Craig Speirs

B.Sc. (Hons), B.Sc. (Hons)

University of Strathclyde

Ph.D. Student

IDATA (Learn more about this study)

IDATA-36

Jul 17, 2020

The development of novel outcomes from free-living physical behaviour and their relationship with determinants of health

Average daily accumulations of activity (step counts, upright time) derived from accelerometer data are typically used to measure volumes of physical behaviour in clinical and non-clinical groups. However, there is detailed information in the pattern of activities contained in accelerometer data which is not available from summary outcomes. For example, in investigations of existing datasets, we can see two people with the same volume of stepping that are accumulated in very different patterns of behaviour. If we can characterise these differences in behaviour, we may be able to better understand the relationship between physical behaviour and health, including differences in disease severity and capability within clinical groups.

By utilising the thigh inclination position captured by the activPAL activity monitor we can identify additional classes of activity, such as seated transport and lying, as well as measures of capability including sit to stand movements, which should allow us to classify a wider range of physical behaviours. By pairing this physical behaviour data with known indicators of health like BMI and VO2 max, it should be possible to investigate and identify outcomes that could act as digital biomarkers for specific diseases states, such as diabetes.

In this project we will use the accelerometer data captured by the activPAL activity monitor to investigate novel physical behaviour metrics, using the repeated observations to assess the consistency of these outcomes in the study population. We will then use the participant health and fitness characteristics that were collected as part of the iData study to test the value of these metrics as biomarkers for specific disease states and as indicators of general fitness.