Clustering Patterns of Physical Fitness, Physical Activity, Sedentary, and Dietary Behavior among School Children

• Published in: Childhood obesity Vol. 16; no. 8; pp. 564 - 570
• Main Authors: López-Gil, José Francisco, Brazo-Sayavera, Javier, García-Hermoso, Antonio, Camargo, Edina Maria de, Yuste Lucas, Juan Luis
• Format: Journal Article
• Language: English
• Published: United States Mary Ann Liebert, Inc 01.12.2020
• Subjects: Body composition; Body fat; Cluster analysis; Diet; Exercise; Health behavior; Lifestyles; Obesity; Overweight; Physical fitness; Schools; Screen time; Sedentary behavior; Teenagers; Japan; Spain; nutrition; risk behavior; health promotion program; obesity; physical activity
• ISSN: 2153-2168; 2153-2176
• DOI: 10.1089/chi.2020.0185

Some healthy lifestyle behaviors may have a greater impact on childhood obesity in combination, compared to the independent effects of those behaviors in an isolated manner. The present study aimed to identify the different healthy lifestyle patterns of children according to their physical fitness, physical activity (PA) patterns, screen time, and adherence to the Mediterranean Diet, as well as to examine the association between anthropometric indicators and the membership to a certain cluster. A final sample of 353 Spanish school children (45.9% females) from the Region of Murcia (Spain) was included in this study (aged 6-13). First, we conducted a hierarchical cluster analysis using Ward's method; based on squared Euclidean distances. Second, we used the k-means cluster analysis to get the final cluster solution. Three different clusters were established: Cluster 1 [high cardiorespiratory fitness (CRF), PA, and Mediterranean Diet], Cluster 2 (low CRF, PA, and Mediterranean Diet + high muscular strength), and Cluster 3 (low physical fitness and PA). Cluster 3 had negative values in all the health-related variables analyzed. Regarding the anthropometric parameters analyzed (BMI, tri-ponderal mass index, waist-to-height ratio, and body fat percentage), Cluster 3 presented the highest values in all anthropometric parameters than the other two clusters (  < 0.001), while Cluster 1 showed the lowest values. The study has identified three clusters respect to health-related variables with the higher prevalence in the cluster established as the unhealthiest lifestyle. Also, cluster classification is associated to obesity indicators such as BMI, tri-ponderal mass index, waist-to-height ratio, and body fat percentage.