Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

Background

Low physical activity is associated with an increased risk of cardiovascular disease. However previous studies are largely based on self-reported data which are imprecise and prone to measurement error. Therefore uncertainty exists on the level of physical activity people should engage in, and the exact amount and type of activity that should be recommended. In response, large cohorts such as UK Biobank now aim to provide new insights on population physical activity through distributing wrist-worn accelerometers to participants.

However, flaws exist with current epidemiological methods to exploit recent advances in processing device data to define physical activity as an exposure variable. For example parametric statistics are used to describe non-normal time series data, and adjustments are not made for sleep time. Furthermore, time spent in different activities (sleep, sedentary behaviour, moderate intensity activity) are all co-dependent which creates issues of collinearity and spurious correlations. Therefore, new statistical methods are needed to assess objectively measured physical activity patterns and their association with cardiovascular disease.

Research Experience, Research Methods and Trainin

This project will involve:

  1. Conducting a systematic review of current methods to associate objectively measured physical activity with health outcomes
  2. Collaborating with biomedical engineers to extract key attributes of physical activity from activity monitors
  3. Developing statistical methods to robustly describe physical activity status from rich time-series device data
  4. Investigating the association between objectively measured physical activity patterns and subsequent risk of cardiovascular disease

Field Work, Secondments, Industry Placements and Training

To develop skills in big data as applied to biomedical research, this project will be placed in the Big Data Institute which combines researchers from genomics, epidemiology and infectious disease alongside those from computer science, statistics and engineering.

Prospective Candidate

A BSc, or ideally MSc, in epidemiology, statistics or a closely related discipline (with a substantive statistical component) and some previous statistical programming experience (e.g. R, Python, or Stata). 

Supervisors

Projects