The application of multivariate adaptive regression splines in exploring the influencing factors and predicting the prevalence of HbA1c improvement.

BACKGROUND: Glycosylated hemoglobin (HbA1c) is directly proportional to the level of glucose in the blood, and it has been the gold standard to evaluate the status of long-term blood glucose levels. Exploring the factors that lead to HbA1c improvement is beneficial for effectively controlling of HbA1c levels. METHODS: Data collected from 52 hospitals in five cities in northern China were divided into training and test sets at a ratio of 7:3. The training set was used to build models, and the test set was used to evaluate the generalizability of the models. The performance of multivariate adaptive regression splines (MARS) models and logistic regression was evaluated, namely, the accuracy, Youden's index, recall rate, G-mean and area under the ROC curve (AUC) with 95% confidence intervals (CIs). RESULTS: The prevalence of improvements in HbA1c levels was 38.35%. Doses of insulin less than 13 U, more than 3 kinds of oral medicine, exercise frequency greater than once per week and 2 h postprandial blood glucose (2hPBG) less than 10.56 mmol/L were found to improve HbA1c. The following interactions were negatively associated with improvement in HbA1c levels: patients with relative complications and 2hPBG less than 10.56 mmol/L, type 2 diabetes mellitus (T2DM) duration more than 7 years and insulin dose less than 13U. Compared to logistic regression, the MARS model performed better in the above aspects, except for accuracy. CONCLUSIONS: Given the interaction between factors affecting HbA1c improvement, medical staff should conduct comprehensive interventions to further reduce HbA1c levels in patients. In this study, the MARS model was superior to the traditional logistic regression in improving HbA1c levels. MARS had greater generalizability because it not only considered nonlinear relations in the process of model fitting but also adopted cross-validation. Nevertheless, more studies are needed to provide evidence for this result.

The Diabetes Risk and Determinants of Transition from Metabolically Healthy to Unhealthy Phenotypes in 49,702 Older Adults: 4-Year Cohort Study.

OBJECTIVE: This study aimed to assess whether metabolically healthy obesity (MHO) increases the risk of diabetes and to explore how the occurrence of metabolic disorders affects the risk of diabetes and which factors determine metabolic health. METHODS: This study examined 49,702 older people without diabetes via the Binhai Health Screening Program in Tianjin. RESULTS: Compared with individuals with metabolic health and normal weight, the risk of diabetes was increased in older adults with MHO (hazard ratio [HR]: 1.786, 95% CI: 1.407-2.279) but was not significantly increased when metabolic health was characterized by the absence of metabolic abnormalities. The older adults who were initially affected by MHO and then converted to having an unhealthy phenotype had a higher diabetes risk than older individuals with stable and healthy normal weight (HR: 3.727, 95% CI: 2.721-5.105). Waist circumference was an independent predictor of the transition from a metabolically healthy status to an unhealthy status in all BMI categories (odds ratio: 1.059, 95% CI: 1.026-1.032). CONCLUSIONS: The MHO phenotype was associated with an increased incidence of diabetes in older adults. The presence of metabolic disorders in the group with MHO was associated with an increased diabetes risk and was predicted by the waist circumference at baseline.

Status of glycosylated hemoglobin and prediction of glycemic control among patients with insulin-treated type 2 diabetes in North China: a multicenter observational study.

BACKGROUND: Blood glucose control is closely related to type 2 diabetes mellitus (T2DM) prognosis. This multicenter study aimed to investigate blood glucose control among patients with insulin-treated T2DM in North China and explore the application value of combining an elastic network (EN) with a machine-learning algorithm to predict glycemic control. METHODS: Basic information, biochemical indices, and diabetes-related data were collected via questionnaire from 2787 consecutive participants recruited from 27 centers in six cities between January 2016 and December 2017. An EN regression was used to address variable collinearity. Then, three common machine learning algorithms (random forest [RF], support vector machine [SVM], and back propagation artificial neural network [BP-ANN]) were used to simulate and predict blood glucose status. Additionally, a stepwise logistic regression was performed to compare the machine learning models. RESULTS: The well-controlled blood glucose rate was 45.82% in North China. The multivariable analysis found that hypertension history, atherosclerotic cardiovascular disease history, exercise, and total cholesterol were protective factors in glycosylated hemoglobin (HbA1c) control, while central adiposity, family history, T2DM duration, complications, insulin dose, blood pressure, and hypertension were risk factors for elevated HbA1c. Before the dimensional reduction in the EN, the areas under the curve of RF, SVM, and BP were 0.73, 0.61, and 0.70, respectively, while these figures increased to 0.75, 0.72, and 0.72, respectively, after dimensional reduction. Moreover, the EN and machine learning models had higher sensitivity and accuracy than the logistic regression models (the sensitivity and accuracy of logistic were 0.52 and 0.56; RF: 0.79, 0.70; SVM: 0.84, 0.73; BP-ANN: 0.78, 0.73, respectively). CONCLUSIONS: More than half of T2DM patients in North China had poor glycemic control and were at a higher risk of developing diabetic complications. The EN and machine learning algorithms are alternative choices, in addition to the traditional logistic model, for building predictive models of blood glucose control in patients with T2DM.

A novel model for malaria prediction based on ensemble algorithms.

BACKGROUND AND OBJECTIVE: Most previous studies adopted single traditional time series models to predict incidences of malaria. A single model cannot effectively capture all the properties of the data structure. However, a stacking architecture can solve this problem by combining distinct algorithms and models. This study compares the performance of traditional time series models and deep learning algorithms in malaria case prediction and explores the application value of stacking methods in the field of infectious disease prediction. METHODS: The ARIMA, STL+ARIMA, BP-ANN and LSTM network models were separately applied in simulations using malaria data and meteorological data in Yunnan Province from 2011 to 2017. We compared the predictive performance of each model through evaluation measures: RMSE, MASE, MAD. In addition, gradient-boosting regression trees (GBRTs) were used to combine the above four models. We also determined whether stacking structure improved the model prediction performance. RESULTS: The root mean square errors (RMSEs) of the four sub-models were 13.176, 14.543, 9.571 and 7.208; the mean absolute scaled errors (MASEs) were 0.469, 0.472, 0.296 and 0.266 and the mean absolute deviation (MAD) were 6.403, 7.658, 5.871 and 5.691. After using the stacking architecture combined with the above four models, the RMSE, MASE and MAD values of the ensemble model decreased to 6.810, 0.224 and 4.625, respectively. CONCLUSIONS: A novel ensemble model based on the robustness of structured prediction and model combination through stacking was developed. The findings suggest that the predictive performance of the final model is superior to that of the other four sub-models, indicating that stacking architecture may have significant implications in infectious disease prediction.

Cost-effectiveness analysis of comprehensive intervention programs to control blood glucose in overweight and obese type 2 diabetes mellitus patients based on a real-world setting: Markov modeling.

BACKGROUND: Blood glucose control management in overweight and obese diabetic patients poses heavy public health and economic burdens on the health system. This study aimed to evaluate the short-term cost-effectiveness of a comprehensive intervention program for blood glucose management in different groups using a Markov model. METHODS: Based on real-world data, a Markov model was developed to calculate the cost per quality-adjusted life-year (QALY) gained. The division of Markov states was in accordance with clinical practice. A three-month cycle length and a 5-year time horizon were applied. A 3% discounting rate was applied for both the costs and utilities. RESULTS: The incremental cost-effectiveness ratios (ICER) was more favorable for the male group than the female group, with an associated ICER of 104 K RMB per QALY gained. Compared with the younger group, the incremental gain of the middle-aged group was -0.062 QALY, and the incremental cost was -3,198.64 RMB; meanwhile, the incremental gain of the elderly group was -0.176 QALY, and the incremental cost was 4,485.746 RMB. The sensitivity analysis showed that the ICER is sensitive to the costs of this program and less sensitive to the discounting rate and the time horizon. CONCLUSIONS: The comprehensive intervention program for blood glucose management of overweight and obese patients with diabetes is cost-effective for the middle-aged male group and elderly female group, respectively. Moreover, the male group was more favorable than the female group if three times the gross domestic product (GDP) per capita was adopted as the maximum willingness to pay (WTP) for a QALY in China.

Asymmetric dimethylarginine predicts left atrial appendage thrombus in patients with non-valvular atrial fibrillation.

INTRODUCTION: Atrial fibrillation (AF) is associated with increased thrombo-embolic events, with thrombi most frequently located in the left atrial appendage (LAA). Asymmetric dimethylarginine (ADMA), a potent endogenous inhibitor of nitric oxide synthase, is elevated in subjects with AF. We investigated the relationship between ADMA and risk of LAA thrombus in patients with non-valvular AF. METHODS AND RESULTS: A total of 223 consecutive patients with non-valvular AF were enrolled (63 ± 10 years, 65% male). LAA thrombus was detected in 27 subjects by transesophageal echocardiography. Serum ADMA was significantly increased (2.5 ± 0.6 µmol/L vs 1.8 ± 0.4 µmol/L, p<0.001), while serum nitrite/nitrate was significantly reduced in patients with LAA thrombus (30.3 ± 7.1 µmol/L vs 48.4 ± 9.3 µmol/L, p<0.001). ADMA was positively correlated with age and left atrium diameter, and was negatively correlated with nitrite/nitrate and LAA peak flow velocity. The area under the receiver operating characteristic curve of ADMA predicting thrombus was 0.84. In multivariate logistic regression analysis, ADMA (OR 4.0, 95% CI 1.2-13.0; p=0.003) was one of independent risk factors for LAA thrombus. CONCLUSIONS: Our study suggested that high ADMA was independently associated with the presence of LAA thrombus in patients with non-valvular AF.