Internship: Association of profiles of physical activity, sleep and circadian rhythm with cardiovascular and mortality risk
Team CRESS: EpiAgeing
Activity location: Université de Paris, Inserm U1153, 10 avenue de Verdun, 75010 Paris
Supervision: Séverine Sabia, Ian Meneghel Danilevicz, Sam Vidil
Topic: Physical activity, sleep and circadian rhythm have been suggested as important factors to maintain good health. However, as most studies have been conducted using self-reported data, many questions remain to be answered. Objective measurement using accelerometers enables different aspects of physical activity, sleep and circadian rhythm to be measured objectively, while accounting for their dependency. In the Whitehall II cohort, 4,000 participants aged between 60 and 83 wore a wrist accelerometer for 9 full days (9 * 24h).
These data are used to estimate numerous parameters of physical activity (in terms of intensity of activity, duration, distribution over the day), sleep (quantity, quality, fragmentation), rest/activity rhythm using parametric and non-parametric approaches, and also chronotype. Based on these data, which represent different dimensions of circadian rhythm through their measurement of the alternation between wakefulness and sleep, phenotypes (profiles constructed using clustering methods) have been identified.
The aims of this project are as follows:
- to study the association between the phenotypes identified and the risk of cardiovascular disease and mortality in the Whitehall II cohort, taking into account potential confounding factors,
- to replicate this analysis in the UK Biobank cohort in order to assess the robustness of the results.
To meet these objectives, methods specific to the analysis of survival data will be used. Thanks to a detailed collection of socio-demographic, lifestyle and health measures, it will be possible to control the analyses for numerous confounding factors. A very good knowledge of programming (Stata or R software) is essential to carry out the analyses successfully.
Contact : Séverine Sabia, email : severine.sabia@inserm.fr
