AUTHOR=Qasrawi Radwan , Abu Al-Halawa Diala 

TITLE=Cluster Analysis and Classification Model of Nutritional Anemia Associated Risk Factors Among Palestinian Schoolchildren, 2014

JOURNAL=Frontiers in Nutrition

VOLUME=Volume 9 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/nutrition/articles/10.3389/fnut.2022.838937

DOI=10.3389/fnut.2022.838937

ISSN=2296-861X

ABSTRACT=Nutritional inadequacy has been a major health problem worldwide. One of many health problems that results from it is anemia. Anemia is considered a health concern among all ages, particularly children, as it has been associated with cognitive and developmental delays. Researchers have investigated the association between nutritional deficiencies and anemia via various methods. As novel analytical methods are needed to ascertain the association, and reveal indirect ones, we aim to classify nutritional anemia using the cluster analysis approach. In this study, we included 4762 students between the ages of 10 and 17, attending public and UNRWA schools in the West Bank. Students’ 24-hour food recall and blood samples data were collected for nutrient intake and hemoglobin analysis. The K-means cluster analysis was used to cluster the hemoglobin levels into two groups. Vitamin B12, folate and iron intakes were used as indicators of nutrient intake associated with anemia and were classified per Recommended Dietary Allowance values (RDA). We applied the Classification and Regression Tree (CRT) model for studying the association between hemoglobin clusters and vitamin B12, folate and iron intakes, sociodemographic variables, and health-related risk factors, accounting for gender and age. Results indicated that 46.4% of the students were classified into the low hemoglobin cluster, and 60.7%,72,5% and 30.3% of vitamin B12, folate and iron intake were below RDA. The CRT analysis indicated that vitamin B12, iron, and folate intakes are important factors related to anemia in girls associated with age, locality, food consumption patterns and physical activity levels, while iron and folate intakes were significant in young boys associated with the place of residence and their mothers’ educational level. The deployment of clustering and classification techniques for identifying the association between anemia and nutritional factors might facilitate the development of nutritional anemia prevention and intervention programs that will improve schoolchildren health and wellbeing.