Framingham Risk Score Model Research Articles

Steps to developing a DXA-based risk score for cardiovascular outcomes among older adults: the health, aging, and body composition study

Cardiovascular disease (CVD) is a major chronic disease worldwide and its risk factors have long been investigating in epidemiological studies. Our study aims to develop a body composition-based risk score and integrate it into the Framingham Risk Score (FRS) to improve CVD prediction among well-functioning older adults. We included 1882 older adults from the Health, Aging and Body Composition (Health ABC) study to screen body composition variables obtained from the Dual-energy X-ray absorptiometry (DXA). Three models were developed and compared: the 4-DXA model, the refit FRS, and the refit FRS plus 4-DXA model. C-statistics were 0.62 (95% CI: 0.59, 0.65) for the refit FRS, 0.58 (95% CI: 0.55, 0.61) for the 4-DXA model, and 0.63 (95% CI: 0.60, 0.66) for the refit FRS plus 4-DXA model. Compared to the refit FRS, the refit FRS plus 4-DXA model slightly improved CVD outcome prediction as the discrimination slope, net reclassification index, and the integrated discrimination index were 0.053 (95% CI: 0.041, 0.066), 0.098 (95% CI = – 0.0033, 0.20) and 0.013 (95% CI: 0.0069–0.019). This study provides a model for more accurate risk stratification and draws more attention on DXA-based indices in the clinical setting. It also encourages further research in validating the developed risk score in more diverse population and in investigating a broader range of CVD risk factors.

Mammographic breast density and cardiovascular disease in women. A literature review

The world is searching for new simple and economically available gender-specific markers to improve cardiovascular risk stratification in women. The aim of this review was to analyze the association of mammographic density (MD) with cardiovascular disease (CVD). In low MD, i.e., high relative mammary gland fat content, there is a higher incidence of the main risk factors for CVD: hypertension, hyperlipidemia, hyperglycemia, excess body weight, as well as an increase in the volume of fat depots, visceral and ectopic fat. Low MD is associated with a higher 10-year risk of adverse cardiovascular events such as coronary artery disease, stroke, peripheral arterial disease, revascularization, and heart failure, and may serve as a predictor of their development. Including MD in the Framingham Risk Score model improves its accuracy. Identification of low MD, as a marker of high cardiovascular risk, allows the use of mammography for early detection and prevention of the two most dangerous diseases among the female population — breast cancer and CVD.

Superior predictive value of estimated pulse wave velocity for all-cause and cardiovascular disease mortality risk in U.S. general adults

BackgroundEstimated pulse wave velocity (ePWV) has been proposed as a potential approach to estimate carotid-femoral pulse wave velocity. However, the potential of ePWV in predicting all-cause mortality (ACM) and cardiovascular disease mortality (CVM) in the general population is unclear.MethodsWe conducted a prospective cohort study using the data of 33,930 adults (age ≥ 20 years) from the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2014 until the end of December 2019. The study outcomes included ACM and CVM. Survey-weighted Cox proportional hazards models were used to assess hazard ratios (HRs) and 95% confidence intervals (CIs) to determine the association between ePWV and ACM and CVM. To further investigate whether ePWV was superior to traditional risk factors in predicting ACM and CVM, comparisons between ePWV and the Framingham Risk Score (FRS) and Pooled Cohort Equations (PCE) models were performed. Integrated Discriminant Improvement (IDI) and Net Reclassification Improvement (NRI) were employed to analyze differences in predictive ability between models.ResultsThe weighted mean age of the 33,930 adults included was 45.2 years, and 50.28% of all participants were men. In the fully adjusted Cox regression model, each 1 m/s increase in ePWV was associated with 50% and 49% increases in the risk of ACM (HR 1.50; 95% CI, 1.45–1.54) and CVM (HR 1.49; 95% CI, 1.41–1.57), respectively. After adjusting for FRS, each 1 m/s increase in ePWV was still associated with 29% (HR 1.29; 95% CI, 1.24–1.34) and 34% (HR 1.34; 95% CI, 1.23–1.45) increases in the risk of ACM and CVM, respectively. The area under the curve (AUC) predicted by ePWV for 10-year ACM and CVM were 0.822 and 0.835, respectively. Compared with the FRS model, the ePWV model improved the predictive value of ACM and CVM by 5.1% and 3.8%, respectively, with no further improvement in event classification. In comparison with the PCE model, the ePWV model’s ability to predict 10-year ACM and CVM was improved by 5.1% and 3.5%, and event classification improvement was improved by 34.5% and 37.4%.ConclusionsIn the U.S. adults, ePWV is an independent risk factor for ACM and CVM and is independent of traditional risk factors. In the general population aged 20 to 85 years, ePWV has a robust predictive value for the risk of ACM and CVM, superior to the FRS and PCE models. The predictive power of ePWV likely originates from the traditional risk factors incorporated into its calculation, rather than from an indirect association with measured pulse wave velocity.

Comparison of arterial stiffness indices measured by pulse wave velocity and pulse wave analysis for predicting cardiovascular and all-cause mortality in a Chinese population.

Arterial stiffness measured by pulse wave velocity and pulse wave analysis has been widely studied in different populations in terms of its correlation with cardiovascular events and all-cause mortality. It remains unknown which arterial stiffness index is better for risk stratification in the general population. We included 4129 participants from Gaoyou County, Jiangsu Province, China, with a median follow-up of 11 years. The primary endpoint was cardiovascular mortality, and the secondary endpoint was all-cause mortality. Harrell's C-index, net reclassification improvement (NRI) and integrated discrimination improvement (IDI) based on the Cox proportional hazards regression model were evaluated to assess predictive discrimination and accuracy. The associations between the 4 indices and cardiovascular mortality remained significant after adjusting for the Framingham Risk Score (FRS) and/or associated risk factors. Considering reclassification based on the newly integrated models (FRS model combined with the 4 indices), NRI for cardiovascular mortality showed that haPWV and baPWV had more significant improvement in reclassification compared with C1 and C2 [NRI with 95% CI: haPWV 0.410 (0.293, 0.523); baPWV 0.447 (0.330, 0.553); C1 0.312 (0.182, 0.454); C2 0.328 (0.159, 0.463); all P < 0.05]. This study showed that pulse wave velocity (haPWV and baPWV) provides better discrimination of long-term risk than arterial elasticity indices (C1 and C2) in the general population.

Serum lipoprotein(a) and reclassification of coronary heart disease risk; application of prediction in a cross-sectional analysis of an ongoing Iranian cohort

IntroductionRecent studies have introduced elevated lipoprotein(a) (Lp(a)) as a risk factor for coronary heart disease (CHD). This study investigated whether the addition of Lp(a) as a novel biomarker to the Framingham Risk Score (FRS) model improves CHD risk prediction.MethodsThe study included 1101 Iranian subjects (443 non-diabetic and 658 diabetic patients) who were followed for 10 years (2003–2013). Lp(a) levels and CHD events were recorded for each participant.ResultsThe Net Reclassification Index (NRI) after adding Lp(a) to the FRS model was 19.57% and the discrimination slope was improved (0.160 vs. 0.173). The Akaike Information Criterion (AIC), a measure of model complexity, decreased significantly after adding Lp(a) to the FRS model (691.9 vs. 685.4, P value: 0.007).ConclusionsThe study concluded that adding Lp(a) to the FRS model improves CHD risk prediction in an Iranian population without making the model too complex. This could help clinicians to better identify individuals who are at risk of developing CHD and to implement appropriate preventive measures.

Mammographic breast density and cardiovascular disease risk in women

AimsWe aimed to determine the predictive role of mammographic breast density in addition to the Framingham Risk Score (FRS) on subsequent CVD events in women. Methods and ResultsThis cohort study included 4,268,579 women aged ≥40 years who underwent mammography screening between 2009 and 2010 with follow-up until 2020. Breast density was reported following the Breast Imaging Reporting and Data System. Primary outcomes included coronary heart disease, cerebrovascular disease, peripheral arterial disease, and heart failure. The incremental predictive ability of breast density added to the FRS model was assessed using the ROC and net reclassification index (NRI) among all women and strata based on FRS risk categories (<5% as low-risk, 5%–10% as moderate-risk, and ≥10% as high-risk). In total, 135,475 CVD events were recorded after a median follow-up of 10.9 years. A lower category of breast density was associated with a higher risk of CVD. Compared to the extremely dense breast group, the hazard ratios (95% CI) for CVDs were 1.12 (1.09–1.14), 1.19 (1.17–1.22), and 1.29 (1.26–1.32) in women with heterogeneously dense, scattered fibroglandular densities, and almost entirely fat breast density, respectively. Adding breast density to the FRS showed a slight improvement in AUROC but a modest improvement in NRI; the C-statistic difference was 0.083% (95% CI 0.069–0.096) with a 7.15% (6.85–7.69) increase in NRI, with the strongest improvement observed in the low-risk group. ConclusionsMammographic breast density is an independent predictor of incident CVD among women. The addition of mammographic breast density to FRS improves the prediction of CVDs, especially in low-risk individuals.

The predictive accuracy of coronary heart disease risk prediction models in rural Northwestern China

Cardiovascular risk models developed may have limitations when applied to rural Chinese. This study validated and compared the Framingham Risk Score (FRS) and Prediction for Atherosclerotic Cardiovascular Disease Risk in China (PAR) models in predicting 10-year risk of coronary heart disease (CHD) in a rural cohort in Ningxia, China from 2008 to 2019. The FRS and PAR models were validated by estimating predicted events, C index, calibration χ2 and plots. 1381 adults without CHD at baseline were followed up for 9.75 years on average. 168 CHD cases were observed. The FRS and PAR underestimated CHD events by 22 % and 46 % for the total population, while overestimated for males by 152 % and 78 %, respectively. The C index was slightly higher for PAR than FRS. Both models showed weak calibration with chi-square values above 20 (p < 0.001). Bland-Altman plots indicated FRS predicted higher CHD risk than PAR, lacking consistency. Overall, FRS and PAR demonstrated limited performance in predicting 10-year CHD risk in this rural population. PAR had slightly better discrimination than FRS, but require further improvement in calibration and individual risk estimation to suit the rural population in Northwest China.

Agreement between ten-years cardiovascular disease risk assessment tools: An application to Iranian population in Shahedieh Cohort Study

Background and aimCardiovascular risk-prediction models are efficient primary prevention tools to detect high-risk individuals. The study aims to use three tools to estimate the 10-year risk of developing cardiovascular disease (CVD) and investigate their agreement in an Iranian adult population. MethodsThe current cross-sectional study was carried out on 8569 adults between 35 and 70 who participated in the first phase of the Shahedieh cohort study in Yazd, Iran, and were free of CVDs (cardiac ischemia or myocardial infarction or stroke). World Health Organization/International Society of Hypertension (WHO/ISH) chart, Laboratory-Based (LB) and Non-Laboratory-Based (NLB) Framingham Risk Score (FRS) were used to predict the 10-year risk of developing CVD. The agreement across tools was determined by Kappa. ResultsWHO/ISH chart indicated the highest prevalence of low CVD risk for males (96.10%) and females (96.50%), while NLB Framingham had the highest prevalence of high CVD risk for males (19.40%) and females (5.30%). In total, there was substantial agreement between both FRS models (Kappa = o.70), while there was a slight agreement between WHO/ISH and both FRS tools. For under 60 years males and females, substantial agreements were observed between FRS methods (kappa = 0.73 and kappa = 0.68). For males and females over 60 years, this agreement was moderate and substantial, respectively (kappa = 0.54 and kappa = 0.64). WHO/ISH and LB Framingham model had substantial agreement for over 60 years females (kappa = 0.61). ConclusionsFramingham models classified more participants in the high-risk category than WHO/ISH. Due to the lethality of CVDs, categorizing individuals based on FRS can ensure that most of the real high-risk people are detected. Remarkable agreement between FRS methods in all sex-age groups suggested using the NLB Framingham model as a primary screening tool, especially in a shortage of resources condition.

Can spirometry improve the performance of cardiovascular risk model in high-risk Eastern European countries?

Impaired lung function has been strongly associated with cardiovascular disease (CVD) events. We aimed to assess the additive prognostic value of spirometry indices to the risk estimation of CVD events in Eastern European populations in this study. We randomly selected 14,061 individuals with a mean age of 59 ± 7.3 years without a previous history of cardiovascular and pulmonary diseases from population registers in the Czechia, Poland, and Lithuania. Predictive values of standardised Z-scores of forced expiratory volume measured in 1 s (FEV1), forced vital capacity (FVC), and FEV1 divided by height cubed (FEV1/ht3) were tested. Cox proportional hazards models were used to estimate hazard ratios (HRs) of CVD events of various spirometry indices over the Framingham Risk Score (FRS) model. The model performance was evaluated using Harrell's C-statistics, likelihood ratio tests, and Bayesian information criterion. All spirometry indices had a strong linear relation with the incidence of CVD events (HR ranged from 1.10 to 1.12 between indices). The model stratified by FEV1/ht3 tertiles had a stronger link with CVD events than FEV1 and FVC. The risk of CVD event for the lowest vs. highest FEV1/ht3 tertile among people with low FRS was higher (HR: 2.35; 95% confidence interval: 1.96-2.81) than among those with high FRS. The addition of spirometry indices showed a small but statistically significant improvement of the FRS model. The addition of spirometry indices might improve the prediction of incident CVD events particularly in the low-risk group. FEV1/ht3 is a more sensitive predictor compared to other spirometry indices.

Association of fruit and vegetable intake with predicted 10-year cardiovascular disease risk among hypertensive patients in Addis Ababa, Ethiopia: a cross-sectional study

ObjectiveIn low-income countries, such as Ethiopia, few studies have evaluated the risk of cardiovascular disease (CVD) among hypertensive patients. We assessed the 10-year CVD risk of hypertensive patients.DesignThis cross-sectional study...

Relative contributions of arterial stiffness to cardiovascular disease risk score in Chinese women in framingham and China-PAR model.

Arterial stiffness played an important role in the development of cardiovascular disease (CVD) events. The aim of this study was to verify the relative importance of arterial stiffness for different CVD risk scores in a large sample of Chinese women. We measured arterial velocity pulse index (AVI) and CVD risk scores in 2220 female participants (mean age 57 years). Framingham Risk Score (FRS), and the prediction for Atherosclerotic Cardiovascular Disease Risk in China (China-PAR) were used to estimate CVD risk, respectively. The relationships between AVI and risk scores were investigated by linear regressions and restricted cubic spline (RCS) analysis. To determine the relative importance of AVI in predicting CVD risk scores, random forest analysis was used. There was a significant positive correlation between AVI and FRS, China-PAR in all subgroup groups stratified by age, blood pressure and BMI. AVI showed higher importance in predicting CVD risk scores in FRS model, compared with these traditional risk factors. In China-PAR model, although AVI was not as predictive as SBP, it had better predictive power than many known risk factors such as lipids. Furthermore, AVI had significant J-shaped associations both with FRS and China-PAR scores. AVI was significantly associated with CVD risk score. In FRS and China-PAR model, AVI showed relatively high importance in predicting CVD risk scores. These findings may support the use of arterial stiffness measurements in CVD risk assessment.

#4666 ASSOCIATION BETWEEN URINARY ALBUMIN CREATININE RATIO AND CARDIOVASCULAR DISEASE

Abstract Background and Aims The association between microalbuminuria and cardiovascular disease (CVD) is accumulating in various patient populations. However, when stratified by sex, the relationship between microalbuminuria and CVD remains unclear. Method We obtained data from the 2011–2014 and 2019–2020 Korea National Health and Nutrition Examination Survey (KNHANES). Microalbuminuria was measured based on spot urine albumin-creatinine ratio (UACR). The Framingham risk score (FRS) model was implemented to evaluate the CVD risk. Linear and logistic regression models were used to identify the associations of microalbuminuria status with cardiometabolic predictors and CVD status determined by the FRS score. Results Among 19,340 representative Korean participants, the (UACR) in Korean women and men with history of CVD was higher than in those without history of CVD. Among patients without history of CVD, multivariate regression analysis showed that a high UACR was related to older age, lower high-density lipoprotein cholesterol level, higher total cholesterol level, higher systolic blood pressure, higher prevalence of current smoking, higher prevalence of diabetes, and higher anti-hypertensive medication use in both women and men. The UACR showed a positive linear correlation with the Framingham risk score in both women and men. Conclusion The presence of microalbuminuria was significantly associated with the cardiometabolic risk factors and the increased risk of CVD evaluated by FRS model in both women and men in a nationally representative sample of Korea.

Effectiveness of China-PAR and Framingham risk score in assessment of 10-year cardiovascular disease risk in Chinese hypertensive patients

ObjectivesEstimating the total risk of cardiovascular disease (CVD) using risk prediction models represents a huge improvement in identifying and treating each of the risk factors. The objective of this study was to evaluate the effectiveness of the China-PAR (Prediction of atherosclerotic CVD risk in China) and Framingham risk score (FRS) in predicting the 10-year risk of CVD in Chinese hypertensive patients. The results of the study can be used to design health promotion strategies. Study designA large cohort study was used to assess the validity of models by comparing model predictions with actual incidence rates. MethodsIn total, 10,498 hypertensive patients aged 30–70 years in Jiangsu Province, China, participated in the baseline survey that took place between January and December 2010 and were followed up to May 2020. China-PAR and FRS were used to calculate the predicted 10-year risk of CVD. The 10-year observed incidence of new cardiovascular events was adjusted by the Kaplan–Meier method. The ratio of the predicted risk to the actual incidence was calculated to evaluate the effectiveness of the model. The discrimination Harrell's C statistics and calibration Chi-square value were used to evaluate the predictive reliability of the models. ResultsOf the 10,498 participants, 4411 (42.02%) were male. During the mean follow-up of 8.30 ± 1.45 years, a total of 693 new cardiovascular events occurred. Both models overestimated the risk of morbidity to varying degrees, and the FRS overestimated to a greater extent. After adjustment for covariates, the results of Cox proportional hazards regression showed that the risk of CVD in the high-risk group was higher than in low-risk group. The degree of discrimination in both models was approximately 0.6, which showed that discrimination was not ideal in the models. In addition, Chi-square calibrations of the two models were <20 in males, which showed that calibration of the models was better for men than women. ConclusionsThe China-PAR and FRS models overestimated the risk of CVD for participants in this study. In addition, the degree of discrimination was not ideal, and both models performed better in males than in females in terms of calibration. The results of this study suggest that a more suitable risk prediction model should be established according to the characteristics of the hypertensive population in Jiangsu Province.

Blood immune inflammatory indicators predict prognosis in patients with coronary artery disease

AbstractBackgroundThe relationship between the combined hematological parameters and echocardiography and long‐term prognosis in patients with coronary artery disease (CAD) remains unclear.MethodsWe examined the ability of hematological parameters to predict all‐cause death and major adverse cardiovascular events (MACE) based on Lasso Cox regression analysis. The significant predictors of hematological parameters from the Lasso Cox model were analyzed via multivariate Cox regression analysis and by adjusting for echocardiographic data. We calculated the continuous net reclassification improvement (cNRI) and integrated discrimination improvement (IDI) of the hematological parameters to assess the improvement in prediction.ResultsA low hemoglobin and lymphocyte ratio and high hematocrit, red blood cell distribution width‐coefficient of variation, and monocyte ratio significantly increased the risk of MACE and death in CAD patients. Neutrophil‐to‐lymphocyte ratio was associated with MACE but not death in CAD patients. After adjustment for echocardiographic parameters, hemoglobin, hematocrit, and lymphocyte ratio remained independently related to death and MACE. The addition of hematological and echocardiographic parameters to the Framingham risk score model significantly improved the area under the curve of mortality (0.794 vs. 0.713, p = 0.0007) and reclassification with cNRI of 30.6% (p = 0.002) and IDI of 0.055 (p &lt; 0.001). Mendelian randomization analyses identified that fibrinogen and neutrophil‐to‐lymphocyte ratio were associated with increased brain natriuretic peptide and decreased left ventricular ejection fraction.ConclusionsThese findings suggest that the blood immune inflammatory indicators fibrinogen and neutrophil‐to‐lymphocyte ratio were causally associated with the risk of heart failure after CAD. The combination of hematological biomarkers and echocardiography parameters as predictor variables is a useful predictive tool for all‐cause mortality in patients with CAD.

Lipidomics Profiling and Risk of Coronary Artery Disease in the BioHEART-CT Discovery Cohort

The current coronary artery disease (CAD) risk scores for predicting future cardiovascular events rely on well-recognized traditional cardiovascular risk factors derived from a population level but often fail individuals, with up to 25% of first-time heart attack patients having no risk factors. Non-invasive imaging technology can directly measure coronary artery plaque burden. With an advanced lipidomic measurement methodology, for the first time, we aim to identify lipidomic biomarkers to enable intervention before cardiovascular events. With 994 participants from BioHEART-CT Discovery Cohort, we collected clinical data and performed high-performance liquid chromatography with mass spectrometry to determine concentrations of 683 plasma lipid species. Statin-naive participants were selected based on subclinical CAD (sCAD) categories as the analytical cohort (n = 580), with sCAD+ (n = 243) compared to sCAD− (n = 337). Through a machine learning approach, we built a lipid risk score (LRS) and compared the performance of the existing Framingham Risk Score (FRS) in predicting sCAD+. We obtained individual classifiability scores and determined Body Mass Index (BMI) as the modifying variable. FRS and LRS models achieved similar areas under the receiver operating characteristic curve (AUC) in predicting the validation cohort. LRS enhanced the prediction of sCAD+ in the healthy-weight group (BMI &lt; 25 kg/m2), where FRS performed poorly and identified individuals at risk that FRS missed. Lipid features have strong potential as biomarkers to predict CAD plaque burden and can identify residual risk not captured by traditional risk factors/scores. LRS compliments FRS in prediction and has the most significant benefit in healthy-weight individuals.

Association between urinary albumin creatinine ratio and cardiovascular disease.

The association between microalbuminuria and cardiovascular disease (CVD) is accumulating in various patient populations. However, when stratified by sex, the relationship between microalbuminuria and CVD remains unclear. We obtained data from the 2011-2014 and 2019-2020 Korea National Health and Nutrition Examination Survey (KNHANES). Microalbuminuria was measured based on spot urine albumin-creatinine ratio (UACR). The Framingham risk score (FRS) model was implemented to evaluate the CVD risk. Linear and logistic regression models were used to identify the associations of microalbuminuria status with cardiometabolic predictors and CVD status determined by the FRS score. Among 19,340 representative Korean participants, the (UACR) in Korean women and men with history of CVD was higher than in those without history of CVD. Among patients without history of CVD, multivariate regression analysis showed that a high UACR was related to older age, lower high-density lipoprotein cholesterol level, higher total cholesterol level, higher systolic blood pressure, higher prevalence of current smoking, higher prevalence of diabetes, and higher anti-hypertensive medication use in both women and men. The UACR showed a positive linear correlation with the Framingham risk score in both women and men. The presence of microalbuminuria was significantly associated with the cardiometabolic risk factors and the increased risk of CVD evaluated by FRS model in both women and men in a nationally representative sample of Korea.

Association between NMR metabolomic signatures of healthy lifestyle and incident coronary artery disease.

To identify metabolites associated with a healthy lifestyle and explore the possible mechanisms of lifestyle in coronary artery disease (CAD). The nuclear magnetic resonance metabolomics platform was applied to perform metabolomic profiling of baseline plasma samples from a randomly selected subset of 121 733 UK Biobank participants. Cox proportional hazards models with covariate adjustments were used to investigate the associations between validated lifestyle-associated metabolites and incident CAD and to estimate the accuracy of the inclusion of metabolites to predict CAD compared with traditional prediction models. The discriminatory ability of each model was evaluated using Harrell's C statistic, integrated discrimination improvement (IDI), and continuous net reclassification improvement (NRI) indexes. During a median of 8.6 years of follow-up, 5513 incident CAD cases were documented. Among the 111 lifestyle-associated metabolites, 65 were significantly associated with incident CAD after multivariate adjustment (Bonferroni P < 3.11 × 10-04). The addition of these metabolites to classic risk prediction models [Framingham Risk Score (FRS) using lipids; FRS using body mass index] improved CAD prediction accuracy as assessed by the C statistic (increasing to 0.739 [95% CI, 0.731-0.747] and 0.752 [95% CI, 0.746-0.758]), respectively; continuous NRI (0.274 [0.227-0.325] and 0.266 [0.223-0.317]) and IDI (0.003 [0.002-0.004] and 0.003 [0.002-0.004]). Healthy lifestyle-associated metabolites are associated with the incidence of CAD and may help improve the prediction of CAD risk. The use of metabolite information combined with the FRS model warrants further investigation before clinical implementation.

Comparative Performance Analysis of Random Forests against AutoPrognosis for predicting Coronary Heart Disease Risk and Metabolic Syndrome: A Retrospective Cohort Study

Cardiovascular Disease (CVD) is the leading cause of mortality worldwide. Amongst them, Coronary Heart Disease (CHD) is the most common type of CVD. The consequences of the presence of CVD risk factors often manifest as Metabolic Syndrome (MetS). In this study, data from the Framingham Heart Study (FHS), consisting of 4240 records and 17 variables, was used to build two types of 10-year CHD risk prediction models based on Random Forests (RF) and AutoPrognosis. The Framingham Risk Score model (AUC-ROC: 0.633) was used as a baseline model for performance evaluation. Results showed that the RF model with optimized hyperparameters had the best performance (AUC-ROC: 0.728). Furthermore, a dataset of 7821 records and 77 variables from the National Health and Nutrition Examination Survey (NHANES) was used to assess the predictive performance of RF against AutoPrognosis for determining the presence of MetS. The RF model with optimized hyperparameters had the best performance (AUC-ROC: 0.851). The performance of RF against AutoPrognosis on different sample sizes of data, ranging from 100 to 4900, was tested. The RF model with optimized hyperparameters had the best overall performance, followed by AutoPrognosis with an ensemble pipeline, then AutoPrognosis with a single pipeline and finally the RF model with default hyperparameter values.

Improvement in the Prediction of Coronary Heart Disease Risk by Using Artificial Neural Networks.

Cardiovascular diseases, such as coronary heart disease (CHD), are the main cause of mortality and morbidity worldwide. Although CHD cannot be entirely predicted by classic risk factors, it is preventable. Therefore, predicting CHD risk is crucial to clinical cardiology research, and the development of innovative methods for predicting CHD risk is of great practical interest. The Framingham risk score (FRS) is one of the most frequently implemented risk models. However, recent advances in the field of analytics may enhance the prediction of CHD risk beyond the FRS. Here, we propose a model based on an artificial neural network (ANN) for predicting CHD risk with respect to the Framingham Heart Study (FHS) dataset. The performance of this model was compared to that of the FRS. A sample of 3066 subjects from the FHS offspring cohort was subjected to an ANN. A multilayer perceptron ANN architecture was used and the lift, gains, receiver operating characteristic (ROC), and precision-recall predicted by the ANN were compared with those of the FRS. The lift and gain curves of the ANN model outperformed those of the FRS model in terms of top percentiles. The ROC curve showed that, for higher risk scores, the ANN model had higher sensitivity and higher specificity than those of the FRS model, although its area under the curve (AUC) was lower. For the precision-recall measures, the ANN generated significantly better results than the FRS with a higher AUC. The findings suggest that the ANN model is a promising approach for predicting CHD risk and a good screening procedure to identify high-risk subjects.

Place of cardiovascular risk prediction models in South Asians; agreement between Framingham risk score and WHO/ISH risk charts

There are no cardiovascular risk prediction models developed in South Asian cohorts. Therefore, different risk models not validated in South Asians are being used. We aimed to compare cardiovascular risk predictions of Framingham risk score (FRS) and World Health Organization/International Society of Hypertension (WHO/ISH) charts for agreement in a sample of South Asians. Ten-year cardiovascular risk predictions of patients without previous cardiovascular diseases attending a non-communicable disease clinic were calculated using FRS (with BMI and with cholesterol) and WHO/ISH charts (with and without cholesterol). Patients were categorised into low (<20%) and high (≥20%) cardiovascular risk groups on risk predictions. Agreement in risk categorisation with different prediction models was compared using Cohen's kappa coefficient (κ). One hundred sixty-nine patients (females 130 (81.1%)) mean age 65±6.9years were studied. Of the participants, 80 (47.3%), 62 (36.7%), 18 (10.7%), 16 (9.5%), were predicted high risk by FRS BMI-based, FRS cholesterol-based, WHO/ISH without-cholesterol and WHO/ISH with-cholesterol models, respectively. Agreement between the two FRS models (κ=0.736, P<.0001) and the two WHO/ISH models (κ=0.804, P<.0001) in stratifying patients into high and low-risk groups, was "good." However, the agreements between FRS BMI-based and WHO/ISH without-cholesterol models (κ=0.234, P<.0001) and FRS cholesterol-based and WHO/ISH with-cholesterol models (κ=0.306, P<.0001) were only "fair." Cardiovascular risk predictions of FRS were higher than WHO/ISH charts and the agreement in risk stratification was not satisfactory in Sri Lankans. Therefore, different cardiovascular risk prediction models should not be used interchangeably in the follow-up of South Asians.