Ian is a fourth-year student at UNSW Sydney, where he is undertaking a Bachelor of Advanced Science, majoring in Mathematics, Statistics, and Molecular and Cell Biology. His research interests include applying probability and statistics, with a particular focus on biochemistry and public health. In his spare time, Ian participates in the university aikido club, has a passion for singing musical theatre, and enjoys the production and consumption of baked goods.
Estimating Population Attributable Fractions in the Presence of Competing Risks
How many lung cancer cases can we attribute to smoking? What proportion of cardiovascular disease cases would never have happened in a population without obesity?
These two questions are specific examples of a more general class of question – what is the probability that an incidence of disease is caused by exposure to some (preferably
modifiable) risk factor? Having an answer to such a question is vital for guiding public health policy decisions and targeting health initiatives.
The proportion in question is termed a Population Attributable Fraction (PAF), which can be used to assess the impact of certain risk factors (e.g. smoking and obesity). This
quantity is particularly useful as it takes into account not only relative risk (which measures the strength of association between a risk factor and outcome), but also prevalence of the risk factor in the population.
In this research project, we will develop a package for the R programming language to calculate PAFs from cohort studies with competing risks, along with their 95% confidence intervals.
Due to its ability to account for competing risks and support for external prevalences, this package will allow for projection into future situations, with applications in areas such as public health.