Risk of Hospitalisation among Patients with Type 2 Diabetes and its Determinants: A Logistic Regression Model

2020 ◽  
Vol 33 (02) ◽  
Author(s):  
Subha P P ◽  
◽  
Roy Scaria ◽  
Keyword(s):  
Type 2 Diabetes ◽  
Logistic Regression ◽  
Regression Model ◽  
Logistic Regression Model

scholarly journals Application of three statistical models for predicting the risk of diabetes

2019 ◽  
Vol 19 (1) ◽  
Author(s):  
Siyu Liu ◽  
Yue Gao ◽  
Yuhang Shen ◽  
Min Zhang ◽  
Jingjing Li ◽  
...  
Keyword(s):  
Neural Network ◽  
Risk Factors ◽  
Type 2 Diabetes ◽  
Logistic Regression ◽  
Decision Tree ◽  
Regression Model ◽  
Neural Network Model ◽  
Logistic Regression Model ◽  
Tree Model

Abstract Background At present, the proportion of undiagnosed diabetes in Chinese adults is as high as 15.5%. People with diabetes who are not treated and controlled in time may have various complications, such as cardiovascular and cerebrovascular diseases and diabetic foot disorders, which not only seriously affect the quality of life of people with diabetes but also impose a heavy burden on families and society. Therefore, prevention and control of type 2 diabetes is of great significance. Methods We constructed a logistic regression model, a neural network model and a decision tree model to analyse the risk factors for type 2 diabetes and then compared the prediction accuracy of the different models by calculating the area under the relative operating characteristic (ROC) curve and back-inputting the data into the model. Results The prevalence of type 2 diabetes in 4177 subjects who were not diagnosed with type 2 diabetes was 9.31%. The most influential factors associated with type 2 diabetes were triglyceride (TG) ≥ 1.17 mmol/L (odds ratio (OR) =2.233), age ≥ 70 years (OR = 1.734), hypertension (OR = 1.703), alcohol consumption (OR = 1.674), and total cholesterol≥5.2 mmol/L (TC) (OR = 1.463). The prediction accuracies of the three prediction models were 90.8, 91.2, and 90.7%, respectively, and the areas under curve (AUCs) were 0.711, 0.780, and 0.698, respectively. The differences in the AUCs after back propagation (BP) of the neural network model, logistic regression model and decision tree model were statistically significant (P < 0.05). Conclusion BP neural networks have a higher predictive power for identifying the associated risk factors of type 2 diabetes than the other two models, but it is necessary to select a suitable model for specific situations.

scholarly journals A Study on Significant Predictors for Prediction of Undiagnosed T2DM Using Binary Logistic Regression Model

2018 ◽  
Vol 7 (4.30) ◽  
pp. 355
Author(s):  
S. S. N. Zainal ◽  
M. J. Masnan ◽  
A. Ahmed ◽  
N. A. M. Amin
Keyword(s):  
Logistic Regression ◽  
Regression Model ◽  
Logistic Regression Model ◽  
Binary Logistic Regression ◽  
World Health ◽  
Binary Logistic Regression Model ◽  
Premature Deaths ◽  
The Many ◽  
Health Organization

Type 2 Diabetes Mellitus (T2DM) is a chronic disease that can cause premature deaths worldwide. Malaysia is one of the many countries that facing this serious epidemic. The World Health Organization (WHO) has also estimated that Malaysia would have 2.8 million people having T2DM disease in 2030. This study aims to identify significant predictors for prediction of undiagnosed T2DM patients in one of the highest prevalence states of T2DM. Binary logistic regression model proposed to predict the presence of T2DM among undiagnosed respondents. The selection of significant predictors using univariate, multivariate and backward stepwise selection was implemented in this study. The study concludes that four predictors were found significant for prediction of undiagnosed T2DM patients.

scholarly journals The frailty syndrome in older adults with type 2 diabetes mellitus and associated factors

2020 ◽  
Vol 23 (1) ◽  
Author(s):  
Bartolomeu Fagundes de Lima Filho ◽  
Antônia Gilvanete Duarte Gama ◽  
Vanessa da Nóbrega Dias ◽  
Eliza Mikaele Tavares da Silva ◽  
Fabricia Azevedo da Costa Cavalcanti ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Type 2 Diabetes ◽  
Older Adults ◽  
Type 2 Diabetes Mellitus ◽  
Logistic Regression ◽  
Logistic Regression Model ◽  
Lower Limbs ◽  
Frailty Syndrome ◽  
Frail Phenotype

Abstract Objective: To compare clinical-functional factors among groups in relation to the frailty syndrome (pre-frail and frail) phenotype profile in older adults with type 2 diabetes mellitus (DM 2). Methods: A descriptive, analytical, cross-sectional study with a quantitative approach was performed. A total of 113 diabetic older adults of both sexes were evaluated in terms of their personal, socio-demographic, clinical-functional, mental, cognitive and fragility phenotype data. The Chi-square test and a logistic regression model were used. Results: The mean age was 68.66±6.62 years, and the sample was mostly female (61.9%), illiterate or with an incomplete primary education (60.2%), pre-frail (52.2%), sedentary (79.6%), and had been diagnosed with DM2 for more than 5 years (58.3%). There was a significant association between “pre-frail and frail” individuals and schooling (p=0.004), social participation (p=0.004), a subjective perception of vision (p=0.004), glycated hemoglobin (p=0.036), limb pain (p=0.012), depressive symptoms (p=0.002) and mobility (p=0.004). The logistic regression model showed an accuracy of 93.6% and the significant variables were education (p=0.039), pain in the lower limbs (p=0.025) and risk of falls (p=0.033). Conclusion: among all the factors related to the “pre-frail” and “frail” phenotype, schooling, pain in the lower limbs and mobility were most related to the worsening of the syndrome and its progress.

scholarly journals Consumption of Nutrition Supplements Is Associated with Less Hypoglycemia during Admission—Results from the MENU Project

Nutrients ◽  
2019 ◽  
Vol 11 (8) ◽  
pp. 1832 ◽  
Author(s):  
Eyal Leibovitz ◽  
Fariba Moore ◽  
Irina Mintser ◽  
Anat Levi ◽  
Ron Dubinsky ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Regression Model ◽  
Medical Records ◽  
Logistic Regression Model ◽  
Medical Center ◽  
Nutrition Support ◽  
Oral Nutrition ◽  
Nutrition Supplement ◽  
Increased Risk

Aim: We studied the effect of the addition of an oral nutrition supplement (ONS) on the rate of hypoglyemia among hospitalized type 2 diabetes mellitus (DM) patients. Methods: In this retrospective analysis, all DM patients with hypoalbuminemia (albumin < 3.5 g/dL) admitted to internal medicine “E” at Wolfson Medical Center between 1 June 2016 and 30 April 2017 were included. One bottle of ONS (Glucerna, 330 KCAL, 28 g carbohydrates, 17 g protein, 17 g fat) was added to the morning meal. The consumption of the ONS was verified during the morning rounds. All glucose measurements were recorded automatically in the patients’ electronic medical records. A logistic regression model was used to evaluate the effect of the nutrition support on the occurrence of hypoglycemia. Results: 218 patients (mean age 77.4 ± 12.0 years, 63.3% female, mean albumin 3.13 ± 0.32 g/dL), of whom 27.9% had documented hypoglycemia during hospitalization were included. The patients consumed 69.5% ± 37.1 of the ONS provided, and ONS was started 4.3 ± 5.3 days from admission. A logistic regression model indicated that age (Odds ratio [OR] 1.048, 95% CI 1.014–1.083, p = 0.005), insulin treatment (OR 3.059, 95% CI 1.497–6.251, p = 0.002), and the day of ONS started from admission (OR 1.094, 95% CI 1.021–1.173, p = 0.011) were associated with an increased risk of hypoglycemia. Complete consumption of the ONS was associated with a reduced risk of hypoglycemia: OR 0.364, 95% CI 0.149–0.890, p = 0.027. Age, other DM medications and serum albumin did not affect the risk. Conclusion: The intake of a complete serving of ONS may be associated with a reduction of the risk of hypoglycemia among diabetes in-patients with hypoalbuminemia.

scholarly journals Single Nucleotide Polymorphisms Associated with Metformin and Sulphonylureas’ Glycaemic Response among South African Adults with Type 2 Diabetes Mellitus

2021 ◽  
Vol 11 (2) ◽  
pp. 104
Author(s):  
Charity Masilela ◽  
Brendon Pearce ◽  
Joven Jebio Ongole ◽  
Oladele Vincent Adeniyi ◽  
Mongi Benjeddou
Keyword(s):  
Diabetes Mellitus ◽  
Logistic Regression ◽  
Combination Therapy ◽  
Regression Model ◽  
South African ◽  
Logistic Regression Model ◽  
Model Analysis ◽  
Glycaemic Response ◽  
Genotype Combination

Aims: To examine the association of polymorphisms belonging to SLC22A1, SP1, PRPF31, NBEA, SCNN1B, CPA6 and CAPN10 genes with glycaemic response to metformin and sulphonylureas (SU) combination therapy among South African adults with diabetes mellitus type 2 (T2DM). Methods: A total of 128 individuals of Swati (n = 22) and Zulu (n = 106) origin attending chronic care for T2DM were recruited. Nine SNPs previously associated with metformin and SUs were selected and genotyped using MassArray. Uncontrolled T2DM was defined as HbA1c > 7%. The association between genotypes, alleles and glycaemic response to treatment was determined using multivariate logistic regression model analysis. Results: About 85.93% (n = 110) of the study participants were female and 77.34% (n = 99) had uncontrolled T2DM (HbA1c > 7%). In the multivariate (adjusted) logistic regression model analysis, the CC genotype of rs2162145 (CPA6), GG and GA genotypes of rs889299 (SCNN1B) were significantly associated with uncontrolled T2DM. On the other hand, the C allele of rs254271 (PRPF31) and the GA genotype of rs3792269 (CAPN10) were associated with controlled T2DM. A significant interaction between rs2162145 and rs889299 in response to metformin and SU combination therapy was observed. Conclusions: In this study, we reported the association of rs2162145 (CC) and rs889299 (GG and GA) with uncontrolled T2DM. We also reported the association of rs254271 (C) and rs3792269 (GA) with controlled T2DM in response to metformin and SU combination therapy. Furthermore, an interaction between rs2162145 and rs889299 was established, where the genotype combination GA (rs889299) and TT (rs2162145) was associated with uncontrolled T2DM.

scholarly journals Insulin Resistance and Contrainsular Response in Type 2 Diabetes Mellitus Patients with Acute Coronary Syndrome

2020 ◽  
Vol 8 (B) ◽  
pp. 943-948
Author(s):  
Dinara Sheryazdanova ◽  
Yelena M. Laryushina ◽  
Larissa Muravlyova ◽  
Lyudmila G. Turgunova ◽  
Assel R. Alina ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Insulin Resistance ◽  
Acute Coronary Syndrome ◽  
Logistic Regression ◽  
Regression Model ◽  
Logistic Regression Model ◽  
Diabetic Patients ◽  
Positive Correlation ◽  
Coronary Syndrome

BACKGROUND: The number of patients with diabetes mellitus (DM) is progressively increasing all over the world. Over the past three decades, the global burden of diabetes has increased from 30 million in 1985 to 382 million in 2015, and current trends indicate that the prevalence of diabetes grows progressively. The phenomenon of insulin resistance established in the majority of type 2 DM (T2DM) patients. T2DM is associated with β-cell deficiency, α-cell resistance to insulin, and reduced effects of incretin. However, the role of insulin and glucagon in the process of cardiovascular complications in diabetic patients is a matter of debate. AIM: Our study aims to estimate insulin resistance and the contrainsular response in patients with T2DM and acute coronary syndrome (ACS). METHODS: The 104 T2DM patients aged 18–70 years participated in the observational study carried out in the Karaganda regional cardiosurgery hospital and ambulatory. The first group included 37 patients hospitalized for ACS in the first 24 h of admission. The second group included 67 patients without ACS. Determination of insulin resistance and contrainsular response was provided using a multiplex immunological assay with XMap technology on Bioplex 3D. RESULTS: During the research, we have discovered a decreased level of glucagon and increased homeostasis model assessment of insulin resistance (HOMA-IR) in patients with T2DM diabetes and ACS. Evaluation of traditional correlation interactions of HOMA-IR and indicators of carbohydrate metabolism showed a positive correlation with fasting plasma glucose in both study groups (Group 1: R = 0.47, p = 0.003; Group 2: R = 0.41, p = 0.024). Glucagon-like peptide (GLP)-1 has a weak positive correlation with HOMA-IR only in the first group (R = 0.32, p = 0.006). Increased insulin resistance was associated with high GLP-1 levels and low glucagon. The logistic regression model established that an increased HOMA-IR index rises the chance of ACS by 10.6% (OR = 1.106 [95% CI 1.105–1.206], p = 0,021). The logistic regression model, reflecting the relation between glucagon and ACS, shows that increased glucagon reduces the ACS odds (OR = 0.989 [95% CI 0.979–0.999], p = 0.026). The adjusted regression model showed no significant influence of early presented factors on the probability of ACS. CONCLUSION: There is a trend toward elevated HOMA-IR insulin resistance index and decreased level of glucagon in diabetic patients with ACS.

Sign in / Sign up

Export Citation Format

Share Document