Using data mining technology to explore homocysteine at low levels

Abstract

A high homocysteine level is known to be an independent risk factor for cardiovascular diseases; however, whether or not low homocysteine level contributes to any damage to the body has not been extensively studied. Furthermore, acquiring healthy subject databases from domestic studies on homocysteine is not trivial. Therefore, we aimed to investigate the causality between serum homocysteine levels and health status and lifestyle factors, particularly with a focus on low serum homocysteine levels. Additionally, we discussed a systematic methodical platform for data collection and statistical analysis, using the descriptive analysis of the chi-square test, t test, multivariate analysis of variance, and logistic regression. This study was a cross-sectional analysis of 5864 subjects (i.e., clients of a health examination clinic) in Taipei, Taiwan during a general health check-up in 2017. The patients’ personal information and associated links were excluded. A sample group was selected as per the health criteria defined for this research whose data were processed using SPSS for descriptive statistical analysis using chi-square test, t test, multivariate analysis of variance, and logistic regression analysis. Those working for >12 hours/day had a higher homocysteine level than those working for <12 hours/day (P