Dynamic Development And Risk Prediction Of Metabolic Syndrome Based On A Health Check-up Cohort Study

BackgroundMetabolic syndrome(MS)is a clinical syndrome of obesity,hypertension,dyslipidemia and hyperglycemia,which seriously affects human health.At present,about one third of adults in the United States and one fifth of adults in China suffer from MS,and the prevalence of MS is still increasing year by year.The prevention and treatment of this disease is urgent.In view of this,scholars have done a lot of research,including looking for MS’s risk factors,biological markers,and the correlation between MS and some diseases,which is of great significance for the prevention and control of MS.As a cluster of diseases,the occurrence and development of MS is very complex.According to the diagnostic criteria of MS from the Chinese Diabetes Society.MS consists of four components,overweight or obesity,dyslipidemia,hypertension and hyperglycemia.Patients with at least three of the four components are diagnosed as MS.It can be seen that the process from healthy to MS should consider not only the change of the number of components,but also the combination of components.There are 12 different states,one non-component state,four single-component states,six two-component combination states,one state of MS.Each individual can maintain its original state or transfer to other states within a certain period of time.There are 144 forms of transition.There must be a pathophysiological process in which multiple components occur simultaneously in individuals.If we can clarify the sequence of MS components,it can be helpd ① find out which component is most likely to start MS,and which component will appear next,at the same time or in succession,②find out how much each component contributes to the development of MS,and whether there is one component which is the cause of other components.For healthy people,preventing the start of MS from the source,and preventing and controlling the core components that promote the rapid development of MS for the people who have already been ill can block the progress of MS to some extent.Markov model,as a well-known method to simulate the occurrence of chronic diseases,was used to describe and predict the development of MS in a study from Taiwan and in a study by our team in 2014.However,Taiwanese study only focus on people aged 18-45.Our team study have covered all people aged 18 and over.But the combination of the two components has not been divided in both researches.It is difficult to determine the whole development process of MS without further study of the dynamic changes of the two components state.After that,with a larger sample size,we can make a more comprehensive study.In this study,the subjects were divided into two gender groups.Each gender group was divided into 18-49 and over 50 years old.Therefore,this study establishes a 12-state Markov model in each gender and age group.We explore the whole process of MS,and predict the risk of MS from all possible initial states,which is of great significance for the prevention and control of MS.Objectives1.To search for the dynamic development of MS by establishing a 12-state Markov model to explore the process from non-component state to a single component or any two-component state and eventually develops into MS.2.To predict the risk of MS for people starting from different states,and explore the different risk of MS in the future for people starting from non-component state,any single component state or any combination of two components state.MethodsSubjects came from the health examination center of a hospital in Dongying City from January 2010 to December 2015.Subjects who age 18 years old and above and had at least two consecutive years to complete the health examination were involved.Subjects who had coronary heart disease,type I diabetes,stroke history and the absence of necessary information at baseline were excluded.The indicators used in the study include general information of the subjects,physical examination data and laboratory test data.The required sample size is estimated to be 5921.Finally,21 777 subjects aged 18 to 88 were included in this study.The diagnostic criteria of MS in this study is the Diabetes Society of Chinese Medical Association criteria.SPSS 24.0 software was used to make statistical analysis and calculate the transition probability of Markov model.Tree Age Pro 2011 software was used to build a 12-state Markov model and predict the risk of MS.One-step transition probability matrix based on Markov model is adopted.The calculation of the transition probability of Markov model is by calculating the annual probability of the subjects’ transition from one state to another or to maintain their own state every two years in each gender and age group,and then take the average of five years’probability as the transition probability of Markov model.Then,the results are brought into Markov model to predict the development of the subjects starting from different states after several years.Results1.The distribution of MS and its components during baseline and follow-up.A total of 21 777 subjects aged 18 to 88 were enrolled in this study.There were 14 250 males(65.44%)and 7527 females(34.56%).At baseline,the prevalence of MS increased with age in males,16.00%in the 18-49 age group and 29.31%in the over 50 age group.The prevalence of MS in females have the same trend.2.43%in the 18-49 age group and 17.80%in the over 50 age group.After 5 years of follow-up,the prevalence of MS showed an upward trend.In the male 18-49 age group,the baseline was 16.00%,followed by 21.77%,25.54%,36.44%,31.99%and 34.80%in each year.In men over 50 years old,the baseline was 29.31%,followed by 35.76%,39.65%,39.36%,47.26%and 49.16%in each year.In women aged 18-49,the baseline was 2.43%,followed by 3.75%,3.99%,5.11%,7.69%and 6.61%in each year.Among people over 50,the baseline was 17.80%,followed by 21.83%,21.17%,24.13%,25.83%and 27.01%in each year.2.The transition probability of Markov model in different sex and age groups.Males’ transition probability from non-component state or any single component state to MS state was higher than that of females,and the probability of males maintaining MS state is higher than that of females.In the 18-49 age group,during the transition from non-component state to other 11 states,the transition probability to simple dyslipidemia or simple overweight/obesity state was the highest,with 10.78%and 10.15%,8.45%and 7.15%in females,respectively.Among people over 50 years old,the transition probability from non-component state to simple hypertension or simple dyslipidemia was the highest,with 12.11%and 11.44%in males,8.17%and 11.30%in females,respectively.People who started with simple dyslipidemia were more likely to shift to dyslipidemia with overweight/obesity in the age group of 18-49,with 9.47%in males and 4.62%in females.In the age group over 50 years old,11.27%of males and 5.32%of females were more likely to transfer to the state of dyslipidemia with hypertension.In addition,males over 50 years of age who started with simple hyperglycemia had the highest probability to develop hyperglycemia with hypertension(20.63%).Other people are more likely to have dyslipidemia whether they start from simple overweight/obesity,hypertension or hyperglycemia.3.Risk prediction of Markov model in different gender and age groups.In the two age groups,the probability of men’s transition to MS state after 10 years from non-component state was higher than that of women,and 32.17%of men and 9.64%of women aged 18-49.However,with the increase of age,the difference of transition probability between men and women to MS state gradually narrowed,43.96%of men and 32.82%of women over 50 years old.People who started with hyperglycemia were more likely to develop MS after 10 years than other single component.42.15%of men and 23.90%of women aged 18-49,54.31%of men and 42.38%of women aged 50 and over,respectively.At the same time,people with two components state which included hyperglycemia were more likely to develop MS 10 years later.Conclusions1.Men in the same age group are more likely to have MS than women.2.Among 18-49 years old,MS usually starts from dyslipidemia or overweight/obesity,while in over 50 years old,MS usually starts from dyslipidemia or hypertension.3.When people begin with simple overweight/obesity,hypertension or hyperglycemia,dyslipidemia often follows next.4.Compared with other initial states,people who started with simple hyperglycemia or 2-component states that included hyperglycemia were more likely to develop MS.Hyperglycemia may be more likely to accelerate the onset of MS.