Understanding associations between substance use and mental health
Substance use (including tobacco, alcohol, and illicit drug use) and mental health problems co-occur at a much higher rate than in the general population. However, untangling the reasons for these associations is more challenging. Does the use of substances increase an individual's risk for poor mental health, or are people with mental health problems more likely to experiment with drug use? It is also possible that earlier life experiences could predict both substance use and poor mental health, and associations seen between the two are a result of this, rather than causal in nature. I am particularly interested in associations between smoking and cannabis use, and schizophrenia and depression.
Methods for ascertaining causality from observational data
A limitation of traditional observational epidemiology is that residual confounding can never be completely ruled out - there may be poorly measured confounders, or completely unknown confounders driving any associations seen in observational studies. For example, we cannot take a group of teenagers and randomly assign some to drink alcohol, some to smoke cigarettes, and some to use cannabis, and follow them up over time.
I am interested in using study designs to better ascertain causality, including Mendelian randomization, negative and positive control studies, and cross-contextual studies.