Back

24-hour Physical Activity, Sedentary, and Sleep Profiles in Individuals with Cancer: A UK Biobank Cohort Study

Le, F.; Dumuid, D.; D'Elia, O.; Haussmann, A.; Stamatakis, E.; Doherty, A.; Liao, Y.; Wiley, J. F.

2025-02-25 epidemiology
10.1101/2025.02.25.25322841 medRxiv
Show abstract

BackgroundThe 24h behaviour profile, including physical activity, sedentary time, and sleep, is disrupted following a cancer diagnosis and contributes to cancer-related outcomes. This study describes the 24h behaviour profiles of individuals with and without cancer. MethodsSeven days of accelerometer data from the UK Biobank (M{+/-}SDage = 62.3 {+/-} 7.8y; 56.4% female) were derived by machine learning models to assess the 24h behaviour profile in individuals with cancer (n = 10 152; M{+/-}SDyears since diagnosis = 7.4 {+/-} 6.1y) compared to individuals without cancer at their accelerometry measurement (including individuals with other conditions [n = 66 403], and healthy [n = 14 726]). Diagnoses were identified through ICD-10 coding within cancer registries, hospital inpatient records, self-reports, primary care, and death registers. Bayesian compositional data analysis compared 24h behaviour profiles between individuals with and without cancer, considering time since cancer diagnosis and cancer type. ResultsThe 24h behaviour profile of individuals with cancer consisted of 21.6 minutes in moderate-to-vigorous physical activity (MVPA), 303.5 minutes in light physical activity (LPA), 559.2 minutes in sedentary, and 555.7 minutes in sleep. Overall, they spent less time in MVPA -7.1 [-8.0, -6.1] min/day) and LPA -11.2 [-14.7, -7.7] min/day), accompanied by more sedentary time (+11.0 [7.2, 14.9] min/day) and sleep (+7.3 [4.6, 10.0] min/day), compared to healthy individuals. The least physically active profiles were observed within 1 year of cancer diagnoses and in individuals with blood, gastrointestinal tract, and lung cancer. ConclusionsThe 24h behaviour profiles differed by cancer history, time since diagnosis, and type. Specific behavioural trade-off strategies and support should be considered for cancer survivors, particularly in the year after diagnosis.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.