medical conditions that have been strongly associated with alcohol consumption, there's a lot of sort of epidemiological and medical literature that looks at these current links. There's even been a paper that tries to sort of quantify the impact of alcohol consumption, sort of, compared to say smoking or obesity and these kind of things added to inside that it's actually on a global scale sort of comparable to the damage caused by smoking, or by hypertension and these kind of thing so it's quite a major health issue.

Christine Garrington  1:00  

So when it comes to any links that might exist between our genes, which is what we're here to talk about today, and the alcohol we drink, what was it that you wanted to look at specifically and why?

Christian Zünd  1:12  

Part of it is that we're really interested in seeing how different components sort of come together to determine how much you drink, and so part of that is your genetic makeup. So there are certain factors that influence how likely you are to drink, basically by influencing the way your body deals with different kinds of sugars and so that's at least part of the story. And the other part seems to be just a residue of all sorts of behavioural factors that might make you more likely to go to frequent pubs and these kind of things so it's not entirely clear what's going on there but we have some idea. Part of it is a metabolism, breaking down different sugars kind of story and the other policy is not that well understood at the moment, then the other factor is just your local environment so we all know that if you are a student you tend to drink more. So there's this kind of age, lifecycle aspects to it. And there's also the factor, which is basically just how easy it is for you to go to a place and have a beer there. And so what we decided to study is basically looking at how your genetic makeup interacts with your local environment, and we'll look specifically at how many hops are there within one kilometre of the place of your residence, and we also look a bit at supermarkets and these kind of things

Christine Garrington  2:25  

And where did you get your information from?

Christian Zünd  2:28  

Where? Yeah so it's becoming increasingly easy to do this kind of genetic research. And one big factor is the UK has this really amazing data source called the UK Biobank, and it's about half a million individuals, and we do have genetic information for them and a lot of other health related behaviours. So there's a number of questions that deal with how frequently do you drink alcohol? Very specific questions about how much beer do you drink? How much wine you drink? Sparkling wine with these kind of things, so there's quite quite a lot of rich information on health behaviours and we also know where these people live so basically what we were able to do is sort of match this information from the UK Biobank with a large cross section of where all the pubs in the UK are located, and we've we've had over, 10 or so largest supermarket chains, the UK and so we have a pretty good idea of how easy it is for these people to buy alcohol somewhere.

Christine Garrington  3:20 

So, how did you make use of the genetic information that was available then. So what have you done with it, basically, um, what can it tell you what does it tell you about people's drinking habits?

Christian Zünd  3:30  

The approach we use is called a polygenic risk score, and the idea is that instead of trying to identify the one gene that causes you to drink a lot if you just take sort of like bird's eye view of the entire genome and you look at all the different associations and even though each one of them contributes very little to your drinking behaviour, combined, if you look at all of them you can actually tell quite a lot about how much people drink. And so it's this useful approach for all sorts of complex traits where it's not entirely clear which genes are involved in it or whether it is a large combination of them, people have been using it to look at the educational attainment, people have been using it for height and BMI, these kind of things. We use this for alcoholic consumption,

Christine Garrington  4:16  

And so what about how you are able, if you were able to establish the availability of alcohol in an individual's local area? For example, and also how did you sort of find out whether there were any local efforts, if you like - any policies in place to try to restrict the amount of alcohol, the people were consuming? as I mentioned we

Christian Zünd  4:34  

As I mentioned we have information basic data from the internet on the locations of pretty much every single pub in UK, and we do have their coordinates from Google Maps and that allows us to then match them fairly precisely with the place with someone on our lists, and initially what we wanted to do is we wanted to look at like, what's the shortest distance to a pub that you have so how far will you have to walk to get to a pub? But because that measure is only precise to some degree. It turned out that actually a lot of the people in our sample live very close to pub so we didn't really have a lot of variation there so instead we decided to look at how many pubs, do you have within one kilometre? and there we find quite a bit of variation so if you're living in a more rural place, it's actually possibility do not have a single pub within one kilometre, although that tends to be very rare in our data. And if you live in a place like Blackpool, you probably going to have two or 300 pubs or restaurants and these kind of things within one kilometre of your residence, so there's a lot of variation there.

Christine Garrington  5:39  

Okay, so let's start getting into the nitty gritty of what you found so far, appreciate it is probably ongoing, but you have found some things already, just talk us through some of the main things that you found around that.

Christian Zünd  5:52

The first one is not entirely surprising, which is that we're actually able to predict quite a bit of how much people drink. So the people who sort of score lowest on the polygenic risk score they drink about half as much on average as the people who score highest so there's a fairly strong relationship there, the more interesting finding is that if you have a high genetic propensity to drink a lot, you react much less to sort of changes in the environment. So, it seems that if you're not very prone to drink, and you would have to walk quite a bit to get a pub, or there's not that many options, you might just drink whereas if you have a high genetic propensity walking a bit doesn't really seem to deter you. So there's sort of this difference in how much you react to these availabilities in the environment. Then there's a second finding, which is quite interesting, which seems to be that people who have a high genetic risk to drink tend to self-select into places where there's more opportunity to go drinking. And so we tried a number of different approaches and they're basically supplying the same result. So one is just to look at the cross section, then see whether we can predict the number of pubs, based on your polygenic score. And that seems to be the case. The other one looks at people that we can observe at several different points in time. And there it turns out that if you have a high genetic propensity to drink, but you live in a place with few pubs you are more likely to move to a new place, same is true if you have a very low polygenic risk to drink, and you live in a place with a lot of pubs, also more likely to move to new place. And the third thing is that if you compare where someone grew up, and how many pubs there were around that place to where they live now. You do find the people with a high polygenic score, tend to have to move to places have more pubs and people with lower score tend to move to places with fewer pubs. So there seems to be quite a lot of sort of self-selection in the environment there. 

Christine Garrington  7:55  

But what about only sort of the policy from then? Because efforts are made in certain areas, whether it's to stop more drinking outlets or more pubs opening up, or restricting the amount of hours for example that pubs and places where you can buy alcohol, stay open. Where's the link there then with sort of local policies that are in place? What do we learn?

Christian Zünd  8:20  

Alcohol licencing policy tends to be fairly decentralised for UK standards. And so there's quite a bit of variation in how likely these places are to try to restrict drinking. But the interesting thing from our perspective is that they're actually not allowed to make public health considerations in their decision making but only to look at things like public noise and disturbances, the impact that it might have on children, and these kind of things but they're not supposed to try to use this as a public health tool and so in a way we have this variation that's introduced for other reasons that we can try to see how that affects health behaviours, so far is mostly concentrated on the tool called a no sales licence, so there's a lot of different establishments that could apply for licences to sell alcohol so there's not only pubs and restaurants but there's also basically every kind of entertainment area that's open late so that includes sports facilities, it includes theatres and cinemas and batting halls and all these kind of establishments. Some places, tend to hand out no sales licences so even though they're licenced to stay open and provide entertainment and these kind of things they're not allowed to sell alcohol. There's a lot of variation in how likely these local licencing boards are to restrict sales and all of these other places. And what we find there is quite interesting because it's basically just saying that we found, for the different reaction to how easy it is to walk to pubs. So we also find that the people with the highest polygenic risk to drink react to least to these changes, whereas the people with the low polygenic risk, they tend to be the people that actually start to drink less if it's harder for them to go drinking, and then we looked at a number of other tools such as licences that allow you to sell alcohol 24/7, and we would expect to have an effect that goes in the other direction. That seems to be a case so if there's more outlets that are likely to sell alcohol basically all night long. People in this area tend to drink more. And we can see that, sort of, the change is strongest for the people that have the lowest risk score and the people with the highest genetic risk to drink they basically don't change their consumption as much. Which has a number of interesting implications in the sense that if these are the people that we actually want to target, and to have an impact on their behaviour. These tools don't really seem to be effective for that. 

Christine Garrington  10:52  

Yeah, that's brings us nicely to my next question really which is sort of what do we learn about how useful genetic information is in efforts to create and maybe evaluate public health policies?

Christian Zünd  11:02 

The question is obviously that we learn anything more from incorporating this genetic information from what we could get it if we just ask people do they drink a lot or not? And we tried to test this and it's, interestingly enough, you can even if you control for how much people drink, you can still use their genetic information to predict a lot about the risks of developing what, some of these associated diseases so a disease of the liver of the digestive system, these kind of things. Which seems to imply that either there's some other genetic channel, or more likely that the genetic information is actually in a sense a more comprehensive measure of their lifetime alcohol exposure and what you would get if you just ask them how much they drank in the last week or something like this. So there seems to be this, just sort of it being a valuable measure to predict these kind of medical conditions. And the other thing is that it gets at a kind of underlying inequality that is otherwise not observable. If we think about your genetic environment is sort of being inequality that's fixed the conception, where you can differ from your siblings for example for for some reason that sort of completely outside of your control. We can then see whether these policy interventions, mitigate these inequalities and sort of protect you against them or whether they actually lead to stronger inequality in the amount of alcohol that they actually consume. And it turns out that, at least for the policies we looked at they tend to actually increase inequality because the people that we would like to react most tend to be the people who react least these kind of policy interventions. 

Christine Garrington  12:45  

So, just finally, what would you say the takeaway message and then from those really interested in the public health in public health policies and to reduce the burden on overstretched health services, particularly in places like the UK but indeed elsewhere?

Christian Zünd  12:58

It's an open question so it could be done on other other kind of policies are actually able to protect against this sort of genetic inequality. So one thing that we would really like to study but unfortunately can't in the UK, are changes in the age we're allowed to drink. And so if our current data we can’t really look at this but in theory could be that if you've raised the age that people are allowed to drink, you will actually find a policy that mitigates against this inequality instead of sort of increasing it in the end, but that's an open question.

Christine Garrington  13:30  

Genes, pubs and drinks is research presented at the DIAL conference 2019 by Christian Zünd. You can find out more about the NORFACE funded DIAL projects at Thanks for listening to this episode of our podcast, which is presented and produced by Chris Garrington and edited by Elina Kilpi-Jakonen.