| Objective: Diabetes,as a lifelong disease,requires long-term self-management to stabilize blood glucose and prevent diabetic complications.Computer information technology can be applied as an effective management method for diabetics to improve blood glucose control and prevent complications.This study discussed the establishment and application of information technology system in the field of diabetes management and enumerated the relevant application cases.The study has 4 parts,including the establishment of individualized diabetes management expert system based on clinical guidelines,the clinical application method of cross sectional data and follow-up data in management system,the effect of diabetes management by mobile health Application(APP)of smartphone combined with the expert system.Methods: The system was developed with Visual Basic 6.0,using SQL 2005 Server database as database sever.The system has collected clinical data through text data extraction technology,and established an expert system for diagnosis and decision making by setting up a production system based on the 2013 China Guidelines for Type 2 Diabetes as knowledge base.Found out the association between Nonalcoholic fatty liver disease(NAFLD)and C-IMT elevation in patients with Type 1 diabetes(T1DM)by through the analysis on the cross-sectional data in the expert system.Through the analysis of the baseline data and follow-up data in the expert system we found out the relationship between waist circumference and carotid intima media thickness(C-IMT)among newly-diagnosed male diabetics,and explored the waist circumference cutoff for identifying C-IMT elevation.Established management system by mobile health combined with expert system,and evaluated its effect through a randomized controlled trial among diabetics,and explored the effect of smartphone APP on the diabetes blood glucose and the self-management.Results:1.This study successfully developed the “Electronic Medical Record Partner –Diabetes individualized management System”(Software Copyright No.2017SR092402),which has already been on line in Department of Endocrinology and Metabolism in Shanghai Sixth People's Hospital(Sixth People's Hospital Affiliated to Shanghai Jiao Tong University).More than 12,000 diabetics are currently under the management with the system with nearly 1 million data records,of which 2567 patients with follow-up records.The accuracy of the data extraction from the discharge summaries by the system is 98.1% and the recall ratio is 98.2%.The diagnosis of Diabetic nephropathy III and IV Stage,Diabetic peripheral vascular disease by the aid system is basically consistent with the clinical diagnosis(Kappa 0.81~0.97)?The clinical acceptance rate of the treatment proposed by the system was 80%.35 diabetics treated with oral antidiabetic drugs were applied the treatment recommended by the system for 6 months,and got the results of the Hba1 c being significantly reduced by 1.72%(P<0.01)?2.722 patients with Type 1 Diabetes were screened from the system database,in which the prevalence of NAFLD is 15.9%.The subjects were divided into two groups according to with or without NAFLD.We found that the C-IMT of the NAFLD group was significantly higher than the control group(0.81±0.25 vs.0.69±0.18 mm,P<0.001).The detection rate of carotid plague was also higher(28.9% vs.16.9,P<0.05).After adjusting age,sex and common cardiovascular risk factors,the analysis of covariance showed that the difference in C-IMT between the NAFLD group and the control group remained(P <0.01).Multiple linear regression analysis showed that age,NAFLD and C reactive protein were independent risk factors for C-IMT in patients with Type 1 diabetes.After stratifying the patients with T1 DM into 4 groups according to the CRP quartiles,we found that the prevalence of NAFLD increased gradually with growing CRP at the rate of 3.4%,10.6%,20.6% and 33.1% respectively(P<0.01).The mean C-IMT also increased with CRP(P<0.01).3.578 newly diagnosed male diabetes were selected from the database of the system and divided into groups according to the quartiles of C-IMT.We found that the waist circumference increased with C-IMT(Trend P<0.01).The ROC curve analysis showed that 90 cm was the optimal cutoff of waist circumference for identifying C-IMT elevation among male Type 2 diabetic patients,which represented a sensitivity of 50.3% and a specificity of 61.3%.Spearman correlation analysis and partial correlation analysis showed that C-IMT was positively correlated with waist circumference(P<0.01).When the subjects were grouped according to waist circumference by 5cm intervals,we found that C-IMT was significantly elevated with the increase of waist circumference,and reached to a platform in about 90cm(waist <90cm: trend P<0.01;waist >90cm,trend P=0.51).Stepwise multiple regression analysis showed that waist circumference was one of independent risk factors for C-IMT.231 patients in the subjects were followed up for 2.2 years averagely,the C-IMT of the patients of whose the baseline waist circumference was 90 cm was significantly elevated(P<0.05),while no significant elevation in C-IMT was detected in the patients of whose the baseline waist circumference was less than 90 cm.Stepwise multiple regression analysis showed that the odds ratios of baseline waist circumference over 90 cm for C-IMT elevation was 1.132(95% CI 1.043-1.431,P<0.05).4.234 diabetics were selected in the system and managed with smartphone APP combined with the expert system.The patients were divided into 3 groups.Group A: conventional diabetes management group;Group B: Self-management with smartphone APP;Group C: interactive management with smartphone APP.194 of the 234 patients completed 6 month follow-up(83.9%).After 6 month management,the amplitude of Hba1 c reduction in different groups were(1.35±1.52)% in Group A,(1.48±1.51)% in Group B and(2.07±1.58)% in Group C.The HbA1 c in Group C was decreased more significantly than Group A and Group B(P<0.05).The control rate of Hba1 c were 22.2% in Group A,28.4% in Group B and 36.8% in Group C.The control rate in which Group C was remarkably higher than in Group A(P=0.04),and not significantly different from that in Group B(P=0.04).After 6 month management,the fasting blood glucose level of Group C was significantly lower than that of Group A and B.The HDL level of Group C was significantly higher than that of Group A.The score of diabetes knowledge and the empowerment of diabetes in the Group C were significantly higher than those in the Group A,and the Zung self-rating anxiety scale score was significantly lower than that in Group A(P<0.05).Conclusion: The individualized diabetes management expert system was developed successfully based on the clinical guideline with features of powerful functions,friendly interface,and high efficiency in obtaining clinical data.It can be used for diagnosis and decision making on oral antidiabetic drugs treatment,which makes it a useful tool for diabetes management.Through the analysis on the data from the system,we found that NAFLD was an independent risk factor of C-IMT elevation in patients with T1 DM,and among newly-diagnosed diabetic men patients,waist circumference over 90 cm not only was associated with C-IMT elevation,but also predicts the progress of atherosclerosis.Diabetics can further reduce glycated hemoglobin and improve the control rate through the application of smartphone APP based on the expert system.They can reduce Hba1 c faster,improve blood lipid metabolism,and improve their self-management level significantly through smartphone APP online interactive management. |
PDF Full Text Request