To Study The Effects Of Lifestyle On Metabolic Syndrome With Logistic Regression

Objective：The metabolic syndrome (MetS) is a cluster of risk factorsassociated with diabetes mellitus and an increased risk of cardiovasculardisease. It is a serious threat to human health. Studies have shown that avariety of lifestyle factors that affect the metabolic syndrome. This study is todescribe the prevalence of and investigate the lifestyle factors associated withmetabolic syndrome and its components among6703adult railway workers ofa railway company in northern China and then provide scientific basis forhealth promotion.Method：In this cross-sectional survey，6703railway workers wererecruited from a railway company in northern China. We conducted athorough health examination including the measurements of waistcircumference，blood pressure，fasting plasma glucose，triglycerides andhigh-density lipoprotein cholesterol. Simultaneously， an investigationquestionnaire about their health status，marital status，working condition，disease history and lifestyle behaviors were implemented.We examined the prevalence of MetS and its components as defined bythe harmonious definition proposed in the joint interim statement by sixmedical organizations in2009. General characteristics was described withmedian，interquartile range and prevalence or proportion. Chi-square test wasused to compare the prevalence，proportion and do trend test for dose-responserelation between the targeting factors and MetS and its components. Theassociations of the lifestyle factors with the MetS and its components wereevaluated by created logistic regression models used to identify differences fordichotomized dependent variables.Results：1The overall crude prevalence of the MetS in the sample was21.57%， male had a higher standard prevalence (23.53%) than female (10.79%)，andthe prevalence of the MetS increased significantly with aging(tendency testP<0.001).2After adjusting for a series of some unalterable factors like age，gender， marital status and working conditions，significant associations werefound between metabolic syndrome and several lifestyle factors includingsmoking，drinking，physical activity and salty food. Compared with those whoneither smoked nor passive smoking，ORs for MetS were1.235(95%CI，1.004-1.519) for those who didn’t smoke but did passive smoking constantly，1.253(95%CI，1.043-1.505) for those who smoked1~10cigarettes/day，1.389(95%CI，1.145-1.685) for those who smoked11~20cigarettes/day and1.697(95%CI，1.130-2.548) for those who smoked more than21cigarettes/day，respectively. The ORs for MetS were1.354(95%CI，1.149-1.596) for thosewho drank3~4liang each time，1.679(95%CI，1.393-2.023) for those whodrank5~6liang each time，3.158(95%CI，1.924-5.185) for those who drankmore than7liang each time，respectively，compared with those who drank lessthan2liang each time. Compared with the subjects who did physical activityless than30min each time，no significant association were found in those whodid physical activity30~min nor60~min each time，but those who did morethan90~min each time were less likely to have metabolic syndrome than thoseless than30min with a OR=0.464(95%CI，0.304-0.710). Quality of sleep wassignificantly and negatively associated with MetS. The ORs of having MetS ingeneral and poor quality of sleep were0.764(95%CI，0.671-0.871) and0.756(95%CI，0.613-0.931)，respectively，compared with those who was in goodquality of sleep. Those who liked the taste of more salty food were more likelyto have metabolic syndrome with a OR=1.209(95%CI，1.067-1.369) comparedwith the contrast.3Lifestyle factors associated with the MetS components after adjustingfor a series of some unalterable factors like age，gender， marital status andworking conditions.Smoking， drinking， physical activity， sleep and salty food were significantly associated with waist circumference. Compared with subjectswhose smoking age were less than5years，the OR for those who smoked for6~15years was1.235(95%CI，1.085-1.406)， no statistically significantassociation was found in those whose smoking age were more than16years.Drinking amount was positively associated with the risk for elevated waistcircumference. Compared with those who drank less than3liang each time，the ORs of elevated waist circumference were1.354(95%CI，1.149-1.596) forthose who drank3~4liang each time，1.679(95%CI，1.393-2.023) for thosewho drank5~6liang each time and3.158(95%CI，1.924-5.185) for those whodrank more than7liang each time，respectively. Those who did physicalactivity more than5times per week had a lower risk of elevated waistcircumference than those who did sometimes， OR=0.729(95%CI，0.555-0.956).Quality of sleep was significantly and negatively associatedwith elevated waist circumference. The ORs of elevated waist circumferencein general and poor quality of sleep were0.695(95%CI，0.576-0.839) and0.857(95%CI，0.759-0.967)，respectively，compared with those who was ingood quality of sleep. Furthermore, the ORs (95%CI) for elevated waistcircumference in subjects who slept for6~8hours and more than8hours pernight compared with those who slept for fewer than6hours were0.817(95%CI，0.702-0.950) and0.743(95%CI，0.594-0.930)，respectively. TheORs for elevated waist circumference in those who liked the taste of moresalty food and those who liked fried food were1.228(95%CI，1.104-1.365)and1.208(95%CI，1.075-1.358)，respectively，compared with their contrast.Smoking and drinking were significantly associated with evaluated bloodpressure. Compared with current smokers，ORs for elevated blood pressurewere1.151(95%CI，1.005-1.318) in the subjects who neither smoked norpassive smoking，1.283(95%CI，1.102-1.493) in those who didn’t smoke butdid passive smoking constantly and1.287(95%CI，1.052-1.574) inex-smokers，respectively. The ORs in the subjects who drank once or twiceper week and more than3times per week were1.334(95%CI，1.165-1.529)and1.624(95%CI，1.325-1.990)，respectively，compared with those who drank less than twice a month. Furthermore，drinking amount was positivelyassociated with the risk for elevated blood pressure. Compared with those whodrank less than3liang each time， the ORs of elevated blood pressure were1.256(95%CI，1.090-1.447) for those who drank3~4liang each time，1.338(95%CI，1.132-1.580) for those who drank5~6liang each time and2.213(95%CI，1.353-3.621) for those who drank more than7liang each time，respectively.Smoking， drinking， physical activity， sleep and salty food weresignificantly associated with evaluated triglyceride(TG). Compared withsubjects whose smoking age were less than5years，the OR of elevated TGlevel in those who smoked for6~15years was1.233(95%CI，1.072-1.417)，no statistically significant association was found in those whose smoking agewere more than16years. The ORs in the subjects who drank once or twice perweek and more than3times per week were1.337(95%CI，1.158-1.544) and1.240(95%CI，1.008-1.526)，respectively，compared with those who drankless than twice a month. Furthermore， drinking amount was positivelyassociated with the risk for elevated TG level. Compared with those whodrank less than3liang each time，the ORs of elevated blood pressure were1.336(95%CI，1.151-1.552) for those who drank3~4liang each time，1.505(95%CI，1.265-1.790) for those who drank5~6liang each time and2.138(95%CI，1.314-3.448) for those who drank more than7liang each time，respectively. Compared with those who did physical activity occasionally，theORs for elevated TG level in subjects who did physical activity3~4times andwho did more than5times per week were0.756(95%CI，0.618-0.923) and0.509(95%CI，0.374-0.694)，respectively. Quality of sleep was significantlyand negatively associated with elevated TG level. The ORs of elevated TGlevel in general and poor quality of sleep were0.695(95%CI，0.576-0.839) and0.857(95%CI，0.759-0.967)，respectively，compared with those who was ingood quality of sleep. The subjects who liked the taste of more salty food weremore likely to have evaluated TG with a OR=1.134(95%CI，1.013-1.270)，compared with the contrast. Smoking，drinking and physical activity were significantly associatedwith low HDL-C. Compared with the subjects who neither smoked norpassive smoking，only current smokers was statistically significant associatedwith a higher risk for lower HDL-C. The OR was1.380(95%CI，1.186-1.605).No statistically significant association was found in those who didn’t smokebut did passive smoking constantly and ex-smokers. In the different drinkingamount levels，only the subjects who drank more than7liang each time hadhigher risk for low HDL-C compared with the subjects who drank less than2liang each time. The OR was2.056(95%CI，1.275-3.315). But in the differentdrinking frequency，the OR of low HDL-C was0.673(95%CI，0.528-0.858)for those who drank more than3times a week compared with those whodrank less than2times a month.Drinking frequency was significantly and negatively associated withelevated FPG levels. Compared with the subjects who drank less than2timesa month，the ORs for those who drank once or twice a week and those whodank more than3times a week were1.468(95%CI，1.152-1.870) and1.547(95%CI，1.142-2.096)，respectively.Conclusion：1. Male had a higher standard prevalence (23.53%) than female(10.79%)，and the prevalence of the MetS increased significantly with aging.2.After adjusting for variables age，gender，marital status and workingconditions，variables smoking，drinking，sleep and physical activity weresignificantly associated with MetS and its components. Especially drinkingwas the commonly owned factor of all of the MetS components. As a result，drinking was a main lifestyle factor which contributed to a higher risk forMetS，and followed by smoking，sleep and salty food. Do physical activitywas a protect factor. Not drinking, not smoking, more exercise, less salt andfried foods are recommended.