Device-measured movement behaviours and cancer incidence in a population sample of UK adults: Dual 24-hour analyses of postural and intensity compositions

IMPORTANCE Insufficient physical activity (PA) and excessive sedentary behaviour is associated with several cancers. Personalised approaches to increasing healthy movement behaviours and reducing unhealthy behaviours may be more effective than a one-size-fits-all approach. Exploring doses, posture allocation and intensities across the 24-hour spectrum of behaviours are important to develop broader guidelines. OBJECTIVE Using a novel dual-compositional approach, we assessed how time reallocation between 24-hour daily movement (postures and intensity bands) and sleep are associated with PA-related cancer incidence (a composite of 13 sites linked with physical inactivity). DESIGN, SETTING, AND PARTICIPANTS Prospective analysis of adults from the UK Biobank accelerometry subsample, followed-up by health linkage. EXPOSURES Participants daily movement was classified into two 24-hour compositions based on intensity and posture. Composition 1 (posture-focused): sleep duration, sedentary behaviour (SB), standing, moving at any intensity. Composition 2 (intensity-focused): sleep duration, sedentary time (ST), light PA (LPA), moderate PA (MPA), and vigorous PA (VPA). Secondary analysis combined VPA and MPA as moderate-to-vigorous PA (MVPA). MAIN OUTCOMES AND MEASURES PA-related cancer diagnoses were captured from health linkage. Cox-proportional hazards models were adjusted for age, sex, education, smoking, alcohol, diet, parental cancer history, cardiovascular disease and medication use. RESULTS Analyses included 59,218 (55% female) participants (mean [SD] age: 61.7 [7.8] years (y)), with a median follow-up of 8.0y [IQR: 7.4-8.5y; 464,640 person years] with 2,385 (4%) incident cancer events. Even among the average, active participant, replacing any behaviour with moving was associated with lower cancer risk, e.g. theoretically replacing 15 min of sleep or SB with 15 min of moving was associated with hazard ratios (HR) of 0.98 (95% Confidence Interval (95%CI): 0.97-0.99) and 0.98 (95%CI: 0.97-0.99), respectively. Similar risk reduction required 30 min of additional standing in place of sleep or ST. Regarding intensity, replacing even 3 min of any other behaviour with equivalent VPA was associated with lower risk (e.g. LPA HR: 0.96 (95%CI: 0.93-0.99); MPA HR: 0.96 (95%CI: 0.93-0.99). Greater MVPA too, in place of any behaviour, was robustly associated with lower risk. CONCLUSIONS AND RELEVANCE Moving at any intensity, in place of other postures, was associated with reduced risk of incident cancer. VPA, or MVPA in place of other movement intensities, were most strongly associated with lower cancer risk.