| Diabetes mellitus(DM)is a metabolic disorder caused by a variety of causes,which is characterized by chronic hyperglycemia.Data from the International Diabetes Federation(IDF)showed that,in 2014,the global prevalence of diabetes was 8.3%,about 4.9 millions of people died of diabetes and its complications and diabetesrelated medical expenditures shared 11% of the global health expenditures.There is a high prevalence of diabetes in China and it is estimated that there are 113.9 millions of Chinese adults with diabetes.Diabetes,following cancer and cardiovascular disease,has become a serious non-communicable chronic disease which would do great harm to people's health and lives.Type 2 diabetes mellitus(T2DM)accounts for 90% of all the diabetes.Its pathogenic mechanism including insulin resistance and/or insufficient insulin secretion,both of which have a genetic susceptibility,as well as influenced by many environmental factors.Genetic and environmental factors have complicated action model on this disease,which is a challenge to explore the mechanism of the etiology,especially in the multivariate statistical analysis methods.In 1970 s,Structural Equation Modeling(SEM)was first proposed by J?reskog.This method is a confirmatory analysis technique.With this method,researchers can eventually build a path graph which would give a clear illustration of the relationships between the factors or latent variables,at the same time,the coefficients of the path may explain the effects of the factors or latent variables.For the SEM has more advantages than the traditional statistical methods,it will enrich the statistical methodology about multivariate analysis if it is applied to explore the etiology of T2 DM.In addition,there are also a large number of people with impaired glucose regulation(IGR)— glycemic levels that are higher than normal,but lower than diabetes thresholds,which is a high-risk stage of diabetes.There are approximately 493.4 millions of adults with IGR in China and it is estimated that up to 70% of the IGR will eventually progress to DM.Therefore,effective interventions in the IGR groups,focusing on diabetes risk factors,are very important and meaningful in the protection of public health.As lifestyle modification is the cornerstone of diabetes and related complications prevention,it is necessary to explore effective strategies to improve adherence to lifestyle modification.However,promoting adherence to lifestyle modification is a special challenge in the management of chronic disease.Miller(1983)created a novel method—motivational interviewing(MI).MI is defined as a client-centered,directive method for stimulating intrinsic motivation to change behavior by exploring and resolving ambivalence.MI has been widely applied in medical care settings to improve self-management behaviors in subjects with various diseases and has achieved certain success.Far as we know,there have been very small numbers of published studies evaluating the effects of MI in behavior modification in China and there has been no study using MI in people with IGR.Therefore,it is necessary to investigate the effects of MI in people with IGR in the community health service center,which would provide new ideas and methods for lifestyle management in people with diabetes or in higher risk of diabetes.This would also enrich the methodology about chronic disease prevention.Part ? Structural Equation Modeling in the Analysis ofFactors Affecting Glucose Metabolism A cross-sectional survey was conducted among the community residents who had no diabetes before in three communities in Nanjing.Information about the factors that might affect the incidence of T2 DM,such as lifestyle,physiological condition and related diseases,family history of DM,etc.were collected.We also evaluated indicators,such as blood glucose,HbA1 c,insulin resistance,insulin secretion function,etc.through the OGTT.Ordinal logistic regression analysis and SEM analysis were used separately to explore how the various factors affected blood glucose level.The results were as follows: 1.Ordinal logistic regression analysis of factors related with blood glucose level 4727 valid sample(30~89 years old,32.85% of them were men and 67.15% were women)were acquired,including newly diagnosed T2DM(584),IGR(1518)and normal glucose regulation(NGT)(2625).Multivariate ordinal logistic regression analysis with blood glucose level(NGT,IGR and T2 DM were ordinal variables in the model)as dependent variable showed that aging,simple overweight or obesity,abdominal obesity,hypertension,dyslipidemia,insulin resistance and family history of DM were all independent risk factors for elevated blood glucose level(all P<0.05).Increasing INR30 and INR120 were protective factors for blood glucose level.Gender,drinking and physical activity had no significant association with blood glucose level(all P>0.05).2.SEM analysis of the factors affecting glucose metabolismAll of the 16(independent or dependent)variables were included in the analysis.Based on the exploratory factor analysis,three latent variables were constructed: obesity(including BMI and WC),secretion function of the islet ?-cell(including INR30 and INR120)and blood glucose level(including FBG,PBG and HbA1c).Then,a path graph was built based on SEM(weighted least square method)to estimate the effects of related factors on insulin resistance or the secretion function of islet ?-cell,as well as their total effects on the blood glucose level.The results showed that aging,lack of physical activity,obesity,hypertension,dyslipidemia and family history of DM all had positive effects on the final blood glucose level through direct or indirect effects on insulin resistance or the secretion function of islet ?-cell,and the total coefficients(?)of corresponding path were 0.16,0.02,0.13,0.17,0.10 and 0.20 respectively.Smoking had a total positive effect(?=0.07)on the blood glucose level through a negative effect on the secretion function of islet ?-cell.Insulin resistance had a direct,positive effect on blood glucose level(?=0.99),and also had a negative effect(?=-0.65)on the blood glucose level through a positive effect on the secretion function of islet ?-cell.Insulin resistance had a total effect of 0.34 on the final blood glucose level.The secretion function of islet ?-cell had a direct,negative effect on the blood glucose level(?=-0.97).Part ? Effects of Motivational Interviewing on DiabetesPrevention in Adults with Pre-diabetes A randomized controlled trial was conducted in a community health service center in Nanjing.160 eligible residents(40~79 years old)with pre-diabetes(impaired glucose regulation,IGR)were recruited and randomly assigned to either the intervention group(the motivational interviewing group)or the control group(the traditional care group)with 80 subjects in each group.Both groups received one year's lifestyle intervention and 158 subjects were followed up at last.A comprehensive evaluation of the various indicators at baseline,the 6th month and 1 year of the intervention were conducted.The main results were as follows: 1.Comparison of self-management knowledge and self-efficacy Variance analysis for repeated measured variables was conducted to analyze the effects of group,time and interaction effect of group and time,and then pairwise comparison within group and between groups were conducted,with gender,age and education level as covariates.The results showed that,in the 6th month and 1 year of the intervention,the average scores of self-management knowledge and self-efficacy in the two groups were all significantly increased than that at baseline(all P<0.05).At 1 year of the intervention,the intervention group had significant higher scores of self-efficacy than that in the control group(P<0.05).2.Comparison of behavior modification Seven health related behaviors were observed in our study.The results showed that in the 6th month and 1 year of the intervention,the proportions of participants in action and maintenance stages in the 7 target behaviors were all significantly increased compared with that at baseline in both groups(except the item "relieve pressure" in the control group).In the 6th month of the intervention,the proportions of participants in action and maintenance stages in the MI group in items of "calorie control" and "monitoring" were significantly higher than that in the control group(all P<0.05).At 1 year of the intervention,the proportions of participants in action and maintenance stages in the MI group in items of "calorie control","increase dietary fiber intake","weight control" and "physical activity" were significantly higher than that in the control group(all P<0.05).3.Comparison of the outcomes about the glucose metabolic status and otherhealth related indicators after the intervention At 1 year of the intervention,the conversion rate to NGT in the intervention group was higher than that in the control group(55.1% vs 36.6%,adjusted RR=2.03,P<0.05).The conversion rate to DM in the control group was higher,but not statistically significant,than that in the MI group(12.7% vs 5.1%,P>0.05).At 1 year of the intervention,there was significant reduction of FBG,PBG,BMI,SBP,DBP and TG in the intervention group,compared with that at baseline,and the average reduction were 0.37 mmol/L,1.59 mmol/L,0.64 kg/m2,2.78 mmHg,2.28 mmHg and 0.34 mmol/L respectively(all P<0.05).There was significant reduction of FBG,PBG,BMI and DBP in the control group,compared with that at baseline,and the average reduction were 0.20 mmol/L,0.79 mmol/L,0.23 kg/m2 and 2.19 mmHg respectively(all P<0.05).The average reduction of FBG,PBG and BMI in the intervention group were significantly higher than that in the control group(all P<0.05).4.Comparison of quality of life At 1 year of the intervention,except the physical pain dimension,there was significant increase in the other seven dimensions in the intervention group,compared with that at baseline(all P<0.05).There was significant increase in two dimensions(the physiological function and social function dimensions)in the control group,compared with that at baseline.The average increase of the total,role physical and role emotional scores in the MI group were significantly higher than that in the control group(all P<0.05). |
PDF Full Text Request