Article Title



Lauren C. Bates, Gabriel Zieff, Lee Stoner, FACSM, Erik D. Hanson, FACSM. The University of North Carolina Chapel Hill, Chapel Hill, NC.

BACKGROUND: Movement behaviors (MB) across the 24-hour day, including physical activity, sedentary behavior (SB), and sleep, have important health implications. The objective of this study was to examine the clustering of MBs to analyze its association with cancer. METHODS: In December 2020, we surveyed a convenience sample of 746 adults (aged > 18 years) residing in USA. Participants self-reported moderate-to-vigorous physical activity (MVPA), SB, sleep, and cancer history. Z-scores were calculated for weekday (WD) SB, weekend (WE) SB, MVPA, and sleep. Cluster analysis was conducted using a two-step method including agglomerative hierarchical clustering with squared Euclidean distance and visual inspection to identify the number of clusters followed by K-means clustering. Clusters were labeled via distinguishing MBs z-scores (high: > 0.5, low: < -0.5). Partial eta-squared (ŋ2) measured effect size (small: 0.01, medium: 0.06, large: 0.14, respectively) and ANOVA was used to compare group by cluster. RESULTS: 112 cancer participants (40 ± 14 yr., 45% female, 64% employed, 15+ cancer types) and 634 non-cancer participants (38 ± 15 yr., 61% female, 62% employed), with matched socio-economic factors, participated in the study. Cluster analysis identified five MB clusters (ŋ2:0.054, p<0.001). Cluster 2 was characterized by high SB, very low MVPA, and very low sleep and had the most cancer participants (31% cancer). Three other clusters had significantly fewer cancer participants. Cluster 1 was characterized by low SB, low MVPA, low sleep (p=0.002, 16% cancer). Cluster 3 was characterized by high SB, some MVPA, high sleep (p<0.001, 11% cancer). Cluster 4 was characterized by high WD SB, low WE SB, low MVPA, high sleep (p<0.001, 10% cancer). Finally, cluster 5 was characterized by no SB, high MVPA, some sleep (p=0.070, 19% cancer) which tended to have fewer cancer participants. CONCLUSIONS: Lifestyle behaviors tend to cluster. Five 24-hour MB clusters were identified, with the largest number of cancer patients (31%) present in the most negative cluster (high SB, very low MVPA, and very low sleep). Considering negative MBs contribute to increased risk of cancer recurrence and adverse health outcomes, future research is warranted to identify feasible lifestyle interventions strategies targeting these co-occurring behaviors.

This document is currently not available here.