During the initial phase of the RHYTHM IN DEMENTIA project, the main achievements were twofolds.
First, we validated the mathematical properties of rest-activity rhythm accelerometer-derived estimators and implemented these metrics into an open-access R package (
https://wadpac.github.io/GGIR/articles/chapter13_CircadianRhythm.html(opens in new window)) to allow their use in other studies. We also found that to have reliable measures of these metrics, 2 to 5 days of accelerometer wear are needed depending on the metric under consideration. This provided information for the next steps for all work packages.
Second, we set up the start of the data collection of a new cohort of patients from 2 memory clinics in Paris, named CIRCAME (
https://circame.fr/(opens in new window)) including measures of sociodemographic and behavioural factors, as well as health-related characteristics, a blood test to measure Alzheimer disease and neurodegeneration blood biomarkers, and accelerometer wear for 9 days to measure features of circadian rhythm. The number of patients included at this stage is above 500 and we aim to recruit 800-1000 more patients by the end of December 2026. These data will allow to determine specific circadian rhythm disruptions that characterise major dementia subtypes and stages of Alzheimer’s disease (WP2) and ascertain how circadian rhythm affects dementia prognosis in the patient cohort by examining cognitive and functional decline, hospitalisation, institutionalisation, and mortality (WP3).
In parallel, in WP1, we work on the identification of circadian rhythm profiles using data from 36 metrics of rest-activity rhythm, daytime activity, sleep, and chronotype. In order to validate the method used and the clusters identified, analyses are conducted in parallel in two population-based cohorts of older adults. The next step will be to examine the association of these clusters with subsequent cognitive decline and dementia onset.
We are also working on determining core circadian rhythm features to improve prediction of incident dementia as compared to known risk factors and predictors of dementia. The validity of the identified circadian rhythm risk score will be tested in external study to assess robustness of findings.