Research interests

As a large and diverse group, TARG have a number of research interests. Here you can find out about each workstream and what we are studying.

Modification of health behaviours

Workstream lead: Olivia Maynard

The overarching aim of this workstream is to develop behavioural interventions to encourage healthy behaviour change. We primarily investigate the effects of choice architecture interventions on behaviour. These interventions change the environments in which decisions are made and influence population health through an overall effect on the behaviour of many people.  As two of the most-pressing threats to public health, the primary focus of this research is tobacco and alcohol use, but the long-term goal is to extend this research into other health behaviours such as dietary behaviour, exercise and other substance use.

Ongoing work in this workstream has identified a number of choice architecture interventions which show promise as behaviour change techniques for tobacco and alcohol consumption.  For example, we have found that plain cigarette packaging, as compared with branded packaging, increases visual attention towards health warnings among non-smokers and light smokers, but not among daily smokers. Ongoing research funded by an ESRC grant awarded to Olivia Maynard is investigating why daily smokers avoid these health warnings and how we can increase attention to them by changing the warnings themselves.

We are extending this work to alcohol, and PhD student Carlos Sillero is examining ways of increasing attention to alcohol health warnings. We are also interested in the role of calorie and unit labelling for alcoholic beverages. Despite government calls for this information on alcoholic beverages, we find little evidence that they impact drinking behaviour.

We are also interested in the causes and consequences of problematic drug use. PhD student Maddy Dyer is investigating the relationship between alcohol use and anxiety in adolescence and young adulthood. PhD student Vicky Carlisle is working with the Bristol Drugs Project to investigate the causes and consequences of stigma for long-term opioid-substitution therapy users and develop an intervention to increase ‘recovery capital’ among this group. 

Key references: 

1. Maynard, O, 2017, ‘No Impact of Calorie or Unit Information on Ad Libitum Alcohol Consumption’Alcohol and Alcoholism.

2. Maynard, OM, Brooks, JCW, Munafò, MR & Leonards, U, 2017, ‘Neural mechanisms underlying visual attention to healthwarnings on branded and plain cigarette packs’Addiction, vol 112., pp. 662-672

3. Maynard, OM, Leonards, U, Attwood, AS, Bauld, L, Hogarth, L & Munafò, MR, 2015, ‘Effects of first exposure to plain cigarette packaging on smoking behaviour and attitudes: a randomised controlled study’BMC Public Health, vol 15.

Back to top

Acceptability and feasibility of delivering smoking cessation treatment in community mental health settings for people with depression

Workstream lead: Gemma Taylor

I hold a £502,378 Cancer Research UK Population Researcher Postdoctoral Fellowship Award. The fellowship project focuses on designing a smoking cessation intervention for people with depression, and testing the intervention in a feasibility and pilot study. More information about this work is available on the study website here:

I am a mixed methods researcher with interests in behavioural medicine and epidemiology. My work falls within the remit of the Tobacco and Alcohol Research Group and the MRC Integrative Epidemiology Unit at the University of Bristol and primarily focuses on the topic of smoking and mental health, with application of epidemiological, psychological and behavioural theory. I completed my PhD in Epidemiology in 2014 at The University of Birmingham which focused on the association between smoking cessation and mental health - part of this work was awarded BMJ’s “Best Research Paper Award” for 2014.

I have worked with mental health service users in long-term recovery and community-based settings, and my roles have included clinical interviewer and social support worker. During my time in these roles I have administered complex psychometrics and provided social support to service users of varying ages, with different types of psychological difficulties. 

Back to top

Investigating causality in the relationship between health behaviours and physical health

Workstream lead: Robyn Wootton

Lifestyle behaviours such as smoking, caffeine, alcohol and other substance use are strongly associated with many health outcomes. However, inferring causality from observational studies is difficult due to problems of confounding and reverse causality. Understanding the causal nature of these relationships is key to informing public health prevention strategies. 

The overarching aim of this workstream is to assess whether relationships between health behaviours and physical health outcomes are causal, and to investigate the direction of these relationships. To do this, we use causal analysis methods, including Mendelian randomization and negative control

Ongoing work in this workstream is investigating the causal role of smoking in a range of outcomes using a genetic variant associated with smoking heaviness in a Mendelian randomization approach. To achieve this, we have established the consortium for Causal Analysis Research in Tobacco and Alcohol (CARTA), which has over 30 contributing studies from 9 different countries. This work is now being extended to investigate the causal effects of caffeine.

We have also conducted work exploring the validity of using partner smoking during pregnancy as a negative control exposure for the effects of maternal smoking during pregnancy. We found that cotinine levels (a biological marker of tobacco exposure) were much higher in mothers who actively smoked during pregnancy than in mothers who were only exposed to passive smoking from their partner. Using negative control methods, we have also found that there is little evidence that maternal smoking during pregnancy increases risk of offspring smoking initiation via intrauterine effects.  

Key references:

1. Taylor AE, Morris RW, Fluharty ME et al. Stratification by Smoking Status Reveals an Association of CHRNA5-A3-B4 Genotype with Body Mass Index in Never Smokers. PLoS Genet. 2014. 4;10 (12).

2. Taylor AE, Davey Smith G, Bares CB, Edwards AC, Munafo MR. Partner smoking and maternal cotinine during pregnancy: Implications for negative control methods. Drug Alcohol Depend. 2014.1;139:159-63.

3. Taylor AE, Howe LD, Heron JE, Ware JJ, Hickman M, Munafo MR. Maternal smoking during pregnancy and offspring smoking initiation: Assessing the role of intrauterine exposure. Addiction. 2014. 109(6):1013-21.

Back to top

Investigating causality in the relationships between health behaviours and mental health

Workstream lead: Hannah Sallis 

Lifestyle behaviours such as smoking, drinking alcohol and using other substances are often found at higher levels in those with mental health problems such as depression or schizophrenia. Understanding the direction of causation in these relationships is not simple. Associations seen in observational data could be driven by other factors confounding the association, reverse causation, or due to bias. Understanding the causal nature of these relationships is key to informing public health prevention strategies. 

The overarching aim of this workstream is to assess whether relationships between health behaviours and mental health outcomes are causal, and to investigate the direction of these relationships. To do this, we use causal analysis methods, including Mendelian randomization and negative control

Ongoing work in this workstream is attempting to further explore recent evidence suggesting smoking might be a risk factor for schizophrenia, using the techniques outlined above to help infer causality. Polygenic risk scores that predict mental health outcomes are also being used to see if they predict substance use, and vice versa, allowing us to assess the possibility of causal relationships in either direction.  

Key references:

1. Gage, S. H. and Munafo, M. R. 2015. Rethinking the association between smoking and schizophrenia. Lancet Psychiatry, 2(2), 118-119

2. Gage, S. H., Hickman, M., Heron, H., Munafo, M. R., Lewis, G., Macleod, J and Zammit, S. 2014. Associations of cannabis and cigarette use with psychotic experiences at age 18: findings from the Avon Longitudinal Study of Parents and Children Psychological Medicine, Psychological medicine, 44(16), 3435-3444

Back to top

Data capture

Workstream lead: Andy Skinner

Data Capture is a cross-cutting theme developing novel approaches to capturing the variety of data of interest to researchers in the IEU – everything from molecules to behaviours. It supports a wide and varied portfolio of research projects across the IEU, while at the same time horizon scanning and collaborating with other world leading teams to bring innovative data capture technologies and techniques into the Unit. The Theme’s research interests include sensing health behaviours using wearable digital devices, crowd-sourcing of cognitive and behavioural data collection, continuous bio-sensing, low energy sensing technologies, on-body video capture, novel dietary assessment and technologies for behaviour change interventions.

Back to top

Phenotype refinement

Workstream lead: Glenda Lassi

Phenotype refinement is a key theme of ongoing research within the Tobacco and Alcohol Research Group. Phenotype refinement crucially provides a means by which to a) improve statistical power in genetic association studies, which may enable us to identify novel genetic variants associated with drug use and other complex behaviours/traits, and b) determine the specific mechanisms linking genetic variants to behaviours.

We employ a variety of approaches within this workstream. These include recall-by-genotype methods, laboratory-based behavioural assessments (e.g., smoking topography measures), and molecular genetics approaches employing innovative phenotypes (e.g., metabolite levels).

Ongoing studies include laboratory-based studies examining the impact of CHRNA5-A3-B4genotype on smoking topography in current smokers and on the response to nicotine in non-smokers. 

Key references:

1. Lassi G, Taylor AE, Timpson NJ, Kenny PJ, Mather RJ, Eisen T, Munafò MR (207). The CHRNA5-A3-B4 gene cluster and smoking: from discovery to therapeutics. Trends in Neurosciences 39(12): 851-861.

2. Ware, JJ & Munafò, MR (2015). Genetics of smoking behaviourCurrent Topics in Behavioural Neuroscience, 23, 19-36.

3. Ware, JJ, Timpson, N, Davey Smith, G & Munafò, MR (2014). A recall-by-genotype study of CHRNA5-A3-B4 genotype, cotinine and smoking topography: study protocol. BMC Medical Genetics, 15, 13.

Back to top

Investigating causal relationships between substance use and mental health outcomes

Workstrem lead: Jorien Treur

We know from observational studies that substance (ab)use is strongly associated with poorer mental health outcomes. These associations may be the result of shared risk factors (genetic or environmental) and/or they may reflect causal effects. Causality could act in two directions – substance use negatively impacting on mental health, or, mental health problems increasing substance use. Testing such causal effects is difficult with conventional epidemiological research designs, because of confounding and reverse causation. To overcome these difficulties we use Mendelian randomization, an instrumental variable technique that uses genetic variants as an instrument to test causal effects. Because genes are randomly passed on from parents to offspring, and an outcome variable cannot change a person’s genetic code, this approach suffers much less from confounding and reverse causality.

In this workstream, several addictive substances are investigated, including nicotine (smoking), cannabis and caffeine. A special focus in the work on caffeine is on creating high quality genetic instruments to perform Mendelian randomization with. It has been shown that genetic variants that predict an increase in cups of coffee consumed per day are also associate with an increased metabolism of caffeine and thus lower levels of its metabolite in the blood. This is important information when we want to test causal effects of caffeine use on other (mental health) outcomes.

Specific relationships being investigated are; substance use in relation to sleeping problems, substance use in relation to attention problems/ADHD and cannabis use in relation to schizophrenia.

Key references:

1. Bjørngaard JH, Nordestgaard AT, TaylorAE, Treur JL, Munafò MR, Nordestgaard BG, Åsvold BO, Romundstad P, SmithGD. Cigarette smoking increases coffee consumption: findings from a Mendelian randomization analysis. International Journal of Epidemiology, in press

2. Treur JL, Taylor AE, Ware JJ, Nivard MG, Neale MC, McMahon G, Hottenga JJ, Baselmans BM, Boomsma DI, Munafò MR, Vink JM. Smoking and caffeine consumption: A genetic analysis of their association. Addiction Biology, 2016, 22(4):1090-1102

3. Treur JL, Taylor AE, Ware JJ, McMahon G, Hottenga JJ, Baselmans BM, Willemsen G, Boomsma DI, Munafò MR, Vink JM. Associations between smoking and caffeine consumption in two European cohorts. Addiction, 2016, 111(6):1059-1068.

4. Treur JL, Willemsen GBartels M, Geels LM, van Beek JHDA, Huppertz C, van Beijsterveldt CEM, Boomsma DI, Vink JM. Smoking during adolescence as a risk factor for attention problems. Biological Psychiatry, 2015, 78(9):656-663

Back to top

Edit this page