Metabolic syndrome in low-income communities/countries: Support vector machine analysis demonstrates a useful and cost-effective two-step screening strategy
Wen Jie Zhang, Hui Hui Yin, Meng Qing Xu, Kui Wang, Jin Huang, Jun Li, and Feng Li
Shihezi University School of Medicine, China
: Endocrinol Diabetes Res
Abstract
Metabolic syndrome (MS) is common in low-income communities/countries where primary healthcare facilities often have no biochemical blood tests. MS is diagnosed if 3 abnormal findings out of 5 bio-indicators including waist circumference (WC), blood pressure (BP), fasting blood glucose (GLU), triglycerides (TG), and high-density lipoprotein cholesterol (HDL-C). Support Vector Machine (SVM) is a computer algorism capable of recognizing combinational patterns and differentiating them by sensitivity and specificity (Table 1). Employing SVM’s ability of recognizing the best combinations of bio-indicators, this study aimed at establishing a sensitive and cost-effective screening strategy for MS in low-income communities. From 2012-2014 in a typical low-income rural township in China’s far-western Xinjiang Province, 3,276 individuals (1,590 males and 1,686 females) aged ≧18 years without prior diagnoses of MS were physically examined and blood biochemistry tested. MS was first diagnosed as an end result among these individuals based on the Joint-Interim-Statement (JIS) criteria. Following SVM analysis revealed that two non-blood bio-indicators, WC and BP together, were able to detect MS individuals with 57.0% sensitivity and 88.4% specificity. We then used SVM algorism to analyze abilities of diagnosing MS by WC+BP+TG, WC+BP+HDL-C, and WC+BP+GLU combinations. Interestingly, WC+BP+TG gained the best sensitivity of 72.5% with 88.1% specificity; WC+BP+HDL-C had a lesser sensitivity of 56.8% with 89.5% specificity; and WC+BP+GLU showed the least sensitivity of 56.6% with 90.9% specificity, respectively. Based on these observations, we propose a cost-effective “two-step screening strategy” for MS in low-income developing countries by which the “first screening step” is to use only non-blood indicators, WC+BP, which can be feasibly examined by paramedical personnel in “village/community clinic” and, individuals with abnormal WC+BP are then recommended to go through the “second screening step” of blood testing in secondary hospitals. Due to cost effectiveness, we recommend TG, other than HDL-C and GLU, as the first-line blood test for low-income populations
Biography
