Physical activity has proven health benefits to humans. Too often we focus on the benefits of exercise, when we should consider the full range of physical activity undertaken during the day. We seek to help people exploit those benefits more effectively by better quantification of those benefits through improved modelling of the impact of behavior over the waking day (or 24 hour day) on health outcomes using compositional analysis.
Compositions are typically sets of values or amounts that quantify the breakdown of something into exclusive parts. For example the breakdown of the waking day (or the 24 hour day) between different activity types. Such quantities are inherently interdependent and negatively correlated (if you do an extra hour of vigorous activity you must be doing an hour less of some other activity type(s)), and convey only relative information. Conventional multivariate analysis techniques must be adjusted to allow for this. Compositional analysis describes such techniques.
The breakdown of the day by behaviour type is a composition. It is not reasonable to assess the impact of an increase in one behavior type in isolation. Allowance should be made for the behavior type(s) replaced. Isotemporal substitution is currently the most popular approach for allowing for this, but compositional analysis makes available the full range of multivariate data analysis techniques to the problem of physical activity.
We aim to inform and educate about the techniques of compositional analysis, and the interpretation of results.
We hope to share the latest developments in compositional analysis that might be applied to the field of physical activity in future.
We are developing techniques for applying compositional analysis to longitudinal trials, and to the problem of physical activity in general, and hope to share details of our work in this area on this website.
We'll be sharing and commenting on the latest results of applying compositional analysis to the problem of physical activity.
We will be providing tools and articles to assist researchers in applying compositional analysis to their own studies and data, and welcome enquiries from other researchers interested in collaboration.
Ultimately, the objective of our research is to help people live healthier lives. What composition of physical activity will help people achieve this, and how can this best be achieved? This is a place to share our best ideas.