Oral Presentation Clinical Oncology Society of Australia Annual Scientific Meeting 2023

Striving towards better assessment of physical activity and sedentary behaviour: A novel cluster analysis approach using wrist accelerometry in childhood cancer survivors (#51)

Lauren Ha 1 2 , Claire E Wakefield 1 2 , Claudio Diaz 3 , David Mizrahi 4 , Richard J Cohn 1 2 , Karen Johnston 1 , Christina Signorelli 1 2 , Kalina Yacef 3 , David Simar 5
  1. Kids Cancer Centre, Sydney Children's Hospital, Randwick, NSW, Australia
  2. School of Clinical Medicine, Discipline of Paediatrics and Child Health, UNSW Sydney, Sydney, NSW, Australia
  3. School of Computer Science, The University of Sydney, Sydney, NSW, Australia
  4. The Daffodil Centre, The University of Sydney, a joint venture with Cancer Council NSW, Sydney, NSW, Australia
  5. School of Health Sciences, UNSW Sydney, Sydney, NSW, Australia

Aims: Physical activity (PA) levels are typically quantified as a total amount using time spent in light, moderate, or vigorous intensity, yet overlook the transient nature of PA such as whether activity is concentrated in certain parts of the day. We analysed PA behaviours using cluster analysis to explore various behaviour profiles in childhood cancer survivors. Understanding PA patterns may assist in tailoring exercise programs to support sedentary populations such as cancer survivors.

Methods: We measured survivors’ PA levels over seven consecutive days using wrist accelerometery (GeneActiv). To identify PA behaviours, we used bouts of physical activity characterised by various intensities (low/moderate/vigorous) and durations (short/moderate/long), then used these as features to cluster survivors’ daily and hourly behaviours. Using logistic regression, we calculated the likelihood of survivors being in more active clusters adjusting for the potential effects of age, sex, and time since treatment completion.

Results: 37 survivors (aged 11.7±3.0 years) engaged in mean 36.3 (SD=19.0) min/day of moderate-to-vigorous physical activity (MVPA) and 4.1 (SD=1.9) hrs/day of sedentary activity. Most survivors (86%) did not meet recommended PA guidelines (≥60min/day). On average, survivors achieved ≥60min on 1.1 (1.5) days/week. We identified five clusters: (i) most active (prevalence 11%), (ii) active (22%), (iii) moderately active + moderately sedentary (35%), (iv) moderately active + high sedentary (5%), (v) least active (27%). More frequent and sustained bouts of MVPA occurred on weekdays (13:00, 15:00 and 17:00) and prolonged sedentary activity occurred on weekends (8:00, 13:00 and 14:00). Younger survivors and those with less time since treatment completion were more likely to be active.

Conclusions: Many survivors are physically inactive, exacerbating their cardiometabolic risk further. Our approach provides an insightful analysis into the transient nature and timing of survivors’ movement behaviours. Our findings may help to develop targeted interventions to alter patterns of PA and sedentary behaviours in survivors.