scholarly journals Modelling of longitudinal data to predict cardiovascular disease risk: a methodological review

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
David Stevens ◽  
Deirdre A. Lane ◽  
Stephanie L. Harrison ◽  
Gregory Y. H. Lip ◽  
Ruwanthi Kolamunnage-Dona
Keyword(s):  
Cardiovascular Disease ◽  
Longitudinal Data ◽  
Risk Prediction ◽  
Single Stage ◽  
Survival Models ◽  
Joint Models ◽  
Cvd Risk ◽  
Two Stage ◽  
Time Points ◽  
Search Terms

Abstract Objective The identification of methodology for modelling cardiovascular disease (CVD) risk using longitudinal data and risk factor trajectories. Methods We screened MEDLINE-Ovid from inception until 3 June 2020. MeSH and text search terms covered three areas: data type, modelling type and disease area including search terms such as “longitudinal”, “trajector*” and “cardiovasc*” respectively. Studies were filtered to meet the following inclusion criteria: longitudinal individual patient data in adult patients with ≥3 time-points and a CVD or mortality outcome. Studies were screened and analyzed by one author. Any queries were discussed with the other authors. Comparisons were made between the methods identified looking at assumptions, flexibility and software availability. Results From the initial 2601 studies returned by the searches 80 studies were included. Four statistical approaches were identified for modelling the longitudinal data: 3 (4%) studies compared time points with simple statistical tests, 40 (50%) used single-stage approaches, such as including single time points or summary measures in survival models, 29 (36%) used two-stage approaches including an estimated longitudinal parameter in survival models, and 8 (10%) used joint models which modelled the longitudinal and survival data together. The proportion of CVD risk prediction models created using longitudinal data using two-stage and joint models increased over time. Conclusions Single stage models are still heavily utilized by many CVD risk prediction studies for modelling longitudinal data. Future studies should fully utilize available longitudinal data when analyzing CVD risk by employing two-stage and joint approaches which can often better utilize the available data.

scholarly journals Cardiovascular disease risk prediction in older people: a qualitative study

2021 ◽  
pp. BJGP.2020.1038
Author(s):  
Denise Ann Taylor ◽  
Katharine Wallis ◽  
Sione Feki ◽  
Sione Segili Moala ◽  
Manusiu He-Naua Esther Latu ◽  
...  
Keyword(s):  
Cardiovascular Disease ◽  
Risk Assessment ◽  
Qualitative Study ◽  
Older People ◽  
Focus Groups ◽  
Risk Prediction ◽  
Heart Attack ◽  
Clinical Decision Making ◽  
Community Dwelling ◽  
Cvd Risk

Background: Despite cardiovascular disease (CVD) risk prediction equations becoming more widely available for people aged 75 years and over, views of older people on CVD risk assessment are unknown. Aim: To explore older people’s views on CVD risk prediction and its assessment. Design and Setting: Qualitative study of community dwelling older New Zealanders. Methods: We purposively recruited a diverse group of older people. Semi-structured interviews and focus groups were conducted, transcribed verbatim and thematically analysed. Results: Thirty-nine participants (mean age 74 years) of Māori, Pacific, South Asian and European ethnicities participated in one of 26 interviews or three focus groups. Three key themes emerged, (1) Poor knowledge and understanding of cardiovascular disease and its risk assessment, (2) Acceptability and perceived benefit of knowing and receiving advice on managing personal cardiovascular risk; and (3) Distinguishing between CVD outcomes; stroke and heart attack are not the same. Most participants did not understand CVD terms but were familiar with ‘heart attack,’ ‘stroke’ and understood lifestyle risk factors for these events. Participants valued CVD outcomes differently, fearing stroke and disability which might adversely affect independence and quality of life, but being less concerned about a heart attack, perceived as causing less disability and swifter death. These findings and preferences were similar across ethnic groups. Conclusion: Older people want to know their CVD risk and how to manage it, but distinguish between CVD outcomes. To inform clinical decision making for older people, risk prediction tools should provide separate event types rather than just composite outcomes.

Cardiovascular Risk Based on ASCVD and KDIGO Categories in Chinese Adults: A Nationwide, Population-Based, Prospective Cohort Study

2021 ◽  
pp. ASN.2020060856
Author(s):  
Yu Xu ◽  
Mian Li ◽  
Guijun Qin ◽  
Jieli Lu ◽  
Li Yan ◽  
...  
Keyword(s):  
Cardiovascular Disease ◽  
Cohort Study ◽  
Risk Prediction ◽  
Cox Regression ◽  
Rank Test ◽  
Atherosclerotic Cardiovascular Disease ◽  
Cvd Risk ◽  
Cox Regression Analysis ◽  
Ascvd Risk ◽  
Risk Categories

BackgroundThe Kidney Disease Improving Global Outcomes (KDIGO) clinical practice guideline used eGFR and urinary albumin-creatinine ratio (ACR) to categorize risks for CKD prognosis. The utility of KDIGO’s stratification of major CVD risks and predictive ability beyond traditional CVD risk prediction scores are unknown.MethodsTo evaluate CVD risks on the basis of ACR and eGFR (individually, together, and in combination using the KDIGO risk categories) and with the atherosclerotic cardiovascular disease (ASCVD) score, we studied 115,366 participants in the China Cardiometabolic Disease and Cancer Cohort study. Participants (aged ≥40 years and without a history of cardiovascular disease) were examined prospectively for major CVD events, including nonfatal myocardial infarction, nonfatal stroke, and cardiovascular death.ResultsDuring 415,111 person-years of follow-up, 2866 major CVD events occurred. Incidence rates and multivariable-adjusted hazard ratios of CVD events increased significantly across the KDIGO risk categories in ASCVD risk strata (all P values for log-rank test and most P values for trend in Cox regression analysis <0.01). Increases in c statistic for CVD risk prediction were 0.01 (0.01 to 0.02) in the overall study population and 0.03 (0.01 to 0.04) in participants with diabetes, after adding eGFR and log(ACR) to a model including the ASCVD risk score. In addition, adding eGFR and log(ACR) to a model with the ASCVD score resulted in significantly improved reclassification of CVD risks (net reclassification improvements, 4.78%; 95% confidence interval, 3.03% to 6.41%).ConclusionsUrinary ACR and eGFR (individually, together, and in combination using KDIGO risk categories) may be important nontraditional risk factors in stratifying and predicting major CVD events in the Chinese population.

Abstract 36: The Role of Sleep as a Cardiovascular Health Metric: Does It Improve Cardiovascular Disease Risk Prediction? Results From The Multi-Ethnic Study of Atherosclerosis

Circulation ◽  
2020 ◽  
Vol 141 (Suppl_1) ◽  
Author(s):  
Nour Makarem ◽  
Cecilia Castro-Diehl ◽  
Marie-Pierre St-Onge ◽  
Susan Redline ◽  
Steven Shea ◽  
...  
Keyword(s):  
Cardiovascular Disease ◽  
Sleep Duration ◽  
Risk Prediction ◽  
Cardiovascular Health ◽  
Cvd Risk ◽  
Sleep Study ◽  
Lower Odds ◽  
Wrist Actigraphy ◽  
Incident Cases

Background: The AHA Life’s Simple 7 (LS7) is a measure of cardiovascular health (CVH). Sufficient and healthy sleep has been linked to higher LS7 scores and lower cardiovascular disease (CVD) risk, but sleep has not been included as a CVH metric. Hypothesis: A CVH score that includes the LS7 plus sleep metrics would be more strongly associated with CVD outcomes than the LS7 score. Methods: Participants were n=1920 diverse adults (mean age: 69.5 y) in the MESA Sleep Study who completed 7 days of wrist actigraphy, overnight in-home polysomnography, and sleep questionnaires. Logistic regression and Cox proportional hazards models were used to compare the LS7 score and 4 new CVH scores that incorporate aspects of sleep in relation to CVD prevalence and incidence (Table). There were 95 prevalent CVD events at the Sleep Exam and 93 incident cases during a mean follow up of 4.4y. Results: The mean LS7 score was 7.3, and the means of the alternate CVH scores ranged from 7.4 to 7.8. Overall, 63% of participants slept <7h, 10% had sleep efficiency <85%, 14% and 36% reported excess daytime sleepiness and insomnia, respectively, 47% had obstructive sleep apnea, and 39% and 25% had high night-to-night variability in sleep duration and sleep onset timing. The LS7 score was not significantly associated with CVD prevalence or incidence (Table). Those in the highest vs. lowest tertile of CVH score 1, that included sleep duration, and CVH score 2, that included sleep characteristics linked to CVD in the literature, had lower odds of prevalent CVD. Those in the highest vs. lowest tertile of CVH scores 3 and 4, which included sleep characteristics linked to cardiovascular risk in MESA, had lower odds of prevalent CVD and lower risk of developing CVD. Conclusions: CVH scores that include sleep were more strongly associated with CVD prevalence and incidence than the traditional LS7 score. The incorporation of sleep as a metric of CVH, akin to other health behaviors, may improve CVD risk prediction. Findings warrant confirmation in larger samples and over longer follow-up.

scholarly journals A Cox-Based Risk Prediction Model for Early Detection of Cardiovascular Disease: Identification of Key Risk Factors for the Development of a 10-Year CVD Risk Prediction

2019 ◽  
Vol 2019 ◽  
pp. 1-11 ◽  
Author(s):  
Xiaona Jia ◽  
Mirza Mansoor Baig ◽  
Farhaan Mirza ◽  
Hamid GholamHosseini
Keyword(s):  
Risk Factors ◽  
Cardiovascular Disease ◽  
Heart Rate ◽  
Prediction Model ◽  
Risk Prediction ◽  
Risk Model ◽  
Clinical Laboratory ◽  
Risk Prediction Model ◽  
Validation Dataset ◽  
Cvd Risk

Background and Objective. Current cardiovascular disease (CVD) risk models are typically based on traditional laboratory-based predictors. The objective of this research was to identify key risk factors that affect the CVD risk prediction and to develop a 10-year CVD risk prediction model using the identified risk factors. Methods. A Cox proportional hazard regression method was applied to generate the proposed risk model. We used the dataset from Framingham Original Cohort of 5079 men and women aged 30-62 years, who had no overt symptoms of CVD at the baseline; among the selected cohort 3189 had a CVD event. Results. A 10-year CVD risk model based on multiple risk factors (such as age, sex, body mass index (BMI), hypertension, systolic blood pressure (SBP), cigarettes per day, pulse rate, and diabetes) was developed in which heart rate was identified as one of the novel risk factors. The proposed model achieved a good discrimination and calibration ability with C-index (receiver operating characteristic (ROC)) being 0.71 in the validation dataset. We validated the model via statistical and empirical validation. Conclusion. The proposed CVD risk prediction model is based on standard risk factors, which could help reduce the cost and time required for conducting the clinical/laboratory tests. Healthcare providers, clinicians, and patients can use this tool to see the 10-year risk of CVD for an individual. Heart rate was incorporated as a novel predictor, which extends the predictive ability of the past existing risk equations.

scholarly journals Cardiovascular Disease Risk Prediction in Women: Is There a Role for Novel Biomarkers?

2014 ◽  
Vol 60 (1) ◽  
pp. 88-97 ◽  
Author(s):  
Nina P Paynter ◽  
Brendan M Everett ◽  
Nancy R Cook
Keyword(s):  
Cardiovascular Disease ◽  
Risk Prediction ◽  
Predictive Accuracy ◽  
Disease Risk ◽  
Troponin T ◽  
Cardiovascular Disease Risk ◽  
High Sensitivity ◽  
Preventive Therapy ◽  
Cvd Risk ◽  
Novel Biomarkers

Abstract BACKGROUND Risk prediction is an integral part of the current US guidelines for cardiovascular disease in women. Although current risk prediction algorithms exist to identify women at increased 10-year risk of cardiovascular disease (CVD), clinicians and researchers have been interested in developing novel biomarkers that might improve predictive accuracy further. These biomarkers have led to important insights into the pathophysiology of CVD, but results for their ability to improve prediction or guide preventive therapy have been mixed. The incidence of CVD is lower in women than men, and the effects of a number of traditional biomarkers on CVD risk differ in women compared to men. Both of these factors influence the ability to accurately predict CVD risk. CONTENT We review the distinctive aspects of CVD risk prediction in women, discuss the statistical challenges to improved risk prediction, and discuss a number of biomarkers in varying stages of development with a range of performance in prediction. SUMMARY A variety of biomarkers from different pathophysiologic pathways have been evaluated for improving CVD risk. While many have been incompletely studied or have not been shown to improve risk prediction in women, others, such as high-sensitivity troponin T, have shown promise in improving risk prediction. Increasing inclusion of women in CVD studies will be crucial to providing opportunities to evaluate future biomarkers.

scholarly journals An explainable XGBoost–based approach towards assessing the risk of cardiovascular disease in patients with Type 2 Diabetes Mellitus

2020 ◽  
Author(s):  
Maria Athanasiou ◽  
Konstantina Sfrintzeri ◽  
Konstantia Zarkogianni ◽  
Anastasia Thanopoulou ◽  
Konstantina S. Nikita
Keyword(s):  
Diabetes Mellitus ◽  
Cardiovascular Disease ◽  
Risk Prediction ◽  
Prediction Models ◽  
Therapeutic Interventions ◽  
Gradient Boosting ◽  
Cvd Risk ◽  
Medical Visits ◽  
Extreme Gradient Boosting

<div> <div> <div> <p>Cardiovascular Disease (CVD) is an important cause of disability and death among individuals with Diabetes Mellitus (DM). International clinical guidelines for the management of Type 2 DM (T2DM) are founded on primary and secondary prevention and favor the evaluation of CVD related risk factors towards appropriate treatment initiation. CVD risk prediction models can provide valuable tools for optimizing the frequency of medical visits and performing timely preventive and therapeutic interventions against CVD events. The integration of explainability modalities in these models can enhance human understanding on the reasoning process, maximize transparency and embellish trust towards the models’ adoption in clinical practice. The aim of the present study is to develop and evaluate an explainable personalized risk prediction model for the fatal or non-fatal CVD incidence in T2DM individuals. An explainable approach based on the eXtreme Gradient Boosting (XGBoost) and the Tree SHAP (SHapley Additive exPlanations) method is deployed for the calculation of the 5-year CVD risk and the generation of individual explanations on the model’s decisions. Data from the 5- year follow up of 560 patients with T2DM are used for development and evaluation purposes. The obtained results (AUC=71.13%) indicate the potential of the proposed approach to handle the unbalanced nature of the used dataset, while providing clinically meaningful insights about the ensemble model’s decision process. </p> </div> </div> </div>

scholarly journals An assessment of the potential miscalibration of cardiovascular disease risk predictions caused by a secular trend in cardiovascular disease in England

2020 ◽  
Vol 20 (1) ◽  
Author(s):  
Alexander Pate ◽  
Tjeerd van Staa ◽  
Richard Emsley
Keyword(s):  
Cardiovascular Disease ◽  
Risk Prediction ◽  
Structural Model ◽  
Validation Cohort ◽  
Secular Trend ◽  
Marginal Structural Model ◽  
Cvd Risk ◽  
Model Framework ◽  
Statin Use

Abstract Background A downwards secular trend in the incidence of cardiovascular disease (CVD) in England was identified through previous work and the literature. Risk prediction models for primary prevention of CVD do not model this secular trend, this could result in over prediction of risk for individuals in the present day. We evaluate the effects of modelling this secular trend, and also assess whether it is driven by an increase in statin use during follow up. Methods We derived a cohort of patients (1998–2015) eligible for cardiovascular risk prediction from the Clinical Practice Research Datalink with linked hospitalisation and mortality records (N = 3,855,660). Patients were split into development and validation cohort based on their cohort entry date (before/after 2010). The calibration of a CVD risk prediction model developed in the development cohort was tested in the validation cohort. The calibration was also assessed after modelling the secular trend. Finally, the presence of the secular trend was evaluated under a marginal structural model framework, where the effect of statin treatment during follow up is adjusted for. Results Substantial over prediction of risks in the validation cohort was found when not modelling the secular trend. This miscalibration could be minimised if one was to explicitly model the secular trend. The reduction in risk in the validation cohort when introducing the secular trend was 35.68 and 33.24% in the female and male cohorts respectively. Under the marginal structural model framework, the reductions were 33.31 and 32.67% respectively, indicating increasing statin use during follow up is not the only the cause of the secular trend. Conclusions Inclusion of the secular trend into the model substantially changed the CVD risk predictions. Models that are being used in clinical practice in the UK do not model secular trend and may thus overestimate the risks, possibly leading to patients being treated unnecessarily. Wider discussion around the modelling of secular trends in a risk prediction framework is needed.

Sign in / Sign up

Export Citation Format

Share Document