Physical Activity and Health Across the Lifespan (PAHAL)
Our research aims and expertise
Our team’s mission is to enhance our understanding of the relationship between physical activity and health across the lifespan. To do this we work across disciplines, using and combining methods from experimental research, epidemiology, data science, and implementation science.
Our group comprises diverse expertise including:
- Physical activity measurement methodology
- Physical activity epidemiology and public health
- Design and evaluation of activity interventions
- Physical activity data modelling and analysis
- Measuring associations between physical activity and health outcomes
- Systematic reviews and meta-analyses
Take a tour of our lab: Click and drag to rotate the camera.
Helping you with your research
Our complete data analytics service for public-funded research and clinical trials that use wearable devices to measure lifestyle, activity and sleep.
Research themes
Understanding how we move, and what that means for our health
Much of our research aims to determine which features of physical activity (for example, how much, how hard, how often, how long, when, and what) are most important for different health and disease outcomes. The results of this research can be used to inform public health and clinical guidelines. In addition, the results can be used to help understand what works for whom and when, which in turn can guide tailored interventions.
Measuring physical activity
Small, wearable devices called accelerometers enable us to collect very detailed information on the patterns of physical activity (intensity, frequency, duration, timing, type), as well as the total amount of activity, that people do in their everyday lives.
PAHAL has developed specialist research skills in the measurement of physical activity and sedentary behaviours using accelerometers, and supports a number of studies with accelerometer data collection for various research groups and studies, collaborating with Exeter colleagues and external research groups. These include:
Accelerometer data processing and analysis
In collaboration with colleagues in the Institute of Data Science and Artificial Intelligence (IDSAI) and the accelerometer manufacturer Activinsights, we have – and continue to develop – new methods of data processing and analysis for translating raw acceleration signals into a range of physical activity and sedentary behaviours such as walking, running and sitting etc.
In addition to estimating how much physical activity people do overall, we also compute measures of how they accumulate their physical activity. We do this by treating each period of physical activity as an event and then using event-based analysis to summarise how physical activity events are distributed within and between days.
Our own research and that of others has highlighted that different patterns of accumulation of physical activity are important for various health outcomes including bone health, physical function, recovery from total knee replacement and diabetes. Therefore, it is important to not only consider how much activity people do but how they accumulate it.
The PAHAL team have expertise in the statistical modelling of the wide range of accelerometer-derived measures they compute, in a variety of study designs. They regularly work with trial/study statisticians in clinical trials, behavioural trials, observational studies and systematic reviews to support them with the analysis of accelerometer data.
Case studies
Group members
Why we are interested
We are interested in this because three people with the same volume of weekly physical activity could have accumulated it through very different variations of daily volumes.
For example, in the image above, the graph on the left is from someone who is inactive during the working week (sedentary occupation) yet more active at the weekend (e.g. park-runner). In the middle is a shift worker whose physical activity is accumulated differently each day and on the right is from someone with an active occupation (e.g. postal delivery worker) but is relatively inactive at the weekend. All three people achieve the same weekly volume of physical activity.
Physical activity and physical function in older adults: Retirement in Action (REACT)
-study_crop.jpg)
Methods: A longitudinal analysis of n>600 adults (≥ 65 years) with measures of lower limb physical function and 7-day accelerometer at 0, 6 12 and 24 months. A stepping algorithm was applied to the raw accelerometer data to obtain ‘step counts’ achieved at different stepping cadences.
Findings: The 24-month change in physical function associated with a 1000 steps/day increase in ‘faster-paced walking’ was more than double that associated with the equivalent increase in total steps. This demonstrates that the potential benefit of more steps depends on the cadence at which those steps are taken. (Schoenfelder 2023).
Physical activity and bone health in pre- and post-menopausal women: UK Biobank
Methods: N~2500 women with cross-sectional measures of bone health (heel ultrasounds) and physical activity (1wk accelerometery). Multiple regression analyses - with incrementally increasing intensity thresholds - were carried out to determine the optimal impact required to benefit bone health.
Findings: We found that as little as 1-2min/day of physical activity of at least moderate-impact (≥750mg), was favourably associated with better bone health. We also demonstrated that a threshold commonly used in such studies – that of MVPA (≥250mg) - was not associated with bone health. (Stiles 2017)
Healthy Lifestyles Programme (HeLP) to prevent obesity in school children
Methods: 32 schools (N=1324 children were randomly assigned to the 10mth HeLP intervention or control. Physical activity was measured using 1wk accelerometery at 18mth.
Findings: The intervention did not lead to improvements in any measures of body mass or in any of the aggregate physical activity measures. (Lloyd 2017)
Physical activity during pregnancy (Moving through Motherhood)
Methods: Phase 1 involved an online survey of N=553 mothers and expectant mothers asking questions on their views and experiences of physical activity during pregnancy.
Findings: Phase 1 of the study found that i) the diverse experiences of pregnancy make PA challenging, ii) expectant mothers experience stigma when trying to be physically active, and iii) there is an absence of clear accessible guidance regarding physical activity during pregnancy. (Salmon 2021)
Physical activity and rehabilitation from heart failure (REACH-HF)
Methods: N=216 adults with heart failure were randomly assigned to a 12-month self-care, home-based support programme intervention or a ‘usual care’ control group. Physical activity was measured by 1wk accelerometry and fitness by the incremental shuttle walk test.
Findings: The intervention did not lead to any improvements in fitness or whole week physical activity, yet in a secondary analysis of the physical activity data, small increases in MVPA (15mins/day) were observed on weekdays though this did not quite reach statistical significance (p=0.06). (Dibben 2023)
Systematic reviews
- Cardiac rehabilitation and physical activity: systematic review and meta-analysis. (Dibben 2018)
- A comparison of the associations between bone health and three different intensities of accelerometer-derived habitual physical activity in children and adolescents. (Brailey 2022)
- The impact of selected methodological factors on data collection outcomes in observational studies of device-measured physical behaviour in adults (Pulsford 2023)
- Associations of habitual physical activity and carotid-femoral pulse wave velocity (Lear 2023)
