AIM: The purpose of this study was to explore the difference in the pattern of metabolic syndrome (MetS) in urban and rural populations in Korea using data mining techniques. SUBJECTS AND METHODS: In total, 1013 adults >30 yr of age from urban (184 males and 313 females) and rural districts (211 males and 305 females) were recruited from Gyeongsangnam-do, Korea. Modified National Cholesterol Education Program Adult Treatment Panel III criteria were used to identify individuals with MetS. We applied a decision tree analysis to elucidate the differences in the clustering of MetS components between the urban and rural populations. RESULTS: The prevalence of MetS was 33.2% and 35.2% in urban and rural districts, respectively (p=0.598). The decision-tree approach revealed that the combination of high serum triglycerides (TG) + high systolic blood pressure (SBP), high TG + low HDL cholesterol, and high waist circumference (WC) + high SBP + high fasting plasma glucose (FPG) were strong predictors of MetS in the urban population, whereas the combination of TG + SBP + WC and SBP + WC + FPG showed high positive predictive value for the presence of MetS in the rural population. CONCLUSIONS: Although no significant difference was found for the prevalence of MetS between the two populations, the differences in the clustering pattern of MetS components in urban and rural districts in Korea were identified by decision tree analysis. Our findings may serve as a basis to design necessary population-based intervention programs for prevention and progression of MetS and its complications in Korea.
AIM: The purpose of this study was to explore the difference in the pattern of metabolic syndrome (MetS) in urban and rural populations in Korea using data mining techniques. SUBJECTS AND METHODS: In total, 1013 adults >30 yr of age from urban (184 males and 313 females) and rural districts (211 males and 305 females) were recruited from Gyeongsangnam-do, Korea. Modified National Cholesterol Education Program Adult Treatment Panel III criteria were used to identify individuals with MetS. We applied a decision tree analysis to elucidate the differences in the clustering of MetS components between the urban and rural populations. RESULTS: The prevalence of MetS was 33.2% and 35.2% in urban and rural districts, respectively (p=0.598). The decision-tree approach revealed that the combination of high serum triglycerides (TG) + high systolic blood pressure (SBP), high TG + low HDL cholesterol, and high waist circumference (WC) + high SBP + high fasting plasma glucose (FPG) were strong predictors of MetS in the urban population, whereas the combination of TG + SBP + WC and SBP + WC + FPG showed high positive predictive value for the presence of MetS in the rural population. CONCLUSIONS: Although no significant difference was found for the prevalence of MetS between the two populations, the differences in the clustering pattern of MetS components in urban and rural districts in Korea were identified by decision tree analysis. Our findings may serve as a basis to design necessary population-based intervention programs for prevention and progression of MetS and its complications in Korea.
Authors: Soo Lim; Hayley Shin; Jung Han Song; Soo Heon Kwak; Seon Mee Kang; Ji Won Yoon; Sung Hee Choi; Sung Il Cho; Kyong Soo Park; Hong Kyu Lee; Hak Chul Jang; Kwang Kon Koh Journal: Diabetes Care Date: 2011-04-19 Impact factor: 19.112
Authors: Shaoyong Xu; Jie Ming; Chao Yang; Bin Gao; Yi Wan; Ying Xing; Lei Zhang; Qiuhe Ji Journal: BMC Public Health Date: 2014-02-01 Impact factor: 3.295