All-cause mortality prediction models in type 2 diabetes: applicability in the early stage of disease.
Ontology highlight
ABSTRACT: Aims
The rate of all-cause mortality is twofold higher in type 2 diabetes than in the general population. Being able to identify patients with the highest risk from the very beginning of the disease would help tackle this burden.Methods
We tested whether ENFORCE, an established prediction model of all-cause mortality in type 2 diabetes, performs well also in two independent samples of patients with early-stage disease prospectively followed up.Results
ENFORCE's survival C-statistic was 0.81 (95%CI: 0.72-0.89) and 0.78 (95%CI: 0.68-0.87) in both samples. Calibration was also good. Very similar results were obtained with RECODe, an alternative prediction model of all-cause mortality in type 2 diabetes.Conclusions
In conclusion, our data show that two well-established prediction models of all-cause mortality in type 2 diabetes can also be successfully applied in the early stage of the disease, thus becoming powerful tools for educated and timely prevention strategies for high-risk patients.
SUBMITTER: Copetti M
PROVIDER: S-EPMC8164049 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description: Not available
| S-EPMC10310973 | biostudies-literature
Project description:Most screening tests for T2DM in use today were developed using multivariate regression methods that are often further simplified to allow transformation into a scoring formula. The increasing volume of electronically collected data opened the opportunity to develop more complex, accurate prediction models that can be continuously updated using machine learning approaches. This study compares machine learning-based prediction models (i.e. Glmnet, RF, XGBoost, LightGBM) to commonly used regression models for prediction of undiagnosed T2DM. The performance in prediction of fasting plasma glucose level was measured using 100 bootstrap iterations in different subsets of data simulating new incoming data in 6-month batches. With 6 months of data available, simple regression model performed with the lowest average RMSE of 0.838, followed by RF (0.842), LightGBM (0.846), Glmnet (0.859) and XGBoost (0.881). When more data were added, Glmnet improved with the highest rate (+ 3.4%). The highest level of variable selection stability over time was observed with LightGBM models. Our results show no clinically relevant improvement when more sophisticated prediction models were used. Since higher stability of selected variables over time contributes to simpler interpretation of the models, interpretability and model calibration should also be considered in development of clinical prediction models.
| S-EPMC7371679 | biostudies-literature
Project description:BackgroundHigh serum resistin has been associated with increased risk of cardiovascular disease in the general population, Only sparse and conflicting results, limited to Asian individuals, have been reported, so far, in type 2 diabetes. We studied the role of serum resistin on coronary artery disease, major cardiovascular events and all-cause mortality in type 2 diabetes.MethodsWe tested the association of circulating resistin concentrations with coronary artery disease, major cardiovascular events (cardiovascular death, non-fatal myocardial infarction and non-fatal stroke) and all-cause mortality in 2,313 diabetic patients of European ancestry from two cross-sectional and two prospective studies. In addition, the expression of resistin gene (RETN) was measured in blood cells of 68 diabetic patients and correlated with their serum resistin levels.ResultsIn a model comprising age, sex, smoking habits, BMI, HbA1c, and insulin, antihypertensive and antidyslipidemic therapies, serum resistin was associated with coronary artery disease in both cross-sectional studies: OR (95%CI) per SD increment = 1.35 (1.10-1.64) and 1.99 (1.55-2.55). Additionally, serum resistin predicted incident major cardiovascular events (HR per SD increment = 1.31; 1.10-1.56) and all-cause mortality (HR per SD increment = 1.16; 1.06-1.26). Adjusting also for fibrinogen levels affected the association with coronary artery disease and incident cardiovascular events, but not that with all cause-mortality. Finally, serum resistin was positively correlated with RETN mRNA expression (rho = 0.343).ConclusionsThis is the first study showing that high serum resistin (a likely consequence, at least partly, of increased RETN expression) is a risk factor for cardiovascular disease and all-cause mortality in diabetic patients of European ancestry.
| S-EPMC3670852 | biostudies-literature
Project description:OBJECTIVES: This study aimed to identify clinical features associated with premature mortality in a large contemporary cohort of adults with type 1 diabetes. RESEARCH DESIGN AND METHODS: The Finnish Diabetic Nephropathy (FinnDiane) study is a national multicenter prospective follow-up study of 4,201 adults with type 1 diabetes from 21 university and central hospitals, 33 district hospitals, and 26 primary health care centers across Finland. RESULTS: During a median 7 years of follow-up, there were 291 deaths (7%), 3.6-fold (95% CI 3.2-4.0) more than that observed in the age- and sex-matched general population. Excess mortality was only observed in individuals with chronic kidney disease. Individuals with normoalbuminuria showed no excess mortality beyond the general population (standardized mortality ratio [SMR] 0.8, 95% CI 0.5-1.1), independent of the duration of diabetes. The presence of microalbuminuria, macroalbuminuria, and end-stage kidney disease was associated with 2.8, 9.2, and 18.3 times higher SMR, respectively. The increase in mortality across each stage of albuminuria was equivalent to the risk conferred by preexisting macrovascular disease. In addition, the glomerular filtration rate was independently associated with mortality, such that individuals with impaired kidney function, as well as those demonstrating hyperfiltration, had an increased risk of death. CONCLUSIONS: An independent graded association was observed between the presence and severity of kidney disease and mortality in a large contemporary cohort of individuals with type 1 diabetes. These findings highlight the clinical and public health importance of chronic kidney disease and its prevention in the management of type 1 diabetes.
| S-EPMC2699848 | biostudies-other
Project description:Underlying type 1 diabetes is a genetic aetiology dominated by the influence of specific HLA haplotypes involving primarily the class II DR-DQ region. In genetically predisposed children with the DR4-DQ8 haplotype, exogenous factors, yet to be identified, are thought to trigger an autoimmune reaction against insulin, signalled by insulin autoantibodies as the first autoantibody to appear. In children with the DR3-DQ2 haplotype, the triggering reaction is primarily against GAD signalled by GAD autoantibodies (GADA) as the first-appearing autoantibody. The incidence rate of insulin autoantibodies as the first-appearing autoantibody peaks during the first years of life and declines thereafter. The incidence rate of GADA as the first-appearing autoantibody peaks later but does not decline. The first autoantibody may variably be followed, in an apparently non-HLA-associated pathogenesis, by a second, third or fourth autoantibody. Although not all persons with a single type of autoantibody progress to diabetes, the presence of multiple autoantibodies seems invariably to be followed by loss of functional beta cell mass and eventually by dysglycaemia and symptoms. Infiltration of mononuclear cells in and around the islets appears to be a late phenomenon appearing in the multiple-autoantibody-positive with dysglycaemia. As our understanding of the aetiology and pathogenesis of type 1 diabetes advances, the improved capability for early prediction should guide new strategies for the prevention of type 1 diabetes.
| S-EPMC5491594 | biostudies-literature
Project description:This study aimed to develop a model to predict the 5-year risk of developing end-stage renal disease (ESRD) in patients with type 2 diabetes mellitus (T2DM) using machine learning (ML). It also aimed to implement the developed algorithms into electronic medical records (EMR) system using Health Level Seven (HL7) Fast Healthcare Interoperability Resources (FHIR). The final dataset used for modeling included 19,159 patients. The medical data were engineered to generate various types of features that were input into the various ML classifiers. The classifier with the best performance was XGBoost, with an area under the receiver operator characteristics curve (AUROC) of 0.95 and area under the precision recall curve (AUPRC) of 0.79 using three-fold cross-validation, compared to other models such as logistic regression, random forest, and support vector machine (AUROC range, 0.929-0.943; AUPRC 0.765-0.792). Serum creatinine, serum albumin, the urine albumin-to-creatinine ratio, Charlson comorbidity index, estimated GFR, and medication days of insulin were features that were ranked high for the ESRD risk prediction. The algorithm was implemented in the EMR system using HL7 FHIR through an ML-dedicated server that preprocessed unstructured data and trained updated data.
| S-EPMC9253099 | biostudies-literature
Project description:BackgroundMachine learning (ML) can include more diverse and more complex variables to construct models. This study aimed to develop models based on ML methods to predict the all-cause mortality in coronary artery disease (CAD) patients with atrial fibrillation (AF).MethodsA total of 2037 CAD patients with AF were included in this study. Three ML methods were used, including the regularization logistic regression, random forest, and support vector machines. The fivefold cross-validation was used to evaluate model performance. The performance was quantified by calculating the area under the curve (AUC) with 95% confidence intervals (CI), sensitivity, specificity, and accuracy.ResultsAfter univariate analysis, 24 variables with statistical differences were included into the models. The AUC of regularization logistic regression model, random forest model, and support vector machines model was 0.732 (95% CI 0.649-0.816), 0.728 (95% CI 0.642-0.813), and 0.712 (95% CI 0.630-0.794), respectively. The regularization logistic regression model presented the highest AUC value (0.732 vs 0.728 vs 0.712), specificity (0.699 vs 0.663 vs 0.668), and accuracy (0.936 vs 0.935 vs 0.935) among the three models. However, no statistical differences were observed in the receiver operating characteristic (ROC) curve of the three models (all P > 0.05).ConclusionCombining the performance of all aspects of the models, the regularization logistic regression model was recommended to be used in clinical practice.
| S-EPMC8520292 | biostudies-literature
Project description:ObjectiveIn diabetes, it remains unclear whether the coronary artery calcium (CAC) score provides additional information about total mortality risk beyond traditional risk factors.Research design and methodsA total of 1,051 participants, aged 34-86 years, in the Diabetes Heart Study (DHS) were followed for 7.4 years. Subjects were separated into five groups using baseline computed tomography scans and CAC scores (0-9, 10-99, 100-299, 300-999, and ?1,000). Logistic regression was performed adjusting for age, sex, race, smoking, and LDL cholesterol to examine the association between CAC and all-cause mortality. Areas under the curve with and without CAC were compared. Natural splines using continuous measures of CAC were fitted to estimate the relationship between observed CAC and mortality risk.ResultsA total of 17% (178 of 1,051) of participants died during the follow-up. In multivariate analysis, the odds ratios (95% CIs) for all-cause mortality, using CAC 0-9 as the reference group, were CAC 10-99: 1.40 (0.57-3.74); CAC 100-299: 2.87 (1.17-7.77); CAC 300-999: 3.04 (1.32-7.90); and CAC ? 1,000: 6.71 (3.09-16.87). The area under the curve without CAC was 0.68 (95% CI 0.66-0.70), and the area under the curve with CAC was 0.72 (0.70-0.74) (P = 0.0001). Using splines, the estimated risk (95% CI) of mortality for a CAC of 0 was 6.7% (4.6-9.7), and the risk increased nearly linearly, plateauing at CAC ? 1,000 (20.0% [15.7-25.2]).ConclusionsIn diabetes, CAC was shown to be an independent predictor of mortality. Participants with CAC (0-9) were at lower risk (0.9% annual mortality). The risk of mortality increased with increasing levels of CAC, plateauing at approximately CAC ? 1,000 (2.7% annual mortality). More research is warranted to determine the potential utility of CAC scans in diabetes.
| S-EPMC3114476 | biostudies-literature
Project description:BackgroundBoth type 1 and type 2 diabetes are well-established risk factors for cardiovascular disease and early mortality. However, few studies have directly compared the hazards of cardiovascular outcomes and premature death among people with type 1 diabetes to those among people with type 2 diabetes and subjects without diabetes. Furthermore, information about the hazard of cardiovascular disease and early mortality among Asians with type 1 diabetes is sparse, although the clinical and epidemiological characteristics of Asians with type 1 diabetes are unlike those of Europeans. We estimated the hazard of myocardial infarction (MI), hospitalization for heart failure (HF), atrial fibrillation (AF), and mortality during follow-up in Korean adults with type 1 diabetes compared with those without diabetes and those with type 2 diabetes.MethodsWe used Korean National Health Insurance Service datasets of preventive health check-ups from 2009 to 2016 in this retrospective longitudinal study. The hazard ratios of MI, HF, AF, and mortality during follow-up were analyzed using the Cox regression analyses according to the presence and type of diabetes in ≥ 20-year-old individuals without baseline cardiovascular disease (N = 20,423,051). The presence and type of diabetes was determined based on the presence of type 1 or type 2 diabetes at baseline.ResultsDuring more than 93,300,000 person-years of follow-up, there were 116,649 MIs, 135,532 AF cases, 125,997 hospitalizations for HF, and 344,516 deaths. The fully-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for incident MI, hospitalized HF, AF, and all-cause death within the mean follow-up of 4.6 years were higher in the type 1 diabetes group than the type 2 diabetes [HR (95% CI) 1.679 (1.490-1.893) for MI; 2.105 (1.901-2.330) for HF; 1.608 (1.411-1.833) for AF; 1.884 (1.762-2.013) for death] and non-diabetes groups [HR (95% CI) 2.411 (2.138-2.718) for MI; 3.024 (2.730-3.350) for HF; 1.748 (1.534-1.993) for AF; 2.874 (2.689-3.073) for death].ConclusionsIn Korea, the presence of diabetes was associated with a higher hazard of cardiovascular disease and all-cause death. Specifically, people with type 1 diabetes had a higher hazard of cardiovascular disease and all-cause mortality compared to people with type 2 diabetes.
| S-EPMC6858684 | biostudies-literature
Project description:AimsTo assess whether the use of beta-blockers influences mortality and the incidence of major cardiovascular events in patients with diabetes and coronary heart disease (CHD).Materials and methodsUsing data from the Bypass Angioplasty Revascularization Investigation 2 Diabetes trial, we performed Cox proportional hazards analysis to assess the effects of ?-blockers on all-cause mortality in 2244 patients with type 2 diabetes who had stable CHD with and without a history of myocardial infarction (MI)/heart failure with reduced left ventricular ejection fraction (HFrEF).ResultsAll-cause mortality in patients with MI/HFrEF was significantly lower in those receiving ?-blockers than in those not receiving ?-blockers (adjusted hazard ratio [HR] 0.60, 95% confidence interval [CI] 0.37-0.98; P ?=?.04), whereas that in patients without MI/HFrEF did not significantly differ (adjusted HR 0.91, 95% CI 0.76-1.32; P ?=?.64). Among patients with MI/HFrEF, all-cause mortality in those who received intensive medical therapy alone for CHD was significantly lower in those on ?-blockers than in those not on ?-blockers (adjusted HR 0.45, 95% CI 0.23-0.88; P ?=?.02); however, mortality in patients who received early revascularization for CHD was not significantly lower in those on ?-blockers (adjusted HR 0.81, 95% CI 0.40-1.65; P ?=?.57). The risk of major cardiovascular events in patients without MI/HFrEF was not significantly different between those on and those not on ?-blocker treatment.ConclusionsIn patients with diabetes and CHD, the use of ?-blockers was effective in reducing all-cause mortality in those with MI/HFrEF but not in those without MI/HFrEF.
| S-EPMC5484336 | biostudies-literature