Cardiovascular Disease Risk Prediction Models Research Articles

Cardiovascular Risk Estimation Is Suboptimal in People With HIV.

Established cardiovascular disease (CVD) risk prediction functions may not accurately predict CVD risk in people with HIV. We assessed the performance of 3 CVD risk prediction functions in 2 HIV cohorts. CVD risk scores were calculated in the Mass General Brigham and Kaiser Permanente Northern California HIV cohorts, using the American College of Cardiology/American Heart Association atherosclerotic CVD function, the FHS (Framingham Heart Study) hard coronary heart disease function and the Framingham Heart Study hard CVD function. Outcomes were myocardial infarction or coronary death for FHS hard coronary heart disease function; and myocardial infarction, stroke, or coronary death for American College of Cardiology/American Heart Association and FHS hard CVD function. We calculated regression coefficients and assessed discrimination and calibration by sex; predicted to observed risk of outcome was also compared. In the combined cohort of 9412, 158 (1.7%) had a coronary heart disease event, and 309 (3.3%) had a CVD event. Among women, CVD risk was generally underestimated by all 3 risk functions. Among men, CVD risk was underestimated by the American College of Cardiology/American Heart Association and FHS hard CVD function, but overestimated by the FHS hard coronary heart disease function. Calibration was poor for women using the FHS hard CVD function and for men using all functions. Discrimination in all functions was good for women (c-statistics ranging from 0.78 to 0.90) and moderate for men (c-statistics ranging from 0.71 to 0.72). Established CVD risk prediction functions generally underestimate risk in people with HIV. Differences in model performance by sex underscore the need for both HIV-specific and sex-specific functions. Development of CVD risk prediction models tailored to HIV will enhance care for aging people with HIV.

ASSESSMENT OF CARDIOVASCULAR RISK IN PATIENTS WITH RHEUMATOID ARTHRITIS ACCORDING TO CURRENT HYPERTENSION GUIDELINES

Objective: Patients with rheumatoid arthritis (RA) are at increased risk of cardiovascular disease (CVD) by as much as 50% compared to the general population. Considering that disease-specific calculators have failed to provide more accurate CVD risk stratification in RA, CVD risk calculation in RA patients is challenging, especially for those who are not by definition at high risk based on concomitant established CVD or its equivalents. The presence of an immune-mediated inflammatory disease (e.g., RA) has been considered as a CVD risk modifier by the previous 2018 European Society of Hypertension/European Society of Cardiology (ESH/ESC) guidelines. Nevertheless, current 2023 ESH/ESC guidelines recommend assessment in RA using general CVD risk calculators. Design and method: We aimed to calculate CVD risk in RA patients using the most frequently applied CVD risk calculators in the general population. We only stratified patients free from established CVD (heart disease, stroke) or its equivalents (diabetes mellitus, chronic kidney disease) aged 40-74 years. The following algorithms were applied: (i) SCORE2 (Systemic Coronary Risk Estimation-2), (ii) Framingham Risk Score-Adult Treatment Panel (FRS-ATP) CVD risk calculator, and (iii) the American College of Cardiology/American Heart Association (ACC/AHA) Pooled Cohort Equation. We further correlated the estimated CVD risk by each algorithm with surrogate CVD markers, i.e. arterial stiffness and carotid atherosclerosis. For this purpose, carotid-femoral pulse wave velocity (cfPWV) was measured using applanation tonometry (Sphygmocor ®), and carotid intima-media thickness (cIMT) was estimated using carotid ultrasound. Results: A total of 91 patients aged 60.3±8.1 years were included (78% females). Using the FRS-ATP algorithm, more patients were classified as high risk and fewer as low risk compared to the other calculators (Table 1). Moreover, CVD risk calculated with FRS-ATP correlated stronger with both cfPWV and cIMT, compared to CVD risk calculated with the other algorithms (Table 2). Conclusions: In RA patients without established CVD, diabetes, or chronic kidney disease, the FRS-ATP algorithm may be applied as a useful tool for CVD risk stratification. Prospective, large-scale studies are warranted to determine the most reliable CVD risk prediction model for patients with RA.

Development and validation of risk prediction model for recurrent cardiovascular events among Chinese: the Personalized CARdiovascular DIsease risk Assessment for Chinese model.

Cardiovascular disease (CVD) is a leading cause of mortality, especially in developing countries. This study aimed to develop and validate a CVD risk prediction model, Personalized CARdiovascular DIsease risk Assessment for Chinese (P-CARDIAC), for recurrent cardiovascular events using machine learning technique. Three cohorts of Chinese patients with established CVD were included if they had used any of the public healthcare services provided by the Hong Kong Hospital Authority (HA) since 2004 and categorized by their geographical locations. The 10-year CVD outcome was a composite of diagnostic or procedure codes with specific International Classification of Diseases, Ninth Revision, Clinical Modification. Multivariate imputation with chained equations and XGBoost were applied for the model development. The comparison with Thrombolysis in Myocardial Infarction Risk Score for Secondary Prevention (TRS-2°P) and Secondary Manifestations of ARTerial disease (SMART2) used the validation cohorts with 1000 bootstrap replicates. A total of 48 799, 119 672 and 140 533 patients were included in the derivation and validation cohorts, respectively. A list of 125 risk variables were used to make predictions on CVD risk, of which 8 classes of CVD-related drugs were considered interactive covariates. Model performance in the derivation cohort showed satisfying discrimination and calibration with a C statistic of 0.69. Internal validation showed good discrimination and calibration performance with C statistic over 0.6. The P-CARDIAC also showed better performance than TRS-2°P and SMART2. Compared with other risk scores, the P-CARDIAC enables to identify unique patterns of Chinese patients with established CVD. We anticipate that the P-CARDIAC can be applied in various settings to prevent recurrent CVD events, thus reducing the related healthcare burden.

A Klotho-Based Machine Learning Model for Prediction of both Kidney and Cardiovascular Outcomes in Chronic Kidney Disease

Introduction: This study aimed to develop and validate machine learning (ML) models based on serum Klotho for predicting end-stage kidney disease (ESKD) and cardiovascular disease (CVD) in patients with chronic kidney disease (CKD). Methods: Five different ML models were trained to predict the risk of ESKD and CVD at three different time points (3, 5, and 8 years) using a cohort of 400 non-dialysis CKD patients. The dataset was divided into a training set (70%) and an internal validation set (30%). These models were informed by data comprising 47 clinical features, including serum Klotho. The best-performing model was selected and used to identify risk factors for each outcome. Model performance was assessed using various metrics. Results: The findings showed that the least absolute shrinkage and selection operator regression model had the highest accuracy (C-index = 0.71) in predicting ESKD. The features mainly included in this model were estimated glomerular filtration rate, 24-h urinary microalbumin, serum albumin, phosphate, parathyroid hormone, and serum Klotho, which achieved the highest area under the curve (AUC) of 0.930 (95% CI: 0.897–0.962). In addition, for the CVD risk prediction, the random survival forest model with the highest accuracy (C-index = 0.66) was selected and achieved the highest AUC of 0.782 (95% CI: 0.633–0.930). The features mainly included in this model were age, history of primary hypertension, calcium, tumor necrosis factor-alpha, and serum Klotho. Conclusion: We successfully developed and validated Klotho-based ML risk prediction models for CVD and ESKD in CKD patients with good performance, indicating their high clinical utility.

Serum Inflammatory Markers Used in Cardiovascular Disease Risk Prediction Models: A Systematic Review.

The aim of this review was to perform a critical appraisal of serum inflammatory biomarkers used for the prediction of cardiovascular disease (CVD) risk. We conducted a systematic review of studies listed on MEDLINE and Scopus from January 2000 to December 2023, focused on the prognostic value of serum inflammatory biomarkers [i.e., C-reactive protein (CRP), interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α)] in individuals without known CVD. Eligible studies used a multivariate prediction model and included discrimination or risk reclassification analysis. The Quality in Prognostic Studies (QUIPS) tool was used to evaluate study quality and potential bias. Thirty-five studies (i.e., total 208,897 participants) that evaluated the added prognostic value of CRP, IL-6, TNF-α on CVD risk prediction were retrieved. Significant improvements in CVD risk model's predictive ability were observed in 7 out of 32 studies relating CRP and 1 out of 8 studies relating IL-6 with CVD risk. The single study found no added prognostic value of TNF-α use in CVD risk model. The integration of serum inflammatory biomarkers into CVD risk prediction models does not appear to improve risk discrimination models, suggesting that these biomarkers may act as surrogate markers, but not as predictors of atherosclerotic CVD.

Comparative Analysis of Three Atherosclerotic Cardiovascular Disease Risk Prediction Models in Individuals Aged 75 and Older.

To evaluate the performance of the Framingham cardiovascular risk score (FRS)/pooled cohort equations (PCE)/China prediction for atherosclerotic cardiovascular disease (ASCVD) risk (China-PAR model) in a prospective cohort of Chinese older adults. We assessed 717 older adults aged 75-85 years without ASCVD at the baseline from the Sichuan province of China. The participants were followed annually from 2011 to 2021. We obtained the participants' information through the medical records of physical examination and evaluated their 10-year ASCVD risk using FRS, PCE, and China-PAR. We further evaluated the predictive abilities of three assessment models. During the 10-year follow-up, 206 participants developed ASCVD, with an incidence rate of 28.73%. The FRS and China-PAR moderately underestimated the risk of ASCVD (22.1% and 12.4%, respectively), but while PCE overestimated the risk (36.1%). FRS and China-PAR were found to underestimate the risk of ASCVD (26% and 63%, respectively) for men, while PCE overestimated the risk by 8%; For women, FRS and China-PAR were found to underestimate the risk of ASCVD (14% and 35%, respectively), while PCE overestimated the risk by 88%. The 10-year ASCVD risk was found to be overestimated by PCE. China-PAR had the most accurate predictions in women, while FRS was particularly well-calibrated in males. All three risk models have good discrimination, with FRS and PCE being well-calibrated in men and all three being well-calibrated in women. Therefore, accurate risk models are warranted to facilitate the prevention of ASCVD at the baseline among Chinese older adults.

Comparison of the effects of imputation methods for missing data in predictive modelling of cohort study datasets

BackgroundMissing data is frequently an inevitable issue in cohort studies and it can adversely affect the study's findings. We assess the effectiveness of eight frequently utilized statistical and machine learning (ML) imputation methods for dealing with missing data in predictive modelling of cohort study datasets. This evaluation is based on real data and predictive models for cardiovascular disease (CVD) risk.MethodsThe data is from a real-world cohort study in Xinjiang, China. It includes personal information, physical examination data, questionnaires, and laboratory biochemical results from 10,164 subjects with a total of 37 variables. Simple imputation (Simple), regression imputation (Regression), expectation-maximization(EM), multiple imputation (MICE) , K nearest neighbor classification (KNN), clustering imputation (Cluster), random forest (RF), and decision tree (Cart) were the chosen imputation methods. Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) are utilised to assess the performance of different methods for missing data imputation at a missing rate of 20%. The datasets processed with different missing data imputation methods were employed to construct a CVD risk prediction model utilizing the support vector machine (SVM). The predictive performance was then compared using the area under the curve (AUC).ResultsThe most effective imputation results were attained by KNN (MAE: 0.2032, RMSE: 0.7438, AUC: 0.730, CI: 0.719-0.741) and RF (MAE: 0.3944, RMSE: 1.4866, AUC: 0.777, CI: 0.769-0.785). The subsequent best performances were achieved by EM, Cart, and MICE, while Simple, Regression, and Cluster attained the worst performances. The CVD risk prediction model was constructed using the complete data (AUC:0.804, CI:0.796-0.812) in comparison with all other models with p<0.05.ConclusionKNN and RF exhibit superior performance and are more adept at imputing missing data in predictive modelling of cohort study datasets.

Age-specific differences in the association between prediabetes and cardiovascular diseases in China: A national cross-sectional study.

Cardiovascular disease (CVD) is a leading cause of morbidity and mortality worldwide, the global burden of which is rising. It is still unclear the extent to which prediabetes contributes to the risk of CVD in various age brackets among adults. To develop a focused screening plan and treatment for Chinese adults with prediabetes, it is crucial to identify variations in the connection between prediabetes and the risk of CVD based on age. To examine the clinical features of prediabetes and identify risk factors for CVD in different age groups in China. The cross-sectional study involved a total of 46239 participants from June 2007 through May 2008. A thorough evaluation was conducted. Individuals with prediabetes were categorized into two groups based on age. Chinese atherosclerotic CVD risk prediction model was employed to evaluate the risk of developing CVD over 10 years. Random forest was established in both age groups. SHapley Additive exPlanation method prioritized the importance of features from the perspective of assessment contribution. In total, 6948 people were diagnosed with prediabetes in this study. In pre-diabetes, prevalences of CVD were 5 (0.29%) in the younger group and 148 (2.85%) in the older group. Overall, 11.11% of the younger group and 29.59% of the older group were intermediate/high-risk of CVD for prediabetes without CVD based on the Prediction for ASCVD Risk in China equation in ten years. In the younger age group, the 10-year risk of CVD was found to be more closely linked to family history of CVD rather than lifestyle, whereas in the older age group, resident status was more closely linked. The susceptibility to CVD is age-specific in newly diagnosed prediabetes. It is necessary to develop targeted approaches for the prevention and management of CVD in adults across various age brackets.

Alcohol Consumption, High-Density Lipoprotein Particles and Subspecies, and Risk of Cardiovascular Disease: Findings from the PREVEND Prospective Study.

The associations of HDL particle (HDL-P) and subspecies concentrations with alcohol consumption are unclear. We aimed to evaluate the interplay between alcohol consumption, HDL parameters and cardiovascular disease (CVD) risk. In the PREVEND study of 5151 participants (mean age, 53 years; 47.5% males), self-reported alcohol consumption and HDL-P and subspecies (small, medium, and large) by nuclear magnetic resonance spectroscopy were assessed. Hazard ratios (HRs) with 95% CIs for first CVD events were estimated. In multivariable linear regression analyses, increasing alcohol consumption increased HDL-C, HDL-P, large and medium HDL, HDL size, and HDL subspecies (H3P, H4P, H6 and H7) in a dose-dependent manner. During a median follow-up of 8.3 years, 323 first CVD events were recorded. Compared with abstainers, the multivariable adjusted HRs (95% CIs) of CVD for occasional to light, moderate, and heavy alcohol consumers were 0.72 (0.55-0.94), 0.74 (0.54-1.02), and 0.65 (0.38-1.09), respectively. These associations remained consistent on additional adjustment for each HDL parameter. For CVD, only HDL-C was associated with a statistically significant decreased risk of CVD in a fully adjusted analysis (HR 0.84, 95% CI 0.72-0.97 per 1 SD increment). For coronary heart disease, HDL-C, HDL-P, medium HDL, HDL size, and H4P showed inverse associations, whereas HDL-C and HDL size modestly increased stroke risk. Except for H6P, alcohol consumption did not modify the associations between HDL parameters and CVD risk. The addition of HDL-C, HDL size, or H4P to a CVD risk prediction model containing established risk factors improved risk discrimination. Increasing alcohol consumption is associated with increased HDL-C, HDL-P, large and medium HDL, HDL size, and some HDL subspecies. Associations of alcohol consumption with CVD are largely independent of HDL parameters. The associations of HDL parameters with incident CVD are generally not attenuated or modified by alcohol consumption.

Performance of ASCVD Risk Prediction Models in Individuals With Inflammatory Bowel Disease: A UK Biobank Study.

Lay Summary This study suggests that the current atherosclerotic cardiovascular disease risk prediction models used in clinical practice performed better in the non–inflammatory bowel disease (IBD) cohort compared with IBD, highlighting the need for a more specific risk prediction model tailored to the IBD population.

Abstract 15002: Cardiovascular Disease Risk Prediction by Random Survival Forest: The Korean National Health Insurance Service-National Health Screening Cohort

Introduction: Risk prediction is critical for cardiovascular disease (CVD) prevention and management. Existing CVD risk prediction models including the Framingham Risk Score (FRS) are based on the Cox proportional hazard ratio regression. Adaptation of a data mining approach could improve the accuracy of CVD risk prediction by being free from the linearity assumption. The aim of study was to construct Korean population-based CVD risk prediction models using random survival forest and to compare their predictive accuracy to the FRS in a 14-year Korean cohort. Methods: This was a secondary data analysis using a Korean population-based cohort. Out of 463,783 eligible participants, 70% were selected for training dataset using stratified random sampling, and the remaining 30% served as test dataset. Random survival forest was adopted to build risk prediction models using the training dataset. The test dataset was used to test the validity of the new models and the FRS. Results: Three random forest-based models were developed, and they were good in discrimination in both men and women with the c-statistics ranging from 0.698 to 0.747 while the c-statistics for the FRS were 0.692 and 0.733 in men and women, respectively. All random forest-based models showed good calibration, whereas the FRS tended to overestimate the CVD risk for men. Conclusions: The random survival forest-based models showed comparable performance with the FRS in a Korean population-based cohort. Continuous efforts are needed to improve the accuracy of CVD risk prediction for the Korean population and to utilize CVD risk prediction in clinical practice.

Association of fat-to-muscle ratio with non-alcoholic fatty liver disease: a single-centre retrospective study

ObjectivesSarcopenia is a known risk factor for non-alcoholic fatty liver disease (NAFLD). Studies evaluating the association between the fat-to-muscle ratio (FMR) and NAFLD are limited. Therefore, the aim of our...

A comparative assessment between Globorisk and WHO cardiovascular disease risk scores: a population-based study

The Globorisk and WHO cardiovascular risk prediction models are country-specific and region-specific, respectively. The goal of this study was to assess the agreement and correlation between the WHO and Globorisk 10-year cardiovascular disease risk prediction models. The baseline data of 6796 individuals aged 40–74 years who participated in the Fasa cohort study without a history of cardiovascular disease or stroke at baseline were included. In the WHO and Globorisk models scores were calculated using age, sex, systolic blood pressure (SBP), current smoking, diabetes, and total cholesterol for laboratory-based risk and age, sex, SBP, current smoking, and body mass index (BMI) for non-laboratory-based risk (office-based or BMI-based). In Globorisk and WHO risk agreement across risk categories (low, moderate, and high) was examined using the kappa statistic. Also, Pearson correlation coefficients and scatter plots were used to assess the correlation between Globorisk and WHO models. Bland–Altman plots were presented for determination agreement between Globorisk and WHO risk scores in individual’s level. In laboratory-based models, agreement across categories was substantial in the overall population (kappa values: 0.75) and also for females (kappa values: 0.74) and males (kappa values: 0.76), when evaluated separately. In non-laboratory-based models, agreement across categories was substantial for the whole population (kappa values: 0.78), and almost perfect for among males (kappa values: 0.82) and substantial for females (kappa values: 0.73). The results showed a very strong positive correlation (r ≥ 0.95) between WHO and Globorisk laboratory-based scores for the whole population, males, and females and also a very strong positive correlation (r > 0.95) between WHO and Globorisk non-laboratory-based scores for the whole population, males, and females. In the laboratory-based models, the limit of agreements was better in males (95%CI 2.1 to − 4.2%) than females (95%CI 4.3 to − 7.3%). Also, in the non-laboratory-based models, the limit of agreements was better in males (95%CI 2.9 to − 4.0%) than females (95%CI 3.2 to − 6.1%). There was a good agreement between both the laboratory-based and the non-laboratory-based WHO models and the Globorisk models. The correlation between two models was very strongly positive. However, in the Globorisk models, more people were in high-risk group than in the WHO models. The scatter plots and Bland–Altman plots showed systematic differences between the two scores that vary according to the level of risk. So, for these models may be necessary to modify the cut points of risk groups. The validity of these models must be determined for this population.

Development of a Cardiovascular Disease Risk Prediction Model: A Preliminary Retrospective Cohort Study of a Patient Sample in Saudi Arabia.

Saudi Arabia has an alarmingly high incidence of cardiovascular disease (CVD) and its associated risk factors. To effectively assess CVD risk, it is essential to develop tailored models for diverse regions and ethnicities using local population variables. No CVD risk prediction model has been locally developed. This study aims to develop the first 10-year CVD risk prediction model for Saudi adults aged 18 to 75 years. The electronic health records of Saudi male and female patients aged 18 to 75 years, who were seen in primary care settings between 2002 and 2019, were reviewed retrospectively via the Integrated Clinical Information System (ICIS) database (from January 2002 to February 2019). The Cox regression model was used to identify the risk factors and develop the CVD risk prediction model. Overall, 451 patients were included in this study, with a mean follow-up of 12.05 years. Thirty-five (7.7%) patients developed a CVD event. The following risk factors were included: fasting blood sugar (FBS) and high-density lipoprotein cholesterol (HDL-c), heart failure, antihyperlipidemic therapy, antithrombotic therapy, and antihypertension therapy. The Bayesian information criterion (BIC) score was 314.4. This is the first prediction model developed in Saudi Arabia and the second in any Arab country after the Omani study. We assume that our CVD predication model will have the potential to be used widely after the validation study.

Independent external validation of the QRISK3 cardiovascular disease risk prediction model using UK Biobank

ObjectiveTo externally evaluate the performance of QRISK3 for predicting 10 year risk of cardiovascular disease (CVD) in the UK Biobank cohort.MethodsWe used data from the UK Biobank, a large-scale prospective...

Validation of Cardiovascular Disease Risk Prediction Models in Community-dwelling Australian and US Older Adults

Validation of Cardiovascular Disease Risk Prediction Models in Community-dwelling Australian and US Older Adults

Derivation of a Protein Risk Score for Cardiovascular Disease Among a Multiracial and Multiethnic HIV+ Cohort.

Background Cardiovascular disease risk prediction models underestimate CVD risk in people living with HIV (PLWH). Our goal is to derive a risk score based on protein biomarkers that could be used to predict CVD in PLWH. Methods and Results In a matched case-control study, we analyzed normalized protein expression data for participants enrolled in 1 of 4 trials conducted by INSIGHT (International Network for Strategic Initiatives in Global HIV Trials). We used dimension reduction, variable selection and resampling methods, and multivariable conditional logistic regression models to determine candidate protein biomarkers and to generate a protein score for predicting CVD in PLWH. We internally validated our findings using bootstrap. A protein score that was derived from 8 proteins (including HGF [hepatocyte growth factor] and interleukin-6) was found to be associated with an increased risk of CVD after adjustment for CVD and HIV factors (odds ratio: 2.17 [95% CI: 1.58-2.99]). The protein score improved CVD prediction when compared with predicting CVD risk using the individual proteins that comprised the protein score. Individuals with a protein score above the median score were 3.10 (95% CI, 1.83-5.41) times more likely to develop CVD than those with a protein score below the median score. Conclusions A panel of blood biomarkers may help identify PLWH at a high risk for developing CVD. If validated, such a score could be used in conjunction with established factors to identify CVD at-risk individuals who might benefit from aggressive risk reduction, ultimately shedding light on CVD pathogenesis in PLWH.

The Role of Social Determinants of Health in Cardiovascular Diseases: An Umbrella Review.

Cardiovascular disease (CVD) is the leading cause of mortality worldwide. Addressing social determinants of health (SDoH) may be the next forefront of reducing the enormous burden of CVD. SDoH can be defined as any social, economic, or environmental factor that influences a health outcome. Comprehensive evidence of the role of SDoH in CVD is lacking, nevertheless. This umbrella review aims to give a comprehensive overview of the role of SDoH in CVD. We searched systematic reviews (with or without meta-analyses) using 8 databases and included review reference lists. Four themes (economic circumstances, social/community context, early childhood development, and neighbourhood/built environment) and health literacy in the health/health care theme were considered. Seventy reviews were eligible. Despite the quality of the included reviews being low or critically low, there was consistent evidence that factors relating to economic circumstances and early childhood development themes were associated with an increased risk of CVD and CVD mortality. We also found evidence that factors in the social/community context and neighbourhood/built environment themes, such as social isolation, fewer social roles, loneliness, discrimination, ethnicity, neighborhood socioeconomic status, violence, and environmental attributes, had a role in CVD. SDoH factors without (or with minimal) evidence synthesis for CVD were also identified. In sum, this umbrella review offers evidence that SDoH, especially economic circumstance and early childhood development, play a significant role in CVD. This calls for the strengthening of nonmedical interventions that address multiple factors simultaneously and the inclusion of SDoH in future CVD risk prediction models. Registration URL: https://www.crd.york.ac.uk/prospero/; Unique identifier: CRD42022346994.

Validation and comparison of cardiovascular risk prediction equations in Chinese patients with Type 2 diabetes.

For patients with diabetes, the European guidelines updated the cardiovascular disease (CVD) risk prediction recommendations using diabetes-specific models with age-specific cut-offs, whereas American guidelines still advise models derived from the general population. We aimed to compare the performance of four cardiovascular risk models in diabetes populations. Patients with diabetes from the CHERRY study, an electronic health records-based cohort study in China, were identified. Five-year CVD risk was calculated using original and recalibrated diabetes-specific models [Action in Diabetes and Vascular disease: PreterAx and diamicroN-MR Controlled Evaluation (ADVANCE) and the Hong Kong cardiovascular risk model (HK)] and general population-based models [Pooled Cohort Equations (PCE) and Prediction for Atherosclerotic cardiovascular disease Risk in China (China-PAR)]. During a median 5.8-year follow-up, 46 558 patients had 2605 CVD events. C-statistics were 0.711 [95% confidence interval: 0.693-0.729] for ADVANCE and 0.701 (0.683-0.719) for HK in men, and 0.742 (0.725-0.759) and 0.732 (0.718-0.747) in women. C-statistics were worse in two general population-based models. Recalibrated ADVANCE underestimated risk by 1.2% and 16.8% in men and women, whereas PCE underestimated risk by 41.9% and 24.2% in men and women. With the age-specific cut-offs, the overlap of the high-risk patients selected by every model pair ranged from only 22.6% to 51.2%. When utilizing the fixed cut-off at 5%, the recalibrated ADVANCE selected similar high-risk patients in men (7400) as compared to the age-specific cut-offs (7102), whereas age-specific cut-offs exhibited a reduction in the selection of high-risk patients in women (2646 under age-specific cut-offs vs. 3647 under fixed cut-off). Diabetes-specific CVD risk prediction models showed better discrimination for patients with diabetes. High-risk patients selected by different models varied significantly. Age-specific cut-offs selected fewer patients at high CVD risk especially in women.

The Role of Adverse Pregnancy Outcomes in Conventional Cardiovascular Risk Prediction.

Approximately one-third of women in the U.S. experience an adverse pregnancy outcome (APO), which are recognized as sex-specific cardiovascular disease (CVD) risk factors. We examine if APOs confer additional CVD risk beyond that of traditional CVD risk factors. Women, age 40-79, with a pregnancy history and no pre-existing CVD were identified in the electronic health record of one health system (n = 2306). APOs included any APO, hypertensive disease of pregnancy (HDP), and gestational diabetes (GDM). Hazard ratios of time to CVD event were estimated from survival models using Cox proportional hazard regression. Discrimination, calibration, and net reclassification of re-estimated CVD risk prediction models including APOs were examined. There was no significant association between any APO, HDP, or GDM and time to CVD outcome in survival models (95% confidence intervals all include 1). Including any APO, HDP, GDM in the CVD risk prediction model did not significantly improve discrimination and there were no clinically relevant changes in net reclassification of cases and non-cases. The strongest predictor of time to CVD event in the survival models was Black race, with hazard ratios ranging from 1.59 to 1.62, statistically significant for all three models. Women with APOs did not have an additional risk of CVD, controlling for traditional risk factors in the PCE and this sex-specific factor did not improve risk prediction. Black race was consistently a strong predictor of CVD even with data limitations. Further study of APOs can help determine how to best use this information for CVD prevention in women.