Development and evaluation of a simple and effective prediction approach for identifying those at high risk of dyslipidemia in rural adult residents

Novel anthropometric indices for predicting type 2 diabetes mellitus

BACKGROUND

This study aimed to compare anthropometric indices to predict type 2 diabetes mellitus (T2DM) among first-degree relatives of diabetic patients in the Iranian community.

METHODS

In this study, information on 3483 first-degree relatives (FDRs) of diabetic patients was extracted from the database of the Endocrinology and Metabolism Research Center of Isfahan University of Medical Sciences. Overall, 2082 FDRs were included in the analyses. A logistic regression model was used to evaluate the association between anthropometric indices and the odds of having diabetes. Furthermore, a receiver operating characteristic (ROC) curve was applied to estimate the optimal cutoff point based on the sensitivity and specificity of each index. In addition, the indices were compared based on the area under the curve (AUC).

RESULTS

The overall prevalence of diabetes was 15.3%. The optimal cutoff points for anthropometric measures among men were 25.09 for body mass index (BMI) (AUC = 0.573), 0.52 for waist-to-height ratio (WHtR) (AUC = 0.648), 0.91 for waist-to-hip ratio (WHR) (AUC = 0.654), 0.08 for a body shape index (ABSI) (AUC = 0.599), 3.92 for body roundness index (BRI) (AUC = 0.648), 27.27 for body adiposity index (BAI) (AUC = 0.590), and 8 for visceral adiposity index (VAI) (AUC = 0.596). The optimal cutoff points for anthropometric indices were 28.75 for BMI (AUC = 0.610), 0.55 for the WHtR (AUC = 0.685), 0.80 for the WHR (AUC = 0.687), 0.07 for the ABSI (AUC = 0.669), 4.34 for the BRI (AUC = 0.685), 39.95 for the BAI (AUC = 0.583), and 6.15 for the VAI (AUC = 0.658). The WHR, WHTR, and BRI were revealed to have fair AUC values and were relatively greater than the other indices for both men and women. Furthermore, in women, the ABSI and VAI also had fair AUCs. However, BMI and the BAI had the lowest AUC values among the indices in both sexes.

CONCLUSION

The WHtR, BRI, VAI, and WHR outperformed other anthropometric indices in predicting T2DM in first-degree relatives (FDRs) of diabetic patients. However, further investigations in different populations may need to be implemented to justify their widespread adoption in clinical practice.

Prevalence of dyslipidemias in the Middle East region: A systematic review & meta-analysis study

BACKGROUND AND AIMS

Dyslipidemia is a known main risk factor for cardiovascular diseases, and it can be controlled to reduce the incidence of cardiovascular diseases. This meta-analysis aimed to estimate the prevalence of dyslipidemias in the Middle East.

METHODS

The relevant published articles between 2000 and 2021 that investigated the prevalence of dyslipidaemias in the Middle East were found through international data sources such as Medline, PubMed, and Google Scholar. The random-effects model was used to estimate the pooled prevalence with 95% confidence intervals.

RESULTS

The pooled prevalence of dyslipidemias, hypertriglyceridemia, hypercholesterolemia, high levels of low-density lipoprotein cholesterol and low levels of high-density lipoprotein cholesterol in the Middle East were 54.08% (95% CI: 43.83-66.71), 32.51% (95% CI: 28.59-36.43), 29.44% (95% CI: 18.74-40.13), 32.09% (95% CI: 22.17-42.01), 44.71% (95% CI: 37.86-51.57), respectively. During the last two decades, an increasing trend in the prevalence of dyslipidemias was observed overall and in both sexes. Also, the age groups over 30 significantly had the highest prevalence of hypercholesterolemia, high levels of low-density lipoprotein cholesterol, and low levels of high-density lipoprotein cholesterol (p < 0.05).

CONCLUSIONS

The increasing trend in the prevalence of dyslipidemias during the last two decades is an alarming and significant concern in the Middle East. Therefore, special measures are needed to deal with dyslipidemias as a health priority in the Middle East.

Unconventional Cardiovascular Risk Factors and Systematic Coronary Risk Estimation (SCORE) in the Lebanese Rural Population: The Forgotten Factors

Purpose

To evaluate the correlation between unconventional risk factors and the Systematic Coronary Risk Estimation (SCORE), and estimate the prevalence of conventional and unconventional cardiovascular (CV) risk factors in the rural Lebanese population in order to assess their CV risk.

Methods

This is a retrospective descriptive study conducted between November 2017 and June 2019 among the Lebanese rural population. The risk factors were analyzed from the files of the patients who presented for the CV disease screening days organized by a non governmental organization. The CV risk estimation tool is the SCORE. The classification of socio-economic level ranges from zero (low level) to 3 (high level).

Results

A total of 433 patients were included. The prevalence of hypertension, diabetes, dyslipidemia, smoking, and metabolic syndrome was 45.1%, 31.2%, 39.2%, 50% and 42.9% respectively. Only 13.6% of hypertensive patients and 6.7% of diabetics were controlled. A total of 0 or 1 point for the classification of socio-economic status was found in 62.6% of cases. A family history of CV diseases was present in 87.3% of participants. The SCORE was correlated with diabetes and metabolic syndrome (p = 0.000), without being correlated to socio-economic status (HR = -0.104; p = 0.059) or to family history (p = 0.834).

Conclusion

The socio-economic status and the family history of CV disease must be evaluated in addition to the classical risk calculation of the SCORE to better pinpoint the actual risk of the targeted population. The risk factors are prevalent but poorly controlled, hence the need for a national effort to ensure better care for the rural Lebanese population.

Prediction of hypercholesterolemia using machine learning techniques

Purpose

Hypercholesterolemia is a major risk factor for a wide range of cardiovascular diseases. Developing countries are more susceptible to hypercholesterolemia and its complications due to the increasing prevalence and the lack of adequate resources for conducting screening and/or prevention programs. Using machine learning techniques to identify factors contributing to hypercholesterolemia and developing predictive models can help early detection of hypercholesterolemia, especially in developing countries.

Methods

Data from the nationwide 2016 STEPs study in Iran were used to identify socioeconomic, lifestyle, and metabolic risk factors associated with hypercholesterolemia. Furthermore, the predictive power of the identified risk factors was assessed using five commonly used machine learning algorithms (random forest; gradient boosting; support vector machine; logistic regression; artificial neural network) and 10-fold cross validation in terms of specificity, sensitivity, and the area under the receiver operating characteristic curve.

Results

A total of 14,667 individuals were included in this study, of those 12.8% (n = 1879) had (undiagnosed) hypercholesterolemia. Based on multivariate logistic regression analysis the five most important risk factors for hypercholesterolemia were: older age (for the elderly group: OR = 2.243; for the middle-aged group: OR = 1.869), obesity-related factors including high BMI status (morbidly obese: OR = 1.884; obese: OR = 1.499; overweight: OR = 1.426) and AO (OR = 1.339), raised BP (hypertension: OR = 1.729; prehypertension: OR = 1.577), consuming fish once or twice per week (OR = 1.261), and having risky diet (OR = 1.163). Furthermore, all the five hypercholesterolemia prediction models achieved AUC around 0.62, and models based on random forest (AUC = 0.6282; specificity = 65.14%; sensitivity = 60.51%) and gradient boosting (AUC = 0.6263; specificity = 64.11%; sensitivity = 61.15%) had the optimal performance.

Conclusion

The study shows that socioeconomic inequalities, unhealthy lifestyle, and metabolic syndrome (including obesity and hypertension) are significant predictors of hypercholesterolemia. Therefore controlling these factors is necessary to reduce the burden of hypercholesterolemia. Furthermore, machine learning algorithms such as random forest and gradient boosting can be employed for hypercholesterolemia screening and its timely diagnosis. Applying deep learning algorithms as well as techniques for handling the class overlap problem seems necessary to improve the performance of the models.

Genetic factors increase the identification efficiency of predictive models for dyslipidaemia: a prospective cohort study

Background

Few studies have developed risk models for dyslipidaemia, especially for rural populations. Furthermore, the performance of genetic factors in predicting dyslipidaemia has not been explored. The purpose of this study is to develop and evaluate prediction models with and without genetic factors for dyslipidaemia in rural populations.

Methods

A total of 3596 individuals from the Henan Rural Cohort Study were included in this study. According to the ratio of 7:3, all individuals were divided into a training set and a testing set. The conventional models and conventional+GRS (genetic risk score) models were developed with Cox regression, artificial neural network (ANN), random forest (RF), and gradient boosting machine (GBM) classifiers in the training set. The area under the receiver operating characteristic curve (AUC), net reclassification index (NRI), and integrated discrimination index (IDI) were used to assess the discrimination ability of the models, and the calibration curve was used to show calibration ability in the testing set.

Results

Compared to the lowest quartile of GRS, the hazard ratio (HR) (95% confidence interval (CI)) of individuals in the highest quartile of GRS was 1.23(1.07, 1.41) in the total population. Age, family history of diabetes, physical activity, body mass index (BMI), triglycerides (TGs), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) were used to develop the conventional models, and the AUCs of the Cox, ANN, RF, and GBM classifiers were 0.702(0.673, 0.729), 0.736(0.708, 0.762), 0.787 (0.762, 0.811), and 0.816(0.792, 0.839), respectively. After adding GRS, the AUCs increased by 0.005, 0.018, 0.023, and 0.015 with the Cox, ANN, RF, and GBM classifiers, respectively. The corresponding NRI and IDI were 25.6, 7.8, 14.1, and 18.1% and 2.3, 1.0, 2.5, and 1.8%, respectively.

Conclusion

Genetic factors could improve the predictive ability of the dyslipidaemia risk model, suggesting that genetic information could be provided as a potential predictor to screen for clinical dyslipidaemia.

Trial registration

The Henan Rural Cohort Study has been registered at the Chinese Clinical Trial Register. (Trial registration: ChiCTR-OOC-15006699. Registered 6 July 2015 - Retrospectively registered).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12944-021-01439-3.

Develop and Evaluate a New and Effective Approach for Predicting Dyslipidemia in Steel Workers

The convolutional neural network (CNN) has made certain progress in image processing, language processing, medical information processing and other aspects, and there are few relevant researches on its application in disease risk prediction. Dyslipidemia is a major and modifiable risk factor for cardiovascular disease, early detection of dyslipidemia and early intervention can effectively reduce the occurrence of cardiovascular diseases. Risk prediction model can effectively identify high-risk groups and is widely used in public health and clinical medicine. Steel workers are a special occupational group. Their particular occupational hazards, such as high temperatures, noise and shift work, make them more susceptible to disease than the general population, which makes the risk prediction model for the general population no longer applicable to steel workers. Therefore, it is necessary to establish a new model dedicated to the prediction of dyslipidemia of steel workers. In this study, the physical examination information of thousands of steel workers was collected, and the risk factors of dyslipidemia in steel workers were screened out. Then, based on the data characteristics, the corresponding parameters were set for the convolutional neural network model, and the risk of dyslipidemia in steel workers was predicted by using convolutional neural network. Finally, the predictive performance of the convolutional neural network model is compared with the existing predictive models of dyslipidemia, logistics regression model and BP neural network model. The results show that the convolutional neural network has a good predictive performance in the risk prediction of dyslipidemia of steel workers, and is superior to the Logistic regression model and BP neural network model.

Development of a screening method for determining sodium intake based on the Dietary Reference Intakes for Japanese, 2020: A cross-sectional analysis of the National Health and Nutrition Survey, Japan

Background

Although assessing nutrient intake through dietary surveys is desirable, it can be effort- and time-intensive. We aimed to develop a brief screening method for determining sodium intake in order to raise public awareness regarding the Dietary Reference Intakes for Japanese (DRI-J) 2020.

Methods

Using data from the 2015 National Health and Nutrition Survey, we compared dietary behaviours obtained from a self-administered questionnaire according to sodium intake, which was assessed from one-day dietary records by a semi-weighed method. Participants were divided into 4 groups based on the reference values of sodium (salt equivalent) shown in the DRI-J. We also randomly divided the participants into development and validation groups, and used logistic regression analysis to identify predictive factors for sex-specific DRI-J (<7.5 g/day in men and <6.5 g/day in women) and above-average intakes (≥10 g/day in men and women).

Results

Among the 6,172 Japanese individuals aged ≥20 years old, participants with lower sodium intake were found to use nutrition labels and had a lower frequency of eating out than those with higher intakes (P for difference < .001). Our final model for predicting sodium intake included adjusted sex, age, dietary behaviours, and consumption of mainly processed foods. In the development group, areas under the receiver operating characteristics curves were 0.747 and 0.741 for adherence to sex-specific DRI-J and above-average intake, respectively. The corresponding values in the validation group were 0.734 and 0.730, respectively.

Conclusions

This method could easily identify sodium intake using dietary behaviours and specific food consumption, and is expected to be widely useful for health and nutrition education in Japan.

Modeling the risk factors for dyslipidemia and blood lipid indices: Ravansar cohort study

BACKGROUND

Lipid disorder is one of the most important risk factors for chronic diseases. Identifying the factors affecting the development of lipid disorders helps reduce chronic diseases, especially Chronic Heart Disease (CHD). The aim of this study was to model the risk factors for dyslipidemia and blood lipid indices.

METHODS

This study was conducted based on the data collected in the initial phase of Ravansar cohort study (2014-16). At the beginning, all the 453 available variables were examined in 33 stages of sensitivity analysis by perceptron Artificial Neural Network (ANN) data mining model. In each stage, the variables that were more important in the diagnosis of dyslipidemia were identified. The relationship among the variables was investigated using stepwise regression. The data obtained were analyzed in SPSS software version 25, at 0.05 level of significance.

RESULTS

Forty percent of the subjects were diagnosed with lipid disorder. ANN identified 12 predictor variables for dyslipidemia related to nutrition and physical status. Alkaline phosphatase, Fat Free Mass (FFM) index, and Hemoglobin (HGB) had a significant relationship with all the seven blood lipid markers. The Waist Hip Ratio was the most effective variable that showed a stronger correlation with cholesterol and Low-Density Lipid (LDL). The FFM index had the greatest effect on triglyceride, High-Density Lipid (HDL), cholesterol/HDL, triglyceride/HDL, and LDL/HDL. The greatest coefficients of determination pertained to the triglyceride/HDL (0.203) and cholesterol/HDL (0.188) model with nine variables and the LDL/HDL (0.180) model with eight variables.

CONCLUSION

According to the results, alkaline phosphatase, FFM index, and HGB were three common predictor variables for all the blood lipid markers. Specialists should focus on controlling these factors in order to gain greater control over blood lipid markers.

Risk Prediction of Dyslipidemia for Chinese Han Adults Using Random Forest Survival Model

OBJECTIVE

Dyslipidemia has been recognized as a major risk factor of several diseases, and early prevention and management of dyslipidemia is effective in the primary prevention of cardiovascular events. The present study aims to develop risk models for predicting dyslipidemia using Random Survival Forest (RSF), which take the complex relationship between the variables into account.

METHODS

We used data from 6328 participants aged between 19 and 90 years free of dyslipidemia at baseline with a maximum follow-up of 5 years. RSF was applied to develop gender-specific risk model for predicting dyslipidemia using variables from anthropometric and laboratory test in the cohort. Cox regression was also adopted in comparison with the RSF model, and Harrell's concordance statistic with 10-fold cross-validation was used to validate the models.

RESULTS

The incidence density of dyslipidemia was 101/1000 in total and subgroup incidence densities were 121/1000 for men and 69/1000 for women. Twenty-four predictors were identified in the prediction model of males and 23 in females. The C-statistics of the prediction models for males and females were 0.731 and 0.801, respectively. The RSF model shows better discriminative performance than CPH model (0.719 for males and 0.787 for females). Moreover, some predictors were observed to have a nonlinear effect on dyslipidemia.

CONCLUSION

The RSF model is a promising method in identifying high-risk individuals for the prevention of dyslipidemia and related diseases.

Application of a decision tree to establish factors associated with a nomogram of aortic stiffness

Aortic stiffness is a marker of vascular aging and may reflect occurrence of cardiovascular (CV) diseases. Aortic pulse wave velocity (PWV), a marker of aortic stiffness, can be measured by applanation tonometry. A nomogram of aortic stiffness was evaluated by the calculation of PWV index. Theoretical PWV can be calculated according to age, gender, mean blood pressure, and heart rate, allowing to form an individual PWV index [(measured PWV - theoretical PWV)/theoretical PWV]. The purpose of the present cross-sectional study was to investigate the determinants of the PWV index, by applying a decision tree. A cross-sectional study was conducted from 2012 to 2017, and 597 individuals were included. A training decision tree was constructed based on seventy percent of these subjects (N = 428). The remaining 30% (N = 169) were used as the testing dataset to evaluate the performance of the decision trees. The input variables for the models were clinical and biochemical parameters. The different input variables remained in the model were diabetes, tobacco status, carotid plaque, albuminuria, C-reactive protein, total cholesterol, BMI, and previous CV diseases. For the validation decision model, the sensitivity, specificity, and accuracy values for identifying the related risk factors of PWV index were 70%, 78%, and 0.73. Since determinants of PWV index were all well-accepted CV risk factors, a nomogram of aortic stiffness could be considered as an integrator of CV risk factors on their duration of exposure and could be utilized to develop future programs for CV risk assessment and reduction strategies.

Determinants of the magnitude of response to vitamin D supplementation in adolescent girls identified using a decision tree algorithm

Vitamin D (VitD) supplementation is an inexpensive and effective approach for improving VitD insufficiency/deficiency. However, the response to supplementation, with respect to the increase in serum 25(OH)D level varies between individuals. In this study, we have assessed the factors associated with the response to VitD supplementation using a decision-tree algorithm. Serum VitD levels, pre- and post-VitD supplementation was used as the determinant of responsiveness. The model was validated by constructing a receiver operating characteristic curve. Serum VitD at baseline levels was at the apex of the tree in our model, followed by serum low-density lipoprotein cholesterol and triglyceride, age, waist-hip ratio, and high-density lipoprotein cholesterol. Our model suggests that these determinants of responsiveness to VitD supplementation had sensitivity, specificity, and accuracy, 59.4, 75.8 and 69.3%, respectively. The decision tree model appears to be a relatively accurate, specific, and sensitive approach for identifying the factors associated with response to VitD supplementation.

Metabolic-related markers and inflammatory factors as predictors of dyslipidemia among urban Han Chinese adults

Background

Metabolic-related markers and inflammatory factors have been proved to be associated with increased risk of dyslipidemia. Elucidating the mechanisms underlying these associations might provide an important perspective for the prevention of dyslipidemia. In the present study, we aimed to explore the effect of metabolic-related markers on dyslipidemia, and to assess what extent inflammation mediating these associations.

Methods

A total of 25,130 participants without dyslipidemia at baseline were included in the present study during 2010–2015. A partial least squares path model was used to explore possible pathways from metabolic-related markers to dyslipidemia, and the mediation role of inflammation.

Results

Lipid metabolism factor, blood pressure factor, obesity condition factor, glucose metabolism factor, renal function factor and lifestyle factor had diverse impact on development of dyslipidemia, directly and (or) indirectly. Partial least squares path analysis revealed that the determination coefficient of the model (R²) was 0.52. Lipid metabolism factor, obesity condition factor, and glucose metabolism factor had both direct and indirect effect on dyslipidemia through inflammatory factor. Lipid metabolism factor was the most important risk factor (β = 0.68) in the prediction of dyslipidemia, followed by obesity condition factor (β = 0.06) and glucose metabolism factor (β = 0.03).

Conclusions

Metabolic-related markers are strong risk factors for dyslipidemia. Inflammatory factors have significant mediating effect on these relationships. These findings suggested that comprehensive intervention strategies on metabolic biomarkers and inflammatory factors should be taken into consideration in prevention and treatment of dyslipidemia.

Comparative Study of Back Propagation Artificial Neural Networks and Logistic Regression Model in Predicting Poor Prognosis after Acute Ischemic Stroke

Objective

To investigate the predictive value of clinical variables on the poor prognosis at 90-day follow-up from acute stroke onset, and compare the diagnostic performance between back propagation artificial neural networks (BP ANNs) and Logistic regression (LR) models in predicting the prognosis.

Methods

We studied the association between clinical variables and the functional recovery of 435 acute ischemic stroke patients. The patients were divided into 2 groups according to modified Rankin Scale scores evaluated on the 90th day after stroke onset. Both BP ANNs and LR models were established for predicting the poor outcome and their diagnostic performance were compared by receiver operating curve.

Results

Age, free fatty acid, homocysteine and alkaline phosphatase were closely related with the poor outcome in acute ischemic stroke patients and finally enrolled in models. The accuracy, sensitivity and specificity of BP ANNs were 80.15%, 75.64% and 82.07% respectively. For the LR model, the accuracy, sensitivity and specificity was 70.61%, 88.46% and 63.04% respectively. The area under the ROC curve of the BP ANNs and LR model was 0.881and 0.809.

Conclusions

Both BP ANNs and LR model were promising for the prediction of poor outcome by combining age, free fatty acid, homocysteine and alkaline phosphatase. However, BP ANNs model showed better performance than LR model in predicting the prognosis.

Associations between lipid profiles of adolescents and their mothers based on a nationwide health and nutrition survey in South Korea

OBJECTIVES

Dyslipidaemia is a metabolic disease influenced by environmental and genetic factors. Especially, family history related to genetic background is a strong risk factor of lipid abnormality. The aim of this study is to evaluate the association between the lipid profiles of adolescents and their mothers.

DESIGN

A cross-sectional study.

SETTING

The data were derived from the Korea National Health and Nutrition Examination Survey (IV-VI) between 2009 and 2015.

PARTICIPANTS

2884 adolescents aged 12-18 years and their mothers were included.

PRIMARY OUTCOME MEASURES

Outcome variables were adolescents' lipid levels. Mothers' lipid levels were the interesting variables. The lipid profiles included total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C). We identified partial correlation coefficients (r) between the lipids. Multiple linear regressions were performed to identify the amount of change in adolescents' lipid levels for each unit increase of their mothers' lipids. The regression models included various clinical characteristics and health behavioural factors of both adolescents and mothers.

RESULTS

The mean levels of adolescents' lipids were 156.6, 83.6, 50.4 and 89.4 mg/dL, respectively for TC, TG, HDL-C and LDL-C. Positive correlations between lipid levels of adolescents and mothers were observed for TC, TG, HDL-C and LDL-C (r, 95% CI: 0.271, 0.236 to 0.304; 0.204, 0.169 to 0.239; 0.289, 0.255 to 0.322; and 0.286, 0.252 to 0.319). The adolescent TC level was increased by 0.23 mg/dL for each unit increase of the mother's TC (SE, 0.02; p<0.001). The beta coefficients were 0.16 (SE, 0.01), 0.24 (SE, 0.02) and 0.24 (SE, 0.02), respectively, in each model of TG, HDL-C and LDL-C (all p<0.001). The linear relationships were significant regardless of sex and mother's characteristics.

CONCLUSIONS

Mothers' lipid levels are associated with adolescents' lipids; therefore, they can serve as a reference for the screening of adolescent's dyslipidaemia.

Application of Latent Class Analysis to Identify Metabolic Syndrome Components Patterns in adults: Tehran Lipid and Glucose study

In this study, using latent class analysis (LCA), we investigated whether there are any homogeneous subclasses of individuals exhibiting different profiles of metabolic syndrome (MetS) components. The current study was conducted within the framework of the Tehran Lipid and Glucose Study (TLGS), a population-based cohort including 6448 subjects, aged 20-50 years. We carried out a LCA on MetS components and assessed the association of some demographic and behavioral variables with membership of latent subclasses using multinomial logistic regression. Four latent classes were identified:(1) Low riskclass, with the lowest probabilities for all MetS components (its prevalence rate in men: 29%, women: 64.7%), (2) MetS with diabetes medication (men: 1%, women: 2.3%), (3) Mets without diabetes medication (men: 32%, women: 13.4%), (4) dyslipidemia (men: 38%, women: 19.6%). In men the forth subclass was more significantly associated with being smoker (odds ratio: 4.49; 95% CI: 1.89-9.97). Our study showed that subjects with MetS could be classified in sub-classes with different origins for their metabolic disorders including drug treated diabetes, those with central obesity and dyslipidemia associated with smoking.

Risk prediction model of dyslipidaemia over a 5-year period based on the Taiwan MJ health check-up longitudinal database

OBJECTIVE

This study aimed to provide an epidemiological model to evaluate the risk of developing dyslipidaemia within 5 years in the Taiwanese population.

METHODS

A cohort of 11,345 subjects aged 35-74 years and was non-dyslipidaemia in the initial year 1996 and followed in 1997-2006 to derive a risk score that could predict the occurrence of dyslipidaemia. Multivariate logistic regression was used to derive the risk functions using the check-up centre of the overall cohort. Rules based on these risk functions were evaluated in the remaining three centres as the testing cohort. We evaluated the predictability of the model using the area under the receiver operating characteristic (ROC) curve (AUC) to confirm its diagnostic property on the testing sample. We also established the degrees of risk based on the cut-off points of these probabilities after transforming them into a normal distribution by log transformation.

RESULTS

The incidence of dyslipidaemia over the 5-year period was 19.1%. The final multivariable logistic regression model includes the following six risk factors: gender, history of diabetes, triglyceride level, HDL-C (high-density lipoprotein cholesterol), LDL-C (low-density lipoprotein cholesterol) and BMI (body mass index). The ROC AUC was 0.709 (95% CI: 0.693-0.725), which could predict the development of dyslipidaemia within 5 years.

CONCLUSION

This model can help individuals assess the risk of dyslipidaemia and guide group surveillance in the community.

Identification of an obesity index for predicting metabolic syndrome by gender: the rural Chinese cohort study

BACKGROUND

To compare the accuracy of different obesity indexes, including waist circumference (WC), weight-to-height ratio (WHtR), body mass index (BMI), and lipid accumulation product (LAP), in predicting metabolic syndrome (MetS) and to estimate the optimal cutoffs of these indexes in a rural Chinese adult population.

METHODS

This prospective cohort involved 8468 participants who were followed up for 6 years. MetS was defined by the International Diabetes Federation, American Heart Association, and National Heart, Lung, and Blood Institute criteria. The power of the 4 indexes for predicting MetS was estimated by receiver operating characteristic (ROC) curve analysis and optimal cutoffs were determined by the maximum of Youden's index.

RESULTS

As compared with WHtR, BMI, and LAP, WC had the largest area under the ROC curve (AUC) for predicting MetS after adjusting for age, smoking, drinking, physical activity, and education level. The AUCs (95% CIs) for WC, WHtR, BMI, and LAP for men and women were 0.862 (0.851-0.873) and 0.806 (0.794-0.817), 0.832 (0.820-0.843) and 0.789 (0.777-0.801), 0.824 (0.812-0.835) and 0.790 (0.778-0.802), and 0.798 (0.785-0.810) and 0.771 (0.759-0.784), respectively. The optimal cutoffs of WC for men and women were 83.30 and 76.80 cm. Those of WHtR, BMI, and LAP were approximately 0.51 and 0.50, 23.90 and 23.00 kg/m2, and 19.23 and 20.48 cm.mmol/L, respectively.

CONCLUSIONS

WC as a preferred index over WHtR, BMI, and LAP for predicting MetS in rural Chinese adults of both genders; the optimal cutoffs for men and women were 83.30 and 76.80 cm.

A Comparison between Decision Tree and Random Forest in Determining the Risk Factors Associated with Type 2 Diabetes

Background: We aimed to identify the associated risk factors of type 2 diabetes mellitus (T2DM) using data mining approach, decision tree and random forest techniques using the Mashhad Stroke and Heart Atherosclerotic Disorders (MASHAD) Study program.

Study design: A cross-sectional study.

Methods: The MASHAD study started in 2010 and will continue until 2020. Two data mining tools, namely decision trees, and random forests, are used for predicting T2DM when some other characteristics are observed on 9528 subjects recruited from MASHAD database. This paper makes a comparison between these two models in terms of accuracy, sensitivity, specificity and the area under ROC curve.

Results: The prevalence rate of T2DM was 14% among these subjects. The decision tree model has 64.9% accuracy, 64.5% sensitivity, 66.8% specificity, and area under the ROC curve measuring 68.6%, while the random forest model has 71.1% accuracy, 71.3% sensitivity, 69.9% specificity, and area under the ROC curve measuring 77.3% respectively.

Conclusions: The random forest model, when used with demographic, clinical, and anthropometric and biochemical measurements, can provide a simple tool to identify associated risk factors for type 2 diabetes. Such identification can substantially use for managing the health policy to reduce the number of subjects with T2DM .

Prediction of dyslipidemia using gene mutations, family history of diseases and anthropometric indicators in children and adolescents: The CASPIAN-III study

Dyslipidemia, the disorder of lipoprotein metabolism resulting in high lipid profile, is an important modifiable risk factor for coronary heart diseases. It is associated with more than four million worldwide deaths per year. Half of the children with dyslipidemia have hyperlipidemia during adulthood, and its prediction and screening are thus critical. We designed a new dyslipidemia diagnosis system. The sample size of 725 subjects (age 14.66 ± 2.61 years; 48% male; dyslipidemia prevalence of 42%) was selected by multistage random cluster sampling in Iran. Single nucleotide polymorphisms (rs1801177, rs708272, rs320, rs328, rs2066718, rs2230808, rs5880, rs5128, rs2893157, rs662799, and Apolipoprotein-E2/E3/E4), and anthropometric, life-style attributes, and family history of diseases were analyzed. A framework for classifying mixed-type data in imbalanced datasets was proposed. It included internal feature mapping and selection, re-sampling, optimized group method of data handling using convex and stochastic optimizations, a new cost function for imbalanced data and an internal validation. Its performance was assessed using hold-out and 4-foldcross-validation. Four other classifiers namely as supported vector machines, decision tree, and multilayer perceptron neural network and multiple logistic regression were also used. The average sensitivity, specificity, precision and accuracy of the proposed system were 93%, 94%, 94% and 92%, respectively in cross validation. It significantly outperformed the other classifiers and also showed excellent agreement and high correlation with the gold standard. A non-invasive economical version of the algorithm was also implemented suitable for low- and middle-income countries. It is thus a promising new tool for the prediction of dyslipidemia.

Influence of polygenic risk scores on lipid levels and dyslipidemia in a psychiatric population receiving weight gain-inducing psychotropic drugs

OBJECTIVES

Dyslipidemia represents a major health issue in psychiatry. We determined whether weighted polygenic risk scores (wPRSs) combining multiple single-nucleotide polymorphisms (SNPs) associated with lipid levels in the general population are associated with lipid levels [high-density lipoprotein (HDL), low-density lipoprotein (LDL), total cholesterol (TC), and triglycerides] and/or dyslipidemia in patients receiving weight gain-inducing psychotropic drugs. We also determined whether genetics improve the predictive power of dyslipidemia.

PATIENTS AND METHODS

The influence of wPRS on lipid levels was firstly assessed in a discovery psychiatric sample (n=332) and was then tested for replication in an independent psychiatric sample (n=140). The contribution of genetic markers to predict dyslipidemia was evaluated in the combined psychiatric sample.

RESULTS

wPRSs were significantly associated with the four lipid traits in the discovery (P≤0.02) and in the replication sample (P≤0.03). Patients whose wPRS was higher than the median wPRS had significantly higher LDL, TC, and triglyceride levels (0.20, 0.32 and 0.26 mmol/l, respectively; P≤0.004) and significantly lower HDL levels (0.13 mmol/l; P<0.0001) compared with others. Adding wPRS to clinical data significantly improved dyslipidemia prediction of HDL (P=0.03) and a trend for improvement was observed for the prediction of TC dyslipidemia (P=0.08).

CONCLUSION

Population-based wPRSs have thus significant effects on lipid levels in the psychiatric population. As genetics improved the predictive power of dyslipidemia development, only 24 patients need to be genotyped to prevent the development of one case of HDL hypocholesterolemia. If confirmed by further prospective investigations, the present results could be used for individualizing psychotropic treatment.

The application of a decision tree to establish the parameters associated with hypertension

INTRODUCTION

Hypertension is an important risk factor for cardiovascular disease (CVD). The goal of this study was to establish the factors associated with hypertension by using a decision-tree algorithm as a supervised classification method of data mining.

METHODS

Data from a cross-sectional study were used in this study. A total of 9078 subjects who met the inclusion criteria were recruited. 70% of these subjects (6358 cases) were randomly allocated to the training dataset for the constructing of the decision-tree. The remaining 30% (2720 cases) were used as the testing dataset to evaluate the performance of decision-tree. Two models were evaluated in this study. In model I, age, gender, body mass index, marital status, level of education, occupation status, depression and anxiety status, physical activity level, smoking status, LDL, TG, TC, FBG, uric acid and hs-CRP were considered as input variables and in model II, age, gender, WBC, RBC, HGB, HCT MCV, MCH, PLT, RDW and PDW were considered as input variables. The validation of the model was assessed by constructing a receiver operating characteristic (ROC) curve.

RESULTS

The prevalence rates of hypertension were 32% in our population. For the decision-tree model I, the accuracy, sensitivity, specificity and area under the ROC curve (AUC) value for identifying the related risk factors of hypertension were 73%, 63%, 77% and 0.72, respectively. The corresponding values for model II were 70%, 61%, 74% and 0.68, respectively.

CONCLUSION

We have developed a decision tree model to identify the risk factors associated with hypertension that maybe used to develop programs for hypertension management.

Risk factors and quality of life of dyslipidemic patients in Lebanon: A cross-sectional study

The main objective of this study was to identify the risk factors of dyslipidemia and measure its impact on patients' quality of life (QOL). Secondary objectives were to determine the percentage of dyslipidemia and assess the predictive factors affecting patients' QOL. A cross-sectional study was conducted in a sample of Lebanese population. A standardized questionnaire was developed to assess the QOL using the Short form-36 (SF-36) score. A total of 452 individuals were interviewed, of which 59.5% were females. The mean age was 43.3±15.6years, and 24.8% had dyslipidemia. The results show a lower overall QOL score among dyslipidemic patients compared with controls (57.9% and 76.5%, respectively; p<0.001). Waterpipe smoking [adjusted odds ratio (OR_a)=4.113, 95% confidence interval (CI): 1.696-9.971, p=0.002], hypertension (OR_a=3.597, 95% CI: 1.818-7.116, p<0.001), diabetes (OR_a=3.441, 95% CI: 1.587-7.462, p=0.002), cigarette smoking (OR_a=2.966, 95% CI: 1.516-5.804, p=0.001), and passive smoking (OR_a=2.716, 95% CI: 1.376-5.358, p=0.004) were significantly associated with dyslipidemia in individuals older than 30years. A higher overall QOL score (p=0.013) was observed in patients treated with statins in comparison with other lipid-lowering medications. In addition to clinical and economical consequences, dyslipidemia may have a significant impact on patients' QOL. Further research is needed to confirm the impact of treatment on dyslipidemic patients' QOL in order to maximize the overall benefits of therapy.

Effect of Simvastatin on induced apical periodontitis in rats: a tomographic and biochemical analysis

Abstract Introduction Simvastatin is one of several statins that are used to treat hypercholesterolemia, and in dentistry, few studies have attempted to associate the administration of this compound with bone repair. Objective To evaluate the effect of simvastatin on the progression of induced apical periodontitis in rats. Material and method To this end, 36 male Wistar rats were divided into 3 groups (N=12): Induced Apical Periodontitis Associated with Simvastatin Group (APSG N=12), Induced Periodontitis Apical Induced Group (APG N=12) and Negative Control Group (CG). On the first day, APG and APSG were anesthetized, and the coronal opening of the mandibular first molar was performed. For thirty days, the APSG received 6 mg of simvastatin daily via gavage. On the thirty-first day, all groups underwent blood collection and euthanasia. The jaws were removed and fixed in formalin. CT scans were performed to measure the periapical regions. In addition, the body mass and lipid profile were analyzed. The data were subjected to statistical analysis (ANOVA and Tukey’s test). Result The APG (3.42±0.65) showed the highest perimeters for the space periapical ligament, followed by APSG (1.54±0.78) and CG (0.64±0.24) (p<0.05). The lipid profile revealed the effect of simvastatin on the amount of glucose, triglycerides, HDL, and VLDL (p<0.05). Body mass APG showed the most weight gain (264.75±44.11), followed by CG (252.00±44.36) and APSG (245.41±42.56). The three groups showed significant differences in decreasing order (p<0.05). Conclusion The use of simvastatin decreased the progression of the increasing periapical ligament space in rats.

Dyslipidemia and its risk factors among urban middle-aged Iranians: A population-based study

AIMS

Dyslipidemia is a known risk factor for cardiovascular disease and is a leading cause of mortality in developed and developing countries. This study was aimed to determine the prevalence of dyslipidemia and its risk factors in an urban group of Iranian adult population.

METHODS

In this study, based on the criteria set by the National Cholesterol Education Program, the prevalence of dyslipidemia was evaluated in a population of 4737 people aged 45-69 years who participated in the second phase of an ophthalmology cohort study in Shahroud. Dyslipidemia prevalence was determined by age, sex, and risk factors of the disease; the findings were tested by using simple and multiple logistic regression.

RESULTS

The prevalence of dyslipidemia was 66.5% (CI 95%: 64.4-68.6) in males, 61.3% (CI 95%: 59.5-63.2) in females, and 63.4% (CI 95%: 62.0-64.9%) in both sexes. The prevalence of hypertriglyceridemia, hypercholesterolemia, low HDL-C, and high LDL-C, respectively, was 28.8%, 13.4%, 42.3%, and 13.4%, respectively. In multivariate logistic regression model, increase of age (for females), abdominal obesity, overweight and obesity, hypertension, and diabetes were associated with an increased odd of dyslipidemia.

CONCLUSION

The prevalence of dyslipidemia in middle-aged urban population in Iran is high, and with increasing age there is an increased risk of dyslipidemia. Hence, considering the growing trend of aging in Iran, there is need for taking special measures to deal with dyslipidemia as a health priority. Furthermore, the need for planning in order to reduce the risk of dyslipidemia and prevent its complications is greater than ever.

Development and Validation of a Risk-Score Model for Type 2 Diabetes: A Cohort Study of a Rural Adult Chinese Population

Some global models to predict the risk of diabetes may not be applicable to local populations. We aimed to develop and validate a score to predict type 2 diabetes mellitus (T2DM) in a rural adult Chinese population. Data for a cohort of 12,849 participants were randomly divided into derivation (n = 11,564) and validation (n = 1285) datasets. A questionnaire interview and physical and blood biochemical examinations were performed at baseline (July to August 2007 and July to August 2008) and follow-up (July to August 2013 and July to October 2014). A Cox regression model was used to weigh each variable in the derivation dataset. For each significant variable, a score was calculated by multiplying β by 100 and rounding to the nearest integer. Age, body mass index, triglycerides and fasting plasma glucose (scores 3, 12, 24 and 76, respectively) were predictors of incident T2DM. The model accuracy was assessed by the area under the receiver operating characteristic curve (AUC), with optimal cut-off value 936. With the derivation dataset, sensitivity, specificity and AUC of the model were 66.7%, 74.0% and 0.768 (95% CI 0.760–0.776), respectively. With the validation dataset, the performance of the model was superior to the Chinese (simple), FINDRISC, Oman and IDRS models of T2DM risk but equivalent to the Framingham model, which is widely applicable in a variety of populations. Our model for predicting 6-year risk of T2DM could be used in a rural adult Chinese population.

Resting heart rate as a marker for identifying the risk of undiagnosed type 2 diabetes mellitus: a cross-sectional survey

Background

Fast resting heart rate might increase the risk of developing type 2 diabetes mellitus (T2DM). However, it is unclear whether resting heart rate could be used to predict the risk of undiagnosed T2DM. Therefore, the purposes of this study were to examine the association between resting heart rate and undiagnosed T2DM, and evaluate the feasibility of using resting heart rate as a marker for identifying the risk of undiagnosed T2DM.

Methods

A cross-sectional survey was conducted. Resting heart rate and relevant covariates were collected and measured. Fasting blood samples were obtained to measure blood glucose using the modified hexokinase enzymatic method. Predictive performance was analyzed by Receiver Operating Characteristic (ROC) curve.

Results

This study included 16, 636 subjects from rural communities aged 35–78 years. Resting heart rate was significantly associated with undiagnosed T2DM in both genders. For resting heart rate categories of <60, 60–69, 70–79, and ≥80 beats/min, adjusted odds ratios for undiagnosed T2DM were 1.04, 2.32, 3.66 and 1.05, 1.57, 2.98 in male and female subjects, respectively. For male subjects, resting heart rate ≥70 beats/min could predict undiagnosed T2DM with 76.56% sensitivity and 48.64% specificity. For female subjects, the optimum cut-off point was ≥79 beats/min with 49.72% sensitivity and 67.53% specificity. The area under the ROC curve for predicting undiagnosed T2DM was 0.65 (95% CI: 0.64-0.66) and 0.61(95% CI: 0.60-0.62) in male and female subjects, respectively.

Conclusions

Fast resting heart rate is associated with an increased risk of undiagnosed T2DM in male and female subjects. However, resting heart rate as a marker has limited potential for screening those at high risk of undiagnosed T2DM in adults living in rural areas.

Epidemiological survey of dyslipidemia in civil aviators in china from 2006 to 2011

Aim. This study aimed to analyze blood lipid levels, temporal trend, and age distribution of dyslipidemia in civil aviators in China. Methods. The 305 Chinese aviators were selected randomly and followed up from 2006 to 2011. Their total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) levels were evaluated annually. Mean values for each parameter by year were compared using a linear mixed-effects model. The temporal trend of borderline high, high, and low status for each index and of overall borderline high, hyperlipidemia, and dyslipidemia by year was tested using a generalized linear mixed model. Results. The aviators' TC (F = 4.33, P < 0.01), HDL-C (F = 23.25, P < 0.01), and LDL-C (F = 6.13, P < 0.01) values differed across years. The prevalence of dyslipidemia (F = 5.53, P < 0.01), borderline high (F = 6.52, P < 0.01), and hyperlipidemia (F = 3.90, P < 0.01) also differed across years. The prevalence rates for hyperlipidemia and dyslipidemia were the highest in the 41-50-year-old and 31-40-year-old groups. Conclusions. Civil aviators in China were in high dyslipidemia and borderline high level and presented with dyslipidemia younger than other Chinese populations.

Development of clinical decision rules to predict recurrent shock in dengue

INTRODUCTION

Mortality from dengue infection is mostly due to shock. Among dengue patients with shock, approximately 30% have recurrent shock that requires a treatment change. Here, we report development of a clinical rule for use during a patient's first shock episode to predict a recurrent shock episode.

METHODS

The study was conducted in Center for Preventive Medicine in Vinh Long province and the Children's Hospital No. 2 in Ho Chi Minh City, Vietnam. We included 444 dengue patients with shock, 126 of whom had recurrent shock (28%). Univariate and multivariate analyses and a preprocessing method were used to evaluate and select 14 clinical and laboratory signs recorded at shock onset. Five variables (admission day, purpura/ecchymosis, ascites/pleural effusion, blood platelet count and pulse pressure) were finally trained and validated by a 10-fold validation strategy with 10 times of repetition, using a logistic regression model.

RESULTS

The results showed that shorter admission day (fewer days prior to admission), purpura/ecchymosis, ascites/pleural effusion, low platelet count and narrow pulse pressure were independently associated with recurrent shock. Our logistic prediction model was capable of predicting recurrent shock when compared to the null method (P < 0.05) and was not outperformed by other prediction models. Our final scoring rule provided relatively good accuracy (AUC, 0.73; sensitivity and specificity, 68%). Score points derived from the logistic prediction model revealed identical accuracy with AUCs at 0.73. Using a cutoff value greater than -154.5, our simple scoring rule showed a sensitivity of 68.3% and a specificity of 68.2%.

CONCLUSIONS

Our simple clinical rule is not to replace clinical judgment, but to help clinicians predict recurrent shock during a patient's first dengue shock episode.

Artificial neural network analysis for evaluating cancer risk in multinodular goiter

BACKGROUND

The aim of this study was to create a diagnostic model using the artificial neural networks (ANNs) to predict malignancy in multinodular goiter patients with an indeterminate cytology.

MATERIALS AND METHODS

Out of 623 patients, 411 evaluated for multinodular goiter between July 2004 and March 2010 had a fine-needle aspiration biopsy. All patients underwent total thyroidectomy. The interpretation was consistent with an indeterminate lesion in 116 (18.6%) patients. Patient's medical records including age, sex, dominant nodule size, pre-operative serum thyroid-stimulating hormone level, thyroid hormone therapy and final pathologic diagnosis were collected retrospectively.

RESULTS

The mean age of the patients was 44.6 years (range, 17-78 years). About 104 (89.7%) were female and 12 (10.3%) were male patients. Final pathology revealed 24 malignant diseases (20.7%) and 92 (79.3%) benign diseases. After the completion of training, the ANN model was able to predict diagnosis of malignancy with a high degree of accuracy. The area under the curve of ANNs was 0.824.

CONCLUSION

The ANNs technique is a useful aid in diagnosing malignancy and may help reduce unnecessary thyroidectomies in multinodular goiter patients with an indeterminate cytology. Further studies are needed to construct the optimal diagnostic model and to apply it in the clinical practice.

Evaluating the risk of type 2 diabetes mellitus using artificial neural network: an effective classification approach

AIM

To develop and evaluate an effective classification approach without biochemical parameters to identify those at high risk of T2DM in rural adults.

METHODS

A cross-sectional survey was conducted. Of 8640 subjects who met inclusion criteria, 75% (N1=6480) were randomly selected to provide training set for constructing artificial neural network (ANN) and multivariate logistic regression (MLR) models. The remaining 25% (N2=2160) were assigned to validation set for performance comparisons of the ANN and MLR models. Predictive performance of different models was analyzed by the receiver operating characteristic (ROC) curve using the validation set.

RESULTS

The prevalence rates of T2DM were 8.66% (n=561) and 9.21% (n=199) in training and validation sets, respectively. For ANN model, the sensitivity, specificity, positive and negative predictive value for identifying T2DM were 86.93%, 79.14%, 31.86%, and 98.18%, respectively, while MLR model were only 60.80%, 75.48%, 21.78%, and 94.52%, respectively. Area under the ROC curve (AUC) value for identifying T2DM when using the ANN model was 0.891, showing more accurate predictive performance than the MLR model (AUC=0.744) (P=0.0001).

CONCLUSION

The ANN model is an effective classification approach for identifying those at high risk of T2DM based on demographic, lifestyle and anthropometric data.

Relationship between resting pulse rate and lipid metabolic dysfunctions in Chinese adults living in rural areas

Background

Resting pulse rate has been observed to be associated with cardiovascular diseases. However, its association with lipid metabolic dysfunctions remains unclear, especially resting pulse rate as an indicator for identifying the risk of lipid metabolic dysfunctions. The purpose of this study was to examine the association between resting pulse rate and lipid metabolic dysfunctions, and then evaluate the feasibility of resting pulse rate as an indicator for screening the risk of lipid metabolic dysfunctions.

Methods

A cross-sectional survey was performed, and 16,926 subjects were included in this study from rural community residents aged 35–78 years. Resting pulse rate and relevant covariates were collected from a standard questionnaire. The fasting blood samples were collected and measured for lipid profile. Predictive performance was analyzed by receiver operating characteristic (ROC) curve.

Results

A significant correlation was observed between resting pulse rate and TC (r = 0.102, P = 0.001), TG (r = 0.182, P = 0.001), and dyslipidemia (r = 0.037, P = 0.008). In the multivariate models, the adjusted odds ratios for hypercholesterolemia (from 1.07 to 1.15), hypertriglyceridemia (1.11 to 1.16), low HDL hypercholesterolemia (1.03 to 1.06), high LDL hypercholesterolemia (0.92 to 1.14), and dyslipidemia (1.04 to 1.07) were positively increased across quartiles of resting pulse rate (P for trend <0.05). The ROC curve indicated that resting pulse rate had low sensitivity (78.95%, 74.18%, 51.54%, 44.39%, and 54.22%), specificity (55.88%, 59.46%, 57.27%, 65.02%, and 60.56%), and the area under ROC curve (0.70, 0.69, 0.54, 0.56, and 0.58) for identifying the risk of hypercholesterolemia, hypertriglyceridemia, low HDL hypercholesterolemia, high LDL hypercholesterolemia, and dyslipidemia, respectively.

Conclusion

Fast resting pulse rate was associated with a moderate increased risk of lipid metabolic dysfunctions in rural adults. However, resting pulse rate as an indicator has limited potential for screening the risk of lipid metabolic dysfunctions.