Search

Background: Understanding the predictive utility of previously derived polygenic risk scores (PRSs) for long-term risk of coronary heart disease (CHD) and its additive value beyond traditional risk factors can inform prevention strategies. Methods: Data from adults 20 to 59 years of age who were free of CHD from the FOS (Framingham Offspring Study) and the ARIC (Atherosclerosis Risk in Communities) study were analyzed. Because the PRS was derived from samples of predominantly European ancestry, individuals who self-reported White race were included. The sample was stratified by age and cohort: young (FOS, 20–39 years [median, 30 years] of age), early midlife (FOS, 40–59 years [median, 43] years of age), and late midlife (ARIC, 45–59 years [median, 52 years] of age). Two previously derived and validated prediction tools were applied: (1) a 30-year traditional risk factor score and (2) a genome-wide PRS comprising >6 million genetic variants. Hazard ratios for the association between each risk estimate and incident CHD were calculated. Predicted and observed rates of CHD were compared to assess discrimination for each model individually and together with the optimism-corrected C index (95% CI). Results: Among 9757 participants, both the traditional risk factor score (hazard ratio per 1 SD, 2.60 [95% CI, 2.08–3.27], 2.09 [95% CI, 1.83–2.40], and 2.11 [95% CI, 1.96–2.28]) and the PRS (hazard ratio, 1.98 [95% CI, 1.70–2.30], 1.64 [95% CI, 1.47–1.84], and 1.22 [95% CI, 1.15–1.30]) were significantly associated with incident CHD in young, early midlife, and late midlife, respectively. Discrimination was similar or better for the traditional risk factor score (C index, 0.74 [95% CI, 0.70–0.78], 0.70 [95% CI, 0.67–0.72], and 0.72 [95% CI, 0.70–0.73]) compared with an age- and sex-adjusted PRS (0.73 [95% CI, 0.69–0.78], 0.66 [95% CI, 0.62–0.69], and 0.66 [95% CI, 0.64–0.67]) in young, early-midlife, and late-midlife participants, respectively. The ΔC index when PRS was added to the traditional risk factor score was 0.03 (95% CI, 0.001–0.05), 0.02 (95% CI, −0.002 to 0.037), and 0.002 (95% CI, −0.002 to 0.006) in young, early-midlife, and late-midlife participants, respectively. Conclusions: Despite a statistically significant association between PRS and 30-year risk of CHD, the C statistic improved only marginally with the addition of PRS to the traditional risk factor model among young adults and did not improve among midlife adults. PRS, an immutable factor that cannot be directly intervened on, has minimal clinical utility for long-term CHD prediction when added to a traditional risk factor model.

This study examines the risk of new-onset diabetes in patients with hypertriglyceridaemic hyperapolipoprotein B (high triglycerides, high apolipoprotein B [apoB], low LDL cholesterol to apoB ratio, and low HDL cholesterol). The aim was to establish whether this lipoprotein phenotype identified a substantial group at high risk of developing diabetes over the next 20 years.In this prospective, longitudinal, observational cohort study, we used data from the Framingham Offspring cohort (recruited in Framingham, MA, USA). Participants were aged 40-69 years and free of diabetes and cardiovascular disease at a baseline examination done between April, 1987, and November, 1991, and were followed up until March, 2014. Cox proportional hazards regression with hierarchical adjustment for age and sex, waist circumference, and fasting blood glucose were used to model the relationship between each lipid marker and incident diabetes, as well as the relationship between hypertriglyceridaemic hyperapoB (defined as values greater than sample medians of triglycerides and apoB, and less than medians of HDL cholesterol and LDL cholesterol to apoB ratio) and incident diabetes.Of 3446 individuals aged 40-69 years who completed baseline examination, 2515 participants were eligible and included in all analyses. During median 21·1 years (IQR 11·1-23·1) of follow-up, 402 (16·0%) individuals developed diabetes. Age (p=0·032), waist circumference (p0·0001), fasting blood glucose (p0·0001), and natural logarithm-transformed triglycerides (p0·0001) were associated with new-onset diabetes, as were apoB (p=0·0016), LDL cholesterol to apoB ratio (p=0·0018), and HDL cholesterol (p=0·0016) when added to this model. The age and sex-adjusted incidence of diabetes in the hypertriglyceridaemic hyperapoB group was 32·4% (95% CI 27·8-37·7) versus 5·5% (3·5-8·6) in the optimal lipid phenotype group and 15·5% (13·5-17·7) in the mixed lipid phenotype group. The fully adjusted hazard ratio, including glucose and waist circumference, for individuals with hypertriglyceridaemic hyperapoB was 3·30 (95% CI 2·06-5·30; p=0·0008) and for mixed lipid phenotype was 2·17 (1·38-3·40; p0·0001) compared with those with the optimal lipid phenotype.Our findings suggest that individuals with hypertriglyceridaemic hyperapoB are at high risk of new-onset diabetes and might benefit from intensive measures to prevent diabetes. The association between this phenotype and incident diabetes is consistent with a pro-diabetic effect due to increased clearance of apoB particles from plasma, which could injure pancreatic islet cells. This mechanism might explain the increased risk of diabetes with statin therapy.Doggone Foundation.

Aims The aim of our study was to analyse the benefits of a 9 week hybrid comprehensive telerehabilitation (HCTR) programme in heart failure (HF) patients according to aetiology, as a subanalysis of the Telerehabilitation in Heart Failure Patients (TELEREH‐HF) trial. Methods and results Overall, 555 (65.3%) patients with ischaemic (IS) and 295 (34.7%) patients with non‐ischaemic (NIS) HF aetiology were randomized. There were no differences between the effect of HCTR and usual care (UC) on the primary outcome of number of days alive and out of the hospital in 26 months from the time of randomization in either aetiology (Wilcoxon–Mann–Whitney test), and no heterogeneity of effect between the aetiologies was noted (van Elteren test, P = 0.746). In Cox proportional hazards regression analysis, treatment was not independently associated with the secondary outcomes. For all‐cause mortality, the adjusted hazard ratio for HCTR vs. UC was 0.90 (95% confidence interval, 0.54–1.51) in IS and 1.42 (95% confidence interval, 0.69–2.94) in NIS (P interaction = 0.316). Differences between HCTR and UC in terms of change in the 6 min walk test distance and cardiopulmonary exercise test time after 9 weeks reached statistical significance in the IS arm (P = 0.015 and P

Introduction This study examines the dose–response relationship between moderate-to-vigorous physical activity and cardiometabolic measures in adolescents. Methods Cross-sectional spline analyses were performed using 2003–2016 National Health and Nutrition Examination Survey data among adolescents (aged 12–19 years, N=9,195) on objectively measured (2003–2006) and self-reported (2007–2016) weekly mean minutes of moderate-to-vigorous physical activity and cardiometabolic measures (systolic and diastolic blood pressure, total cholesterol, high-density lipoprotein, BMI, and cardiorespiratory fitness). Inflection points were determined for nonlinear relationships. Results For objective moderate-to-vigorous physical activity, female adolescents had significant nonlinear associations with inflection points at 90 minutes/week for BMI percentile and systolic blood pressure. Male adolescents had inflection points at 150 weekly minutes of objective activity for BMI percentile and cardiorespiratory fitness. BMI percentile was about 7% lower for female and male adolescents at 150 weekly minutes of objectively measured moderate-to-vigorous physical activity than at 0 minutes. For self-reported moderate-to-vigorous physical activity, inflection points were at 375 minutes/week (diastolic blood pressure for female adolescents) and 500 minutes/week (systolic blood pressure for male adolescents). Conclusions Among several significant dose–response relationships between physical activity and cardiometabolic health in adolescents, consistent and often nonlinear relationships were identified for BMI, with inflection points at 90–150 minutes of objective moderate-to-vigorous physical activity. Notable differences in associations and linearity were identified by sex and physical activity measure (objective or self-reported). These results support calls for any increase in physical activity among adolescents and suggest that recommendations closer to the adult guidelines of 150 weekly minutes of physical activity may be health promoting and more attainable for youth than the current recommendation of 420 weekly minutes.

Background: Lipid-lowering recommendations for prevention of atherosclerotic cardiovascular disease rely principally on estimated 10-year risk. We sought to determine the optimal time for initiation of lipid lowering in younger adults as a function of expected 30-year benefit. Methods: Data from 3148 National Health and Nutrition Examination Survey (2009–2016) participants, age 30 to 59 years, not eligible for lipid-lowering treatment recommendation under the most recent US guidelines, were analyzed. We estimated the absolute and relative impact of lipid lowering as a function of age, age at initiation, and non–high-density lipoprotein cholesterol (HDL-C) level on the expected rates of atherosclerotic cardiovascular disease over the succeeding 30 years. We modeled expected risk reductions based on shorter-term effects observed in statin trials (model A) and longer-term benefits based on Mendelian randomization studies (model B). Results: In both models, potential reductions in predicted 30-year atherosclerotic cardiovascular disease risk were greater with older age and higher non–HDL-C level. Immediate initiation of lipid lowering (ie, treatment for 30 years) in 40- to 49-year-old patients with non–HDL-C ≥160 mg/dL would be expected to reduce their average predicted 30-year risk of 17.1% to 11.6% (model A; absolute risk reduction [ARR], 5.5%) or 6.5% (model B; ARR 10.6%). Delaying lipid lowering by 10 years (treatment for 20 years) would result in residual 30-year risk of 12.7% (A; ARR 4.4) or 9.9% (B; ARR 7.2%) and delaying by 20 years (treatment for 10 years) would lead to expected mean residual risk of 14.6% (A; ARR 2.6%) or 13.9% (B; ARR 3.2%). The slope of the achieved ARR as a function of delay in treatment was also higher with older age and higher non–HDL-C level. Conclusions: Substantial reduction in expected atherosclerotic cardiovascular disease risk in the next 30 years is achievable by intensive lipid lowering in individuals in their 40s and 50s with non–HDL-C ≥160 mg/dL. For many, the question of when to start lipid lowering might be more relevant than whether to start lipid lowering.

Prediction Models Prediction models’ newfound importance and the emergence of model development based on machine learning raise questions about how to ensure their safety and efficacy, given their ...

Background Medicare insurance claims may provide an efficient means to ascertain follow-up of older participants in clinical research. We sought to determine the accuracy and completeness of claims- versus site-based follow-up with clinical event committee (+CEC) adjudication of cardiovascular outcomes. Methods We performed a retrospective study using linked Medicare and Duke Database of Clinical Trials data. Medicare claims were linked to clinical data from 7 randomized cardiovascular clinical trials. Of 52,476 trial participants, linking resulted in 5,839 (of 10,497 linkage-eligible) Medicare-linked trial participants with fee-for-service A and B coverage. Death, myocardial infarction (MI), stroke, and revascularization incidences were compared using Medicare inpatient claims only, site-reported events (+CEC) only, or a combination of the 2. Randomized treatment effects were compared as a function of whether claims-based, site-based (+CEC), or a combined system was used for event detection. Results Among the 5,839 study participants, the annual event rates were similar between claims- and site-based (+CEC) follow-up: death (overall rate 5.2% vs 5.2%; adjusted κ 0.99), MI (2.2% vs 2.3%; adjusted κ 0.96), stroke (0.7% vs 0.7%; adjusted κ 0.99), and any revascularization (7.4% vs 7.9%; adjusted κ 0.95). Of events detected by claims yet not reported by CEC, a minority were reported by sites but negatively adjudicated by CEC (39% of MIs and 18% of strokes). Differences in individual case concordance led to higher event rates when claims- and site-based (+CEC) systems were combined. Randomized treatment effects were similar among the 3 approaches for each outcome of interest. Conclusions Claims- versus site-based (+CEC) follow-up identified similar overall cardiovascular event rates despite meaningful differences in the events detected. Randomized treatment effects were similar using the 2 methods, suggesting claims data could be used to support clinical research leveraging routinely collected data. This approach may lead to more effective evidence generation, synthesis, and appraisal of medical products and inform the strategic approaches toward the National Evaluation System for Health Technology.

Background There is a paucity of data on the distribution of cardiovascular risk factors in patients with familial hypercholesterolemia (FH) as compared to the general population. The aim of the study was to compare cardiovascular risk factors in a cohort of FH patients to the representative sample of adults in Poland who represent a high–cardiovascular risk European region. Methods We compared the distribution of risk factors in 1,382 individuals with FH phenotype referred for genetic testing between 2006 and 2014 to the National Centre of Familial Hypercholesterolemia in Gdansk, Poland. The cohort was comprised of 637 positive FH(+) and 745 negative FH(−) patients who were compared to a nationally representative sample of 2,413 adults age 18-79, standardized by age and sex, from the NATPOL 2011 study (NATPOL). We analyzed patients' distribution of history of atherosclerotic cardiovascular disease (ASCVD) and standard risk factors including total cholesterol, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol, triglycerides, systolic and diastolic blood pressure (SBP, DBP), body mass index, smoking, and diabetes. Results FH(+) patients (mean age 45.6 years) had the highest LDL-C of 241.7 mg/dL (95% CI 234.8-248.5) compared to 206.1 mg/dL (200.5-211.7) in FH(−) patients (mean age 48.2) and 126.2 mg/dL (124.8-127.6) in NATPOL. Mean SBP was the lowest in FH(+) patients at 128.7 mm Hg (126.7-130.7) compared to 133.4 mm Hg (132.6-134.3) in NATPOL and 134.4 mm Hg (132.3-136.5) in FH(−). No differences were found in the prevalence of diabetes and body mass index. Smoking was less common in FH(+) at 12.4% (9.4-15.4) compared to both FH(−) and NATPOL: 20.4% (16.6-24.1) and 28.4% (26.6-30.2), respectively. The prevalence of individuals with a history of ASCVD in both FH(+) and FH(−) was nearly 3-fold higher compared to NATPOL: 26% (21.8-30.1) and 26.6% (22.2-30.9) versus 9.5% (8.3-10.7), respectively. Conclusions The FH(+) patients had significantly higher mean LDL-C, but the levels of nonlipid factors were lower or similar compared to the other groups. Both FH(+) and FH(−) were characterized by a heavy burden of ASCVD. This suggests that cholesterol, and no other risk factors, is a key contributor to cardiovascular risk in patients with FH, especially those with genetic mutation.

Background Guidelines recommend exercise training as a component of heart failure (HF) management. There are large disparities in access to rehabilitation and introducing hybrid comprehensive telerehabilitation (TR) consisting of remote monitoring of training in patients' homes might be an optimal solution in Poland. Purpose The primary objective of the TELEREH-HF trial is to determine whether introducing TR will significantly increase days alive and out of hospital compared with usual care. The secondary objectives including assessment the effects of TR compared to usual care on all-cause and cardiovascular mortality and all-cause, cardiovascular and HF hospitalization. The tertiary analyses include: evaluation of the safety, effectiveness, quality of life, depression, anxiety, patients' acceptance of and adherence to TR. Methods The TELEREH-HF study is a randomized, multi-center, prospective, open-label, parallel group controlled trial in 850 HF patients after a hospitalization incident in NYHA I-III and LVEF≤40%. Patients were randomized to TR + usual care (TR group) or to usual care only (control group) and are followed for a maximum of 24 months. The TR group patients underwent a 9-week TR program consisting of an initial stage (1 week) conducted at hospital and a basic stage (8-week) home-based TR five times weekly. Results All patients were randomized and completed initial intervention in the TR group. The follow up of both groups is in progress. Conclusion The TELEREH-HF trial will provide novel data on the effects of telerhabilitation on hospitalization and mortality in HF patients, and on safety, qaulity of life, depression, anxiety and acceptance of and adherence to this intervention.

There is a need to identify high-risk features that predict early-onset atherosclerotic cardiovascular disease (ASCVD). The authors provide insights to help clinicians identify and address high-risk conditions in the 20- to 39-year age range (young adults). These include tobacco use, elevated blood pressure/hypertension, family history of premature ASCVD, primary severe hypercholesterolemia such as familial hypercholesterolemia, diabetes with diabetes-specific risk-enhancing factors, or the presence of multiple other risk-enhancing factors, including in females, a history of pre-eclampsia or menopause under age 40. The authors update current thinking on lipid risk factors such as triglycerides, non-high-density lipoprotein cholesterol, apolipoprotein B, or lipoprotein (a) that are useful in understanding an individual's long-term ASCVD risk. The authors review emerging strategies, such as coronary artery calcium and polygenic risk scores in this age group, that have potential clinical utility, but whose best use remains uncertain. Finally, the authors discuss both the obstacles and opportunities for addressing prevention in early adulthood.

Background Exercise training in heart failure (HF) patients should be monitored to ensure patients’ safety. Electrocardiographic (ECG) telemonitoring was used to assess the safety of hybrid comprehensive telerehabilitation (HCTR). Objective Analysis of ECG recorded during HCTR in HF patients. Methods The TELEREH‐HF multicenter, randomized, controlled trial enrolled 850 HF patients with New York Heart Association class I‐III and left ventricular ejection fraction of ≤40%. This subanalysis focuses on 386 patients (aged 62 ± 11 years, LVEF 31 ± 7%) randomized to HCTR. HCTR was telemonitored with a device allowing to record 16‐s fragments of ECG and to transmit the data via mobile phone network to the monitoring center. ResultsIn 386 patients, 16,622 HCTR sessions were recorded and 66,488 ECGs fragments were evaluated. Sinus rhythm was present in 320 (83%) and permanent atrial fibrillation (AF) in 66 (17%) patients, respectively. The most common arrhythmias were ventricular and atrial premature beats, recorded in 76.4% and 27.7% of the patients, respectively. Non‐sustained ventricular tachycardia (21 episodes in 8 patients) and paroxysmal AF episodes (6 in 4 patients) were rare. None of the analyzed demographic and clinical characteristics was predictive for onset of the new arrhythmias on exercise. Conclusion Telerehabilitation in HF patients was safe without the evidence for symptomatic arrhythmias requiring discontinuation of telerehabilitation. Only one mildly symptomatic paroxysmal AF episode led to the short‐term suspension of the training program. The most common arrhythmias were atrial and ventricular premature beats. These arrhythmias did not result in any changes in rehabilitation and therapy regimens.

BACKGROUND: The appropriate dose of aspirin to lower the risk of death, myocardial infarction, and stroke and to minimize major bleeding in patients with established atherosclerotic cardiovascular disease is a subject of controversy. METHODS: Using an open-label, pragmatic design, we randomly assigned patients with established atherosclerotic cardiovascular disease to a strategy of 81 mg or 325 mg of aspirin per day. The primary effectiveness outcome was a composite of death from any cause, hospitalization for myocardial infarction, or hospitalization for stroke, assessed in a time-to-event analysis. The primary safety outcome was hospitalization for major bleeding, also assessed in a time-to-event analysis. RESULTS: A total of 15,076 patients were followed for a median of 26.2 months (interquartile range [IQR], 19.0 to 34.9). Before randomization, 13,537 (96.0% of those with available information on previous aspirin use) were already taking aspirin, and 85.3% of these patients were previously taking 81 mg of daily aspirin. Death, hospitalization for myocardial infarction, or hospitalization for stroke occurred in 590 patients (estimated percentage, 7.28%) in the 81-mg group and 569 patients (estimated percentage, 7.51%) in the 325-mg group (hazard ratio, 1.02; 95% confidence interval [CI], 0.91 to 1.14). Hospitalization for major bleeding occurred in 53 patients (estimated percentage, 0.63%) in the 81-mg group and 44 patients (estimated percentage, 0.60%) in the 325-mg group (hazard ratio, 1.18; 95% CI, 0.79 to 1.77). Patients assigned to 325 mg had a higher incidence of dose switching than those assigned to 81 mg (41.6% vs. 7.1%) and fewer median days of exposure to the assigned dose (434 days [IQR, 139 to 737] vs. 650 days [IQR, 415 to 922]). CONCLUSIONS: In this pragmatic trial involving patients with established cardiovascular disease, there was substantial dose switching to 81 mg of daily aspirin and no significant differences in cardiovascular events or major bleeding between patients assigned to 81 mg and those assigned to 325 mg of aspirin daily. (Funded by the Patient-Centered Outcomes Research Institute; ADAPTABLE ClinicalTrials.gov number, NCT02697916.).

Background Type 2 diabetes mellitus (DM) is one of the most common comorbidities among patients with heart failure (HF) with reduced ejection fraction (HFrEF). There are limited data regarding efficacy of hybrid comprehensive telerehabilitation (HCTR) on cardiopulmonary exercise capacity in patients with HFrEF with versus those without diabetes. Aim The aim of the present study was to analyze effects of 9-week HCTR in comparison to usual care on parameters of cardiopulmonary exercise capacity in HF patients according to history of DM. Methods Clinically stable HF patients with left ventricular ejection fraction [LVEF] Results CPET was performed in 385 patients assigned to HCTR group: 129 (33.5%) had DM (HCTR-DM group) and 256 patients (66.5%) did not have DM (HCTR-nonDM group). Among 397 patients assigned to UC group who had CPET: 137 (34.5%) had DM (UC-DM group) and 260 patients (65.5%) did not have DM (UC-nonDM group). Among DM patients, differences in cardiopulmonary parameters from baseline to 9 weeks remained similar among HCTR and UC patients. In contrast, among patients without DM, HCTR was associated with greater 9-week changes than UC in exercise time, which resulted in a statistically significant interaction between patients with and without DM: difference in changes in exercise time between HCTR versus UC was 12.0 s [95% CI − 15.1, 39.1 s] in DM and 43.1 s [95% CI 24.0, 63.0 s] in non-DM, interaction p-value = 0.016. Furthermore, statistically significant differences in the effect of HCTR versus UC between DM and non-DM were observed in ventilation at rest: − 0.34 l/min [95% CI − 1.60, 0.91 l/min] in DM and 0.83 l/min [95% CI − 0.06, 1.73 l/min] in non-DM, interaction p value = 0.0496 and in VE/VCO2 slope: 1.52 [95% CI − 1.55, 4.59] for DM vs. − 1.44 [95% CI − 3.64, 0.77] for non-DM, interaction p value = 0.044. Conclusions The benefits of hybrid comprehensive telerehabilitation versus usual care on the improvement of physical performance, ventilatory profile and gas exchange parameters were more pronounced in patients with HFrEF without DM as compared to patients with DM. Trial registration: ClinicalTrials.gov Identifier: NCT02523560. Registered 3rd August 2015. https://clinicaltrials.gov/ct2/show/NCT02523560?term=NCT02523560&draw=2&rank=1. Other Study ID Numbers: STRATEGME1/233547/13/NCBR/2015

When comparing performances of two risk prediction models, several metrics exist to quantify prognostic improvement, including the change in the area under the Receiver Operating Characteristic curve, the Integrated Discrimination Improvement, the Net Reclassification Index at event rate, the change in Standardized Net Benefit, the change in Brier score, and the change in scaled Brier score. We explore the behavior and interrelationships between these metrics under multivariate normality in nested and nonnested model comparisons. We demonstrate that, within the framework of linear discriminant analysis, all six statistics are functions of squared Mahalanobis distance, a robust metric that properly measures discrimination by quantifying the separation between the risk scores of events and nonevents. These relationships are important for overall interpretability and clinical usefulness. Through simulation, we demonstrate that the performance of the theoretical estimators under normality is comparable or superior to empirical estimation methods typically used by investigators. In particular, the theoretical estimators for the Net Reclassification Index and the change in Standardized Net Benefit exhibit less variability in their estimates as compared to their empirically estimated counterparts. Finally, we explore how these metrics behave with potentially nonnormal data by applying these methods in a practical example based on the sex-specific cardiovascular disease risk models from the Framingham Heart Study. Our findings aim to give greater insight into the behavior of these measures and the connections existing among them and to provide additional estimation methods with less variability for the Net Reclassification Index and the change in Standardized Net Benefit.

Supplemental Digital Content is available in the text., Background: To optimize preventive strategies for coronary heart disease (CHD), it is essential to understand and appropriately quantify the contribution of its key risk factors. Our objective was to compare the associations of key modifiable CHD risk factors—specifically lipids, systolic blood pressure (SBP), diabetes mellitus, and smoking—with incident CHD events based on their prognostic performance, attributable risk fractions, and treatment benefits, overall and by age. Methods: Pooled participant-level data from 4 observational cohort studies sponsored by the National Heart, Lung, and Blood Institute were used to create a cohort of 22 626 individuals aged 45 to 84 years who were initially free of cardiovascular disease. Individuals were followed for 10 years from baseline evaluation for incident CHD. Proportional hazards regression was used to estimate metrics of prognostic model performance (likelihood ratio, C index, net reclassification, discrimination slope), hazard ratios, and population attributable fractions for SBP, non–high-density lipoprotein cholesterol (non–HDL-C), diabetes mellitus, and smoking. Expected absolute risk reductions for antihypertensive and lipid-lowering treatment were assessed. Results: Age, sex, and race capture 63% to 80% of the prognostic performance of cardiovascular risk models. In contrast, adding either SBP, non–HDL-C, diabetes mellitus, or smoking to a model with other risk factors increases the C index by only 0.004 to 0.013. However, primordial prevention could have a substantial effect as demonstrated by population attributable fractions of 28% for SBP≥130 mm Hg and 17% for non–HDL-C≥130 mg/dL. Similarly, lowering the SBP of all individuals to

The last two decades have witnessed an explosion in research focused on the development and assessment of novel biomarkers for improved prognosis of diseases. As a result, best practice standards guiding biomarker research have undergone extensive development. Currently, there is great interest in the promise of biomarkers to enhance research efforts and clinical practice in the setting of chronic kidney disease, acute kidney injury, and glomerular disease. However, some have questioned whether biomarkers currently add value to the clinical practice of nephrology. The current state of the art pertaining to statistical analyses regarding the use of such measures is critical. In December 2014, the National Institute of Diabetes and Digestive and Kidney Diseases convened a meeting, "Toward Building Better Biomarker Statistical Methodology," with the goals of summarizing the current best practice recommendations and articulating new directions for methodological research. This report summarizes its conclusions and describes areas that need attention. Suggestions are made regarding metrics that should be commonly reported. We outline the methodological issues related to traditional metrics and considerations in prognostic modeling, including discrimination and case mix, calibration, validation, and cost-benefit analysis. We highlight the approach to improved risk communication and the value of graphical displays. Finally, we address some "new frontiers" in prognostic biomarker research, including the competing risk framework, the use of longitudinal biomarkers, and analyses in distributed research networks.

IMPORTANCE: Accurate prediction of adverse outcomes after acute myocardial infarction (AMI) can guide the triage of care services and shared decision-making, and novel methods hold promise for using existing data to generate additional insights. OBJECTIVE: To evaluate whether contemporary machine learning methods can facilitate risk prediction by including a larger number of variables and identifying complex relationships between predictors and outcomes. DESIGN, SETTING, AND PARTICIPANTS: This cohort study used the American College of Cardiology Chest Pain-MI Registry to identify all AMI hospitalizations between January 1, 2011, and December 31, 2016. Data analysis was performed from February 1, 2018, to October 22, 2020. MAIN OUTCOMES AND MEASURES: Three machine learning models were developed and validated to predict in-hospital mortality based on patient comorbidities, medical history, presentation characteristics, and initial laboratory values. Models were developed based on extreme gradient descent boosting (XGBoost, an interpretable model), a neural network, and a meta-classifier model. Their accuracy was compared against the current standard developed using a logistic regression model in a validation sample. RESULTS: A total of 755 402 patients (mean [SD] age, 65 [13] years; 495 202 [65.5%] male) were identified during the study period. In independent validation, 2 machine learning models, gradient descent boosting and meta-classifier (combination including inputs from gradient descent boosting and a neural network), marginally improved discrimination compared with logistic regression (C statistic, 0.90 for best performing machine learning model vs 0.89 for logistic regression). Nearly perfect calibration in independent validation data was found in the XGBoost (slope of predicted to observed events, 1.01; 95% CI, 0.99-1.04) and the meta-classifier model (slope of predicted-to-observed events, 1.01; 95% CI, 0.99-1.02), with more precise classification across the risk spectrum. The XGBoost model reclassified 32 393 of 121 839 individuals (27%) and the meta-classifier model reclassified 30 836 of 121 839 individuals (25%) deemed at moderate to high risk for death in logistic regression as low risk, which were more consistent with the observed event rates. CONCLUSIONS AND RELEVANCE: In this cohort study using a large national registry, none of the tested machine learning models were associated with substantive improvement in the discrimination of in-hospital mortality after AMI, limiting their clinical utility. However, compared with logistic regression, XGBoost and meta-classifier models, but not the neural network, offered improved resolution of risk for high-risk individuals.

The PATH (Predictive Approaches to Treatment effect Heterogeneity) Statement was developed to promote the conduct of, and provide guidance for, predictive analyses of heterogeneous treatment effects (HTE) in clinical trials. The goal of predictive HTE analysis is to provide patient-centered estimates of outcome risks with versus without the intervention, taking into account all relevant patient attributes simultaneously, to support more personalized clinical decision making than can be made based only on an overall average treatment effect. We distinguished two categories of predictive HTE approaches (a “risk modeling” and an “effect modeling” approach) and developed four sets of guidance statements: 1) criteria to determine when risk modeling approaches are likely to identify clinically meaningful HTE; 2) methodological aspects of risk modeling methods; 3) considerations for translation to clinical practice; and 4) considerations and caveats in the use of effect modeling approaches. We also discuss limitations of these methods and enumerate research priorities for advancing methods designed to generate more personalized evidence. This explanation and elaboration document describes the intent and rationale of each recommendation; and discusses related analytic considerations, caveats, and reservations.

IMPORTANCE: The extent to which levels of high-sensitivity C-reactive protein (hs-CRP), a known marker of increased cardiovascular risk, are elevated and are associated with standard cardiovascular risk factors in patients with a history of myocardial infarction (MI) is unknown. OBJECTIVES: To determine the pattern and determinants of the distribution of hs-CRP in those with a prior MI in the United States using a nationally representative sample. DESIGN AND PARTICIPANTS: Adults with hs-CRP data in the National Health and Nutrition Examination Surveys from 1999–2010. RESULTS: Among 1296 individuals in our cohort, the median age was 65 years and the median hs-CRP level was 2.69 mg/L, measured an average of 7.1 years after the MI. Among these patients, 22% had hs-CRP levels of

Supplemental Digital Content is available in the text., Background: Total cholesterol and high-density lipoprotein cholesterol (HDL-C) measurements are central to cardiovascular disease (CVD) risk assessment, but there is continuing debate around the utility of other lipids for risk prediction. Methods: Participants from UK Biobank without baseline CVD and not taking statins, with relevant lipid measurements (n=346 686), were included in the primary analysis. An incident fatal or nonfatal CVD event occurred in 6216 participants (1656 fatal) over a median of 8.9 years. Associations of nonfasting lipid measurements (total cholesterol, HDL-C, non–HDL-C, direct and calculated low-density lipoprotein cholesterol [LDL-C], and apolipoproteins [Apo] A1 and B) with CVD were compared using Cox models adjusting for classical risk factors, and predictive utility was determined by the C-index and net reclassification index. Prediction was also tested in 68 649 participants taking a statin with or without baseline CVD (3515 CVD events). Results: ApoB, LDL-C, and non–HDL-C were highly correlated (r>0.90), while HDL-C was strongly correlated with ApoA1 (r=0.92). After adjustment for classical risk factors, 1 SD increase in ApoB, direct LDL-C, and non–HDL-C had similar associations with composite fatal/nonfatal CVD events (hazard ratio, 1.23, 1.20, 1.21, respectively). Associations for 1 SD increase in HDL-C and ApoA1 were also similar (hazard ratios, 0.81 [both]). Adding either total cholesterol and HDL-C, or ApoB and ApoA, to a CVD risk prediction model (C-index, 0.7378) yielded similar improvement in discrimination (C-index change, 0.0084; 95% CI, 0.0065, 0.0104, and 0.0089; 95% CI, 0.0069, 0.0109, respectively). Once total and HDL-C were in the model, no further substantive improvement was achieved with the addition of ApoB (C-index change, 0.0004; 95% CI, 0.0000, 0.0008) or any measure of LDL-C. Results for predictive utility were similar for a fatal CVD outcome, and in a discordance analysis. In participants taking a statin, classical risk factors (C-index, 0.7118) were improved by non–HDL-C (C-index change, 0.0030; 95% CI, 0.0012, 0.0048) or ApoB (C-index change, 0.0030; 95% CI, 0.0011, 0.0048). However, adding ApoB or LDL-C to a model already containing non–HDL-C did not further improve discrimination. Conclusions: Measurement of total cholesterol and HDL-C in the nonfasted state is sufficient to capture the lipid-associated risk in CVD prediction, with no meaningful improvement from addition of apolipoproteins, direct or calculated LDL-C.

IMPORTANCE: Guidelines recommend exercise training as a component of heart failure management. There are large disparities in access to rehabilitation, and introducing hybrid comprehensive telerehabilitation (HCTR) consisting of remote monitoring of training at patients’ homes might be an appealing alternative. OBJECTIVE: To assess whether potential improvements in quality-of-life outcomes after a 9-week HCTR intervention in patients with heart failure translate into improvement in clinical outcomes during extended 12 to 24 months of follow-up, compared with usual care. DESIGN, SETTING, AND PARTICIPANTS: The Telerehabilitation in Heart Failure Patients (TELEREH-HF) trial is a multicenter, prospective, open-label, parallel-group randomized clinical trial that enrolled 850 patients with heart failure up to 6 months after a cardiovascular hospitalization with New York Heart Association levels I, II, or III and left ventricular ejection fraction of 40% or less. Patients from 5 centers in Poland were randomized 1:1 to HCTR plus usual care or usual care only and followed up for 14 to 26 months after randomization. INTERVENTIONS: During the first 9 weeks, patients underwent either an HCTR program (1 week in hospital and 8 weeks at home) or usual care with observation. The HCTR intervention encompassed telecare, telerehabilitation, and remote monitoring of implantable devices. No intervention occurred in the remaining study period. MAIN OUTCOMES AND MEASURES: The percentage of days alive and out of the hospital from randomization through the end of follow-up at 14 to 26 months. RESULTS: A total of 850 patients were enrolled, with 425 randomized to the HCTR group (377 male patients [88.7%]; mean [SD] age, 62.6 [10.8] years) and 425 randomized to usual care (376 male patients [88.5%]; mean [SD] age, 62.2 [10.2] years). The HCTR intervention did not extend the percentage of days alive and out of the hospital. The mean (SD) days were 91.9 (19.3) days in the HCTR group vs 92.8 (18.3) days in the usual-care group, with the probability that HCTR extends days alive and out of the hospital equal to 0.49 (95% CI, 0.46-0.53; P = .74) vs usual care. During follow-up, 54 patients died in the HCTR arm and 52 in the usual-care arm, with mortality rates at 26 months of 12.5% vs 12.4%, respectively (hazard ratio, 1.03 [95% CI, 0.70-1.51]). There were also no differences in hospitalization rates (hazard ratio, 0.94 [95% CI, 0.79-1.13]). The HCTR intervention was effective at 9 weeks, significantly improving peak oxygen consumption (0.95 [95% CI, 0.65-1.26] mL/kg/min vs 0.00 [95% CI, −0.31 to 0.30] mL/kg/min; P

Much of medical risk prediction involves externally derived prediction equations, nomograms, and point-based risk scores. These settings are vulnerable to misleading findings of incremental value based on versions of the net reclassification index (NRI) in common use. By applying non-nested models and point-based risk scores in the setting of stroke risk prediction in patients with atrial fibrillation (AF), we demonstrate current recommendations for presentation and interpretation of the NRI. We emphasize pitfalls that are likely to occur with point-based risk scores that are easy to neglect when statistical methodology is focused on continuous models. In order to make appropriate decisions about risk prediction and personalized medicine, physicians, researchers, and policy makers need to understand the strengths and limitations of the NRI.

BACKGROUND: Current cholesterol guidelines recommend using 10-year risk of atherosclerotic cardiovascular disease (ASCVD) to guide informed decisions regarding statin therapy, yet patients may have difficulty conceptualizing absolute risk estimates. Peer comparisons may provide an improved tool for patient risk comprehension. METHODS: Using data from the 2009–2014 National Health and Nutrition Examination Survey (NHANES), we estimated standardized cardiovascular risk percentiles for various age-, sex-, and race-specific subgroups based on their 10-year ASCVD risks using the Pooled Cohort Equations. RESULTS: We examined 9,160 adults in NHANES who were free of cardiovascular disease and had complete clinical data. Among specific age, sex, and race groups, we estimated the distribution of 10-year risk, calculating the 10-year risk corresponding to each percentile in order to generate standardized cardiovascular risk percentiles. Estimated 10-year ASCVD absolute risk varied markedly by age, sex, and race subgroups. A 10-year risk of 7.0% would put a 55 year-old black male in the 20(th) percentile relative to his peers (i.e., at lower risk than 80% of his peers), whereas a 10-year risk of 7.0% would put a 55 year-old white female in the 95(th) percentile (i.e., only 5% of her peers would have higher risk). Standardized cardiovascular risk percentiles by age, race, and sex are available online at populationrelativerisk.dcri.org. CONCLUSION: Cardiovascular risk varies substantially by age, sex, and race. These data allow for 10-year absolute risks of ASCVD to be translated into a standardized cardiovascular risk percentile, providing patients with information that is easy to understanding regarding how their personal risk of cardiovascular disease compares with their age-, sex-, and race-matched peers.

Elevated fibroblast growth factor 23 (FGF23) levels, measured at a single time, are strongly associated with increased risk of mortality in patients with CKD. There are minimal data on serial FGF23 measurements in CKD. In a prospective case-cohort study of the Chronic Renal Insufficiency Cohort, we measured FGF23 at two to five annual time points (mean 4.0±1.2) in a randomly selected subcohort of 1135 participants, of whom 203 died, and all remaining 390 participants who died through mid-2013. Higher FGF23 was independently associated with increased risk of death in multivariable-adjusted analyses of time-varying FGF23 (hazard ratio per 1-SD increase in ln-transformed FGF23, 1.84; 95% CI, 1.67 to 2.03). Median FGF23 was stable over 5 years of follow-up, but its gradually right-skewed distribution suggested a subpopulation with markedly elevated FGF23. Trajectory analysis revealed three distinct trajectories: stable FGF23 in the majority of participants (slope of lnFGF23 per year =0.03, 95% CI, 0.02 to 0.04, n=724) and smaller subpopulations with slowly (slope=0.14, 95% CI, 0.12 to 0.16, n=486) or rapidly (slope=0.46, 95% CI, 0.38 to 0.54, n=99) rising levels. Compared with stable FGF23, participants with slowly rising FGF23 trajectories were at 4.49-fold higher risk of death (95% CI, 3.17 to 6.35) and individuals with rapidly rising FGF23 trajectories were at 15.23-fold higher risk of death (95% CI, 8.24 to 28.14) in fully adjusted analyses. Trajectory analyses that used four or three annual FGF23 measurements yielded qualitatively similar results. In conclusion, FGF23 levels are stable over time in the majority of patients with CKD, but serial measurements identify subpopulations with rising levels and exceptionally high risk of death.

IMPORTANCE: There are limited clinical trial data and discrepant recommendations regarding use of anticoagulation therapy in patients with atrial fibrillation (AF) aged 65 to 74 years without other stroke risk factors. OBJECTIVES: To evaluate the risk of stroke without anticoagulation therapy in men and women with AF aged 66 to 74 years without other CHA(2)DS(2)-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes, stroke, vascular disease, age 65-74 years, female sex) risk factors and examine the association of stroke incidence with patient age. DESIGN, SETTING, AND PARTICIPANTS: A population-based retrospective cohort study was conducted using linked administrative databases. The population included 16 351 individuals aged 66 to 74 years who were newly diagnosed with AF in Ontario, Canada, between April 1, 2007, and March 31, 2017. Exclusion criteria included long-term care residence, prior anticoagulation therapy, valvular disease, heart failure, hypertension, diabetes, stroke, and vascular disease. The cumulative incidence function was used to estimate the 1-year incidence of stroke in patients who did not receive anticoagulation therapy. Fine-Gray regression was used to study the association of patient characteristics with stroke incidence and derive estimates of stroke risk at each age. Death was treated as a competing risk and patients were censored if they initiated anticoagulation therapy. Inverse probability of censoring weights was used to account for patient censoring. Data analysis was performed from May 26, 2019, to December 9, 2020. EXPOSURES: Atrial fibrillation and age. MAIN OUTCOMES AND MEASURES: Hospitalizations for stroke. RESULTS: Of the 16 351 individuals with AF (median [interquartile range] age, 70 [68-72] years), 8352 (51.1%) were men; 6314 individuals (38.6%) started anticoagulation therapy during follow-up. The overall 1-year stroke incidence among patients who did not receive anticoagulation therapy was 1.1% (95% CI, 1.0%-1.3%) and the incidence of death without stroke was 8.1% (95% CI, 7.7%-8.5%). The incidence of stroke was not significantly associated with sex. The estimated 1-year stroke risk increased with patient age from 66 years (0.7%; 95% CI, 0.5%-0.9%) to 74 years (1.7%; 95% CI, 1.3%-2.1%). CONCLUSIONS AND RELEVANCE: The risk of stroke more than doubled in this study as men and women with AF but no other CHA(2)DS(2)-VASc risk factors aged from 66 to 74 years. These data suggest that anticoagulation therapy is more likely to benefit older individuals within this group of patients, whereas younger individuals are less likely to gain net clinical benefit from anticoagulation therapy.

We conducted a meta-analysis of randomized controlled trials (RCTs) and quasi-RCTs to synthesize evidence about the efficacy and safety of alternate-day vs daily dosing of statins.We searched selected databases through January 2, 2017 to identify relevant RCTs and quasi-RCTs. The primary outcome was change in low-density lipoprotein cholesterol (LDL-C), total cholesterol (TC), and triglycerides (TG), while secondary outcomes included adverse events and adherence.Twelve RCTs and 1 quasi-RCT (n = 1023 patients) were included in the analysis. Pooled analysis revealed no statistically significant difference between alternate-day and daily regimens of atorvastatin and rosuvastatin in terms of change in LDL-C (mean difference [MD] 6.79 mg/dL, 95% confidence interval [CI] -1.59, 15.17, p = 0.11, and 10.51 mg/dL, 95%CI -0.23, 21.26, p = 0.06, respectively) and TG (p 0.05). Daily regimens of atorvastatin and rosuvastatin were superior to alternate-day regimes in term of change in TC (MD 12.45 mg/L, 95%CI 8.14, 16.76, p 0.00001, and 15.80 mg/dL, 95%CI 5.66, 25.95, p = 0.002, respectively). For all outcomes, there was no statistically significant difference between alternate-day and daily regimens for both fluvastatin and pravastatin (p 0.05). Both regimens of statins were generally well tolerated with good adherence.Alternate-day dosing of individual statins (especially atorvastatin and rosuvastatin) is as efficacious as daily dosing on LDL-C and TG.

Background: Optimal management of patients with stable chest pain relies on the prognostic information provided by noninvasive cardiovascular testing, but there are limited data from randomized trials comparing anatomic with functional testing. Methods: In the PROMISE trial (Prospective Multicenter Imaging Study for Evaluation of Chest Pain), patients with stable chest pain and intermediate pretest probability for obstructive coronary artery disease (CAD) were randomly assigned to functional testing (exercise electrocardiography, nuclear stress, or stress echocardiography) or coronary computed tomography angiography (CTA). Site-based diagnostic test reports were classified as normal or mildly, moderately, or severely abnormal. The primary end point was death, myocardial infarction, or unstable angina hospitalizations over a median follow-up of 26.1 months. Results: Both the prevalence of normal test results and incidence rate of events in these patients were significantly lower among 4500 patients randomly assigned to CTA in comparison with 4602 patients randomly assigned to functional testing (33.4% versus 78.0%, and 0.9% versus 2.1%, respectively; both P P P =0.87], 2.65 [ P =0.001], 3.88 [ P P =0.04). If 2714 patients with at least an intermediate Framingham Risk Score (>10%) who had a normal functional test were reclassified as being mildly abnormal, the discriminatory capacity improved to 0.69 (95% CI, 0.64–0.74). Conclusions: Coronary CTA, by identifying patients at risk because of nonobstructive CAD, provides better prognostic information than functional testing in contemporary patients who have stable chest pain with a low burden of obstructive CAD, myocardial ischemia, and events. Clinical Trial Registration: URL: http://www.clinicaltrials.gov . Unique identifier: NCT01174550.

Background/aims Use of data monitoring committees to oversee clinical trials was first proposed nearly 50 years ago. Since then, data monitoring committee use in clinical trials has increased and evolved. Nonetheless, there are no well-defined criteria for determining the need for a data monitoring committee, and considerable variability exists in data monitoring committee composition and conduct. To understand and describe the role and function of data monitoring committees, and establish best practices for data monitoring committee trial oversight, the Clinical Trials Transformation Initiative—a public–private partnership to improve clinical trials—launched a multi-stakeholder project. Methods The data monitoring committee project team included 16 individuals charged with (1) clarifying the purpose of data monitoring committees, (2) identifying best practices for independent data monitoring committee conduct, (3) describing effective communication practices, and (4) developing strategies for training data monitoring committee members. Evidence gathering included a survey, a series of focus group discussions, and a 2-day expert meeting aimed at achieving consensus opinions that form the foundation of our data monitoring committee recommendations. Results We define the role of the data monitoring committee as an advisor to the research sponsor on whether to continue, modify, or terminate a trial based on periodic assessment of trial data. Data monitoring committees should remain independent from the sponsor and be composed of members with no relevant conflicts of interest. Representation on a data monitoring committee generally should include at least one clinician with expertise in the therapeutic area being studied, a biostatistician, and a designated chairperson who has experience with clinical trials and data monitoring. Data monitoring committee meetings are held periodically to evaluate the unmasked data from ongoing trials, but the content and conduct of meetings may vary depending on specific goals or topics for deliberation. To guide data monitoring committee conduct and communication plans, a charter consistent with the protocol’s research design and statistical analysis plan should be developed and agreed upon by the sponsor and the data monitoring committee prior to patient enrollment. We recommend concise and flexible charters that explain roles, responsibilities, operational issues, and how data monitoring committee recommendations are generated and communicated. The demand for data monitoring committee members appears to exceed the current pool of qualified individuals. To prepare a new generation of trained data monitoring committee members, we encourage a combination of didactic educational programs, practical experience, and skill development through apprenticeships and mentoring by experienced data monitoring committee members. Conclusion Our recommendations address data monitoring committee use, conduct, communication practices, and member preparation and training. Furthermore recommendations form the foundation for ongoing efforts to improve clinical trial oversight and enhance the safety and integrity of clinical research. These recommendations serve as a call to action for implementation of best practices that benefit study participants, study sponsors, and society.

An increasingly important data source for the development of clinical risk prediction models is electronic health records (EHRs). One of their key advantages is that they contain data on many individuals collected over time. This allows one to incorporate more clinical information into a risk model. However, traditional methods for developing risk models are not well suited to these irregularly collected clinical covariates. In this paper, we compare a range of approaches for using longitudinal predictors in a clinical risk model. Using data from an EHR for patients undergoing hemodialysis, we incorporate five different clinical predictors into a risk model for patient mortality. We consider different approaches for treating the repeated measurements including use of summary statistics, machine learning methods, functional data analysis, and joint models. We follow up our empirical findings with a simulation study. Overall, our results suggest that simple approaches perform just as well, if not better, than more complex analytic approaches. These results have important implication for development of risk prediction models with EHRs. Copyright © 2017 John Wiley & Sons, Ltd.

Drug-eluting stents (DES) reduce restenosis but require prolonged antiplatelet therapy, when compared with bare metal stents. Ideally, the patient should be involved in this risk:benefit assessment prior to selecting DES, to maximize the benefits and cost-effectiveness of care, and to improve medication adherence. However, accurate estimation of restenosis risk may require angiographic factors identified at cardiac catheterization.In a large PCI registry, we used logistic regression to identify clinical and angiographic predictors of clinically-driven target lesion revascularization (TLR) over the first year after stent placement. Discrimination c-statistic and integrated discrimination improvement (IDI) were used to calculate the incremental utility of angiographic variables when added to clinical predictors.Of 8501 PCI patients, TLR occurred in 4.5%. After adjusting for DES use, clinical TLR predictors were younger age, female sex, diabetes, prior PCI, and prior bypass surgery (model c-statistic 0.630). Angiographic predictors were vein graft PCI, in-stent restenosis lesion, longer stent length, and smaller stent diameter (c-statistic 0.650). After adding angiographic factors to the clinical model, c-statistic improved to 0.680 and the average separation in TLR risk among patients with and without TLR improved by 1% (IDI=0.010, 95% CI 0.009-0.014), primarily driven by those experiencing TLR (from 5.9% to 6.9% absolute risk).Among unselected PCI patients, the incidence of clinically-indicated TLR is5% at 1-year, and standard clinical variables only moderately discriminate who will and will not experience TLR. Angiographic variables significantly improve TLR risk assessment, suggesting that stent selection may be best performed after coronary anatomy has been delineated.Although several recent studies have challenged traditional expectations regarding the duration of dual antiplatelet therapy, current guidelines recommend at least 6 to 12months of treatment after implantation of a drug eluting stent, with a shorter course for bare metal stents. Stent selection ideally should involve input from the patient receiving these stents, but multiple studies have suggested that angiographic factors - obtained after the patient has received sedation during the diagnostic catheterization - are important predictors of repeat revascularization. In this analysis from a large registry of patients receiving coronary stents, angiographic characteristics were found to significantly improve risk assessment for target lesion revascularization, when added to clinical variables alone.

Polytomous discrimination index is a novel and important diagnostic accuracy measure for multi-category classification. After reconstructing its probabilistic definition, we propose a nonparametric approach to the estimation of polytomous discrimination index based on an empirical sample of biomarker values. In this paper, we provide the finite-sample and asymptotic properties of the proposed estimators and such analytic results may facilitate the statistical inference. Simulation studies are performed to examine the performance of the nonparametric estimators. Two real data examples are analysed to illustrate our methodology.

Background and Purpose: Data monitoring committees are responsible for safeguarding the interests of study participants and assuring the integrity and credibility of clinical trials. The independence of data monitoring committees from sponsors and investigators is essential in achieving this mission. Creative approaches are needed to address ongoing and emerging challenges that potentially threaten data monitoring committees’ independence and effectiveness. Methods: An expert panel of representatives from academia, industry and government sponsors, and regulatory agencies discussed these challenges and proposed best practices and operating principles for effective functioning of contemporary data monitoring committees. Results and Conclusions: Prospective data monitoring committee members need better training. Options could include didactic instruction as well as apprenticeships to provide real-world experience. Data monitoring committee members should be protected against legal liability arising from their service. While avoiding breaches in confidentiality of interim data remains a high priority, data monitoring committees should have access to unblinded efficacy and safety data throughout the trial to enable informed judgments about risks and benefits. Because overly rigid procedures can compromise their independence, data monitoring committees should have the flexibility necessary to best fulfill their responsibilities. Data monitoring committee charters should articulate principles that guide the data monitoring committee process rather than list a rigid set of requirements. Data monitoring committees should develop their recommendations by consensus rather than through voting processes. The format for the meetings of the data monitoring committee should maintain the committee’s independence and clearly establish the leadership of the data monitoring committee chair. The independent statistical group at the Statistical Data Analysis Center should have sufficient depth of knowledge about the study at hand and experience with trials in general to ensure that the data monitoring committee has access to timely, reliable, and readily interpretable insights about emerging evidence in the clinical trial. Contracts engaging data monitoring committee members for industry-sponsored trials should have language customized to the unique responsibilities of data monitoring committee members rather than use language appropriate to consultants for product development. Regulatory scientists would benefit from experiencing data monitoring committee service that does not conflict with their regulatory responsibilities.

Objectives This study sought to determine whether processes of care and long-term clinical outcomes for heart failure (HF) admissions across Get With The Guidelines-Heart Failure (GWTG-HF) program participating centers differ according to HF-specific risk-adjusted 30-day readmission rates (excess readmission ratio [ERR]) as determined by the Hospital Readmission Reduction Program (HRRP). Background HRRP penalizes hospitals with higher than expected risk-adjusted 30-day readmission rates (ERR >1) for common conditions including HF. However, it is unclear whether the differences in this metric of hospital performance used by HRRP and related penalties are associated with measured quality of care and long-term outcomes. Methods We analyzed data from the GWTG-HF registry linked to Medicare claims from July 2008 to June 2011. Using publically available data on HF-ERR in 2013, we stratified the participating centers into groups with low (HF-ERR ≤1) versus high (HF-ERR >1) risk-adjusted readmission rates. We compared the care quality, in-hospital, and 1-year clinical outcomes across the 2 groups in unadjusted and multivariable adjusted analysis. Results The analysis included 171 centers with 43,143 participants; 49% of centers had high risk-adjusted 30-day readmission rates (HF-ERR >1). There were no differences between the low and high risk-adjusted 30-day readmission groups in median adherence rate to all performance measures (95.7% vs. 96.5%; p = 0.37) or median percentage of defect-free care (90.0% vs. 91.1%; p = 0.47). The composite 1-year outcome of death or all-cause readmission rates was also not different between the 2 groups (median 62.9% vs. 65.3%; p = 0.10). The high HF-ERR group had higher 1-year all-cause readmission rates (median 59.1% vs. 54.7%; p = 0.01). However, the 1-year mortality rates were lower among high versus low HF-ERR group with a trend toward statistical significance (median 28.2% vs. 31.7%; p = 0.07). Conclusions Quality of care and clinical outcomes were comparable among hospitals with high versus low risk-adjusted 30-day HF readmission rates. These findings raise questions about the validity of the HRRP performance metric in identifying and penalizing low-performance centers.

BACKGROUND: Trapping of apoB lipoprotein particles within the arterial wall initiates and drives the atherosclerotic process from beginning to end. Very low-density lipoprotein particles (VLDL) contain most of the triglyceride in plasma whereas low-density lipoprotein particles (LDL) particles contain most of the cholesterol. Smaller numbers of chylomicron and Lp(a) particles are also present in plasma. All these particles have one molecule of apoB. Therefore, plasma apoB equals the total number of apoB particles. Because the lipid content of apoB particles is variable, plasma triglyceride and cholesterol are not always accurate-measures of the number of apoB particles. THE CHOLESTEROL MODEL OF ATHEROSCLEROSIS: The conventional model of atherosclerosis presumes that the mass of cholesterol within VLDL and LDL particles is the principal determinant of the mass of cholesterol that will be deposited within the arterial wall and will drive atherogenesis. But cholesterol can only enter the arterial wall within apoB particles and the mass of cholesterol that will be deposited is determined by the rate at which apoB particles are trapped within the arterial wall rather than passing harmless through. THE APOB PARTICLE MODEL OF ATHEROGENESIS: The number of apoB particles that enter and are trapped within the arterial wall is determined primarily by the number of apoB particles within the arterial lumen. However, once within the arterial wall, smaller cholesterol-depleted apoB particles have a greater tendency to be trapped than larger cholesterol-enriched apoB particles because they bind more avidly to the glycosaminoglycans within the subintimal space of the arterial wall. If so, a cholesterol-enriched particle would deposit more cholesterol than a cholesterol-depleted apoB particle. By contrast, more smaller apoB particles that enter the arterial wall will be trapped than larger apoB particles. The net result is, with the exceptions of the abnormal chylomicron remnants in type III hyperlipoproteinemia and Lp(a), all apoB particles are equally atherogenic. ApoB, therefore, unifies, amplifies, and simplifies the information from the conventional lipid markers as to the atherogenic risk attributable to the apoB lipoproteins.

IMPORTANCE: American College of Cardiology/American Heart Association cholesterol guidelines prioritize primary prevention statin therapy based on 10-year absolute risk (AR(10)) of atherosclerotic cardiovascular disease (ASCVD). However, given the same AR(10), patients with higher levels of low-density lipoprotein cholesterol (LDL-C) experience greater absolute risk reduction from statin therapy. OBJECTIVES: To estimate the cost-effectiveness of expanding preventive statin treatment eligibility from standard care to patients at borderline risk (AR(10), 5.0%-7.4%) for ASCVD and with high levels of LDL-C and to estimate cost-effectiveness of statin treatment across ranges of age, sex, AR(10), and LDL-C levels. DESIGN, SETTING, AND PARTICIPANTS: This study evaluated 100 simulated cohorts, each including 1 million ASCVD-free survey respondents (50% men and 50% women) aged 40 years at baseline. Cohorts were created by probabilistic sampling of the 1999-2014 US National Health and Nutrition Examination Surveys from the perspective of the US health care sector. The CVD Policy Model microsimulation version projected lifetime health and cost outcomes. Probability of first-ever coronary heart disease or stroke event was estimated by analysis of 6 pooled US cohort studies and recalibrated to match contemporary event rates. Other model variables were derived from national surveys, meta-analyses, and published literature. Data were analyzed from May 15, 2018, through June 10, 2019. EXPOSURES: Four statin treatment strategies were compared: (1) treat all patients with AR(10) of at least 7.5%, diabetes, or LDL-C of at least 190 mg/dL (standard care); (2) add treatment for borderline risk and LDL-C levels of 160 to 189 mg/dL; (3) add treatment for borderline risk and LDL-C levels of 130 to 159 mg/dL; and (4) add treatment for remainder of patients with AR(10) of at least 5.0%. Statin treatment was also compared with no statin treatment in age, sex, AR(10), and LDL-C strata. MAIN OUTCOMES AND MEASURES: Lifetime quality-adjusted life-years (QALYs) and costs (2019 US dollars) were projected and discounted 3.0% annually. The primary outcome was the incremental cost-effectiveness ratio. RESULTS: In these 100 simulated cohorts, each with 1 million patients aged 40 years at baseline (50% women and 50% men), adding preventive statins to individuals with borderline AR(10) and LDL-C levels of 160 to 189 mg/dL would be cost-saving; further treating borderline AR(10) and LDL-C levels of 130 to 159 mg/dL would also be cost-saving; and treating all individuals with AR(10) of at least 5.0% would be highly cost-effective ($33 558/QALY) and would prevent the most ASCVD events. Within age, AR(10), and sex categories, individuals with higher baseline LDL-C levels gained more QALYs from statin therapy. Cost-effectiveness increased with LDL-C level and AR(10). CONCLUSIONS AND RELEVANCE: In this study, lifetime statin treatment of patients in a hypothetical cohort with borderline ASCVD risk and LDL-C levels of 160 to 189 mg/dL was found to be cost-saving. Results suggest that treating all patients at borderline risk regardless of LDL-C level would likely be highly cost-effective.

BACKGROUND: Extended elevations of non-HDL cholesterol (non-HDL-C) across a lifespan are associated with increased risk of cardiovascular disease (CVD). However, optimal testing intervals to identify individuals with high lipid-related CVD risk are unknown. OBJECTIVES: We determined the extent to which lipid levels in young adulthood predict future lipid trajectories and associated long-term CVD risk. METHODS: A sample of 2516 Framingham Offspring participants age 25–40 free of CVD and diabetes had their non-HDL-C progression modeled over 8 study examinations (mean follow-up 32.6 years) using group-based methods. CVD risk based on 25–30 years of follow up was evaluated using Kaplan-Meier analyses for those with mean non-HDL-C≥160 mg/dl (“high”) and

The associations of egg consumption with total, coronary heart disease (CHD), and stroke mortality are poorly understood. We prospectively evaluated the link between total, CHD, and stroke mortality with egg consumption using a randomly selected sample of U.S. adults. Next we validated these results within a meta-analysis and systematic review of all available prospective results. We assessed the mean of cardiometabolic risk factors across the intake of eggs. We made the analysis based on data from the National Health and Nutrition Examination Surveys (NHANES; 1999-2010). In NHANES, vital status through December 31, 2011, was ascertained. Cox proportional hazard regression models were used to relate baseline egg consumption with all-cause and cause-specific mortality. PubMed, Scopus, Web of Science, and Google Scholar databases were also searched (up to December 2017). The DerSimonian-Laird method and generic inverse variance methods were used for quantitative data synthesis. Overall, 23,524 participants from NHANES were included (mean age of 47.7 years; 48.7% were men). Across increasing the intake of eggs, adjusted mean levels of cardiometabolic risk factors worsened. Adjusted logistic regression showed that participants in the highest category of egg intake had a greater risk of diabetes (T2DM; 30%) and hypertension (HTN; 48%). With regard to total and CHD mortality, multivariable Cox regression in a fully adjusted model showed no link in males and females. In males, egg intake had a reverse (66%) association with stroke mortality, while this link was not significant among females. The results of pooling data from published prospective studies also showed no link between CHD and total mortality with egg consumption, whereas we observed a reverse (28%) association between egg intake and stroke mortality. These findings were robust after sensitivity analysis. According to our findings, egg intake had no association with CHD and total mortality, whereas was associated with lower risk of mortality from stroke. Egg consumption was associated with T2DM, HTN, C-reactive protein, and markers of glucose/insulin homeostasis. If confirmed in clinical trials (causation), this information may have applications for population-wide health measures. Key teaching points No link between total and CHD mortality with eggs intake in males and females. In males, egg intake had a reverse association with stroke mortality, while this link was not significant among females. The results of pooling data from published prospective studies also showed no link between CHD and total mortality with egg consumption, whereas we observed a reverse association between egg intake and stroke mortality.

The American College of Cardiology (ACC)/American Heart Association (AHA) guidelines suggest statin therapy for primary prevention in patients with an estimated 10-year atherosclerotic cardiovascular disease risk ≥7.5%1 as calculated by the Pooled Cohorts Equation. We recently demonstrated that only 53% of individuals eligible for statins based on the ≥7.5% risk criterion would have been eligible for a statin randomized control trial (RCT) in which benefit was demonstrated.2 We sought (1) to evaluate whether the recently published HOPE-3 trial (Heart Outcomes Prevention Evaluation)3 improves the evidence base for the ACC/AHA guidelines and (2) to estimate whether HOPE-3 increases the number of individuals eligible for statins in primary prevention beyond those recommended by the ACC/AHA guidelines. Data from the National Health and Nutrition Examination Survey (NHANES) collected between 2005 and 2010 were used to create a sample of 2134 participants representing 71.8 million Americans without atherosclerotic cardiovascular disease who were not currently taking statins. Participants were categorized based on the following criteria: (1) a 10-year risk of ≥7.5% by Pooled Cohorts Equation; (2) an expected absolute risk reduction of ≥2.3% based on Pooled Cohorts Equation risk and individualized relative risk reductions, as previously described2; …

Importance A 10-year benefit-based approach to statin therapy in primary prevention includes younger individuals with higher low-density lipoprotein cholesterol (LDL-C) and prevents more cardiovascular events than a risk-based approach. However, a 10-year treatment duration likely underestimates the expected benefits of statins. Objective To model the impact of a 30-year benefit approach to select individuals for statin therapy. Design, Setting, and Participants This cross-sectional analysis of the National Health and Nutrition Survey (NHANES) data set included samples of the US population from the 2009-2010, 2011-2012, and 2013-2014 data collection cycles. Individuals between 40 to 60 years old who did not have atherosclerotic cardiovascular disease, diabetes, or LDL-C levels greater than 190 mg/dL and who were not taking statins were included. Data analysis took place from November 2017 to August 2018. Exposures We calculated 10-year risk of atherosclerotic cardiovascular disease and 10-year and 30-year absolute risk reduction (10-year ARR and 30-year ARR) of atherosclerotic cardiovascular disease for each individual. Main Outcomes and Measures Number of individuals meeting eligibility for statins based on 10-year (atherosclerotic) cardiovascular disease risk, 10-year ARR, or 30-year ARR. Results A total of 1688 individuals were included, representing 56.6 million US individuals. Statin eligibility based on 7.5% CVR 10 was 9.5%; based on 2.3% 10-year ARR, 13.0%, and based on 15% 30-year ARR, 17.5%. The 10-year risk, 10-year benefit, and 30-year benefit approaches all led to similar acceptable mean absolute risk reductions at 30 years, with the benefit-based approaches better able to avoid treatment of individuals with low expected benefit. Individuals who met statin eligibility based solely on the 30-year ARR threshold of 15% or greater were younger (mean age, 50 [95% CI, 48-52] years) and more likely to be women (43% [95% CI, 26%-59%]) than those recommended with a 10-year ARR threshold of 2.3% or greater (mean age, 56 [95% CI, 54-57] years; 22% [95% CI, 10%-34%] women). This group also had lower 10-year risk (mean risk, 4.7% [95% CI, 4.4%-5.1%]) and higher LDL-C levels (mean level, 149 mg/dL [95% CI, 142-155 mg/dL]) than those recommended with a 10-year ARR threshold of 2.3% or greater (mean risk, 9.3% [95% CI, 8.3%-10.2%]; mean LDL-C levels, 110 [103-118] mg/dL). Preventable atherosclerotic cardiovascular disease events in 10 and 30 years were highest using the 30-year benefit approach (296 000 at 10 years and 2.03 million at 30 years) and lowest based on 10-year risk (204 000 at 10 years and 1.18 million at 30 years). Conclusions and Relevance A long-term benefit approach to statin eligibility identifies nearly 1 in 6 individuals as having a high degree of expected long-term benefit of statins, with a number needed to treat of less than 7. This approach identifies younger individuals with higher LDL-C levels who would not be currently recommended for treatment and may provide a more optimal approach for determining statin eligibility in primary prevention.

Clinical core laboratories, such as Echocardiography core laboratories, are increasingly used in clinical studies with imaging outcomes as primary, secondary, or surrogate endpoints. While many factors contribute to the quality of measurements of imaging variables, an essential step in ensuring the value of imaging data includes formal assessment and control of reproducibility via intra-observer and inter-observer reliability. There are many different agreement/reliability indices in the literature. However, different indices may lead to different conclusions and it is not clear which index is the preferred choice as an overall indication of data quality and a tool for providing guidance on improving quality and reliability in a core lab setting. In this paper, we pre-specify the desirable characteristics of an agreement index for assessing and improving reproducibility in a core lab setting; we compare existing agreement indices in terms of these characteristics to choose a preferred index. We conclude that, among the existing indices reviewed, the coverage probability for assessing agreement is the preferred agreement index on the basis of computational simplicity, its ability for rapid identification of discordant measurements to provide guidance for review and retraining, and its consistent evaluation of data quality across multiple reviewers, populations, and continuous/categorical data.

Objective: Electronic health records (EHRs) are a resource for “big data” analytics, containing a variety of data elements. We investigate how different categories of information contribute to prediction of mortality over different time horizons among patients undergoing hemodialysis treatment. Material and Methods: We derived prediction models for mortality over 7 time horizons using EHR data on older patients from a national chain of dialysis clinics linked with administrative data using LASSO (least absolute shrinkage and selection operator) regression. We assessed how different categories of information relate to risk assessment and compared discrete models to time-to-event models. Results: The best predictors used all the available data (c-statistic ranged from 0.72–0.76), with stronger models in the near term. While different variable groups showed different utility, exclusion of any particular group did not lead to a meaningfully different risk assessment. Discrete time models performed better than time-to-event models. Conclusions: Different variable groups were predictive over different time horizons, with vital signs most predictive for near-term mortality and demographic and comorbidities more important in long-term mortality.

Objective: Electronic health records (EHRs) are an increasingly common data source for clinical risk prediction, presenting both unique analytic opportunities and challenges. We sought to evaluate the current state of EHR based risk prediction modeling through a systematic review of clinical prediction studies using EHR data.Methods: We searched PubMed for articles that reported on the use of an EHR to develop a risk prediction model from 2009 to 2014. Articles were extracted by two reviewers, and we abstracted information on study design, use of EHR data, model building, and performance from each publication and supplementary documentation.Results: We identified 107 articles from 15 different countries. Studies were generally very large (median sample size = 26 100) and utilized a diverse array of predictors. Most used validation techniques (n = 94 of 107) and reported model coefficients for reproducibility (n = 83). However, studies did not fully leverage the breadth of EHR data, as they uncommonly used longitudinal information (n = 37) and employed relatively few predictor variables (median = 27 variables). Less than half of the studies were multicenter (n = 50) and only 26 performed validation across sites. Many studies did not fully address biases of EHR data such as missing data or loss to follow-up. Average c-statistics for different outcomes were: mortality (0.84), clinical prediction (0.83), hospitalization (0.71), and service utilization (0.71).Conclusions: EHR data present both opportunities and challenges for clinical risk prediction. There is room for improvement in designing such studies.

Background— Current guidelines recommend statins in the primary prevention of cardiovascular disease on the basis of predicted cardiovascular risk without directly considering the expected benefits of statin therapy based on the available randomized, controlled trial evidence. Methods and Results— We included 2134 participants representing 71.8 million American residents potentially eligible for statins in primary prevention from the National Health and Nutrition Examination Survey for the years 2005 to 2010. We compared statin eligibilities using 2 separate approaches: a 10-year risk-based approach (≥7.5% 10-year risk) and an individualized benefit approach (ie, based on predicted absolute risk reduction over 10 years [ARR 10 ] ≥2.3% from randomized, controlled trial data). A risk-based approach led to the eligibility of 15.0 million (95% confidence interval, 12.7–17.3 million) Americans, whereas a benefit-based approach identified 24.6 million (95% confidence interval, 21.0–28.1 million). The corresponding numbers needed to treat over 10 years were 21 (range, 9–44) and 25 (range, 9–44). The benefit-based approach identified 9.5 million lower-risk (10 ) compared with higher-risk individuals. This lower-risk/acceptable-benefit group includes younger individuals (mean age, 55.2 versus 62.5 years; P P =0.01). Statin treatment in this group would be expected to prevent an additional 266 508 cardiovascular events over 10 years. Conclusions— An individualized statin benefit approach can identify lower-risk individuals who have equal or greater expected benefit from statins in primary prevention compared with higher-risk individuals. This approach may help develop guideline recommendations that better identify individuals who meaningfully benefit from statin therapy.

Several studies suggest differences between fibrates and statins in lowering plasma fibrinogen (Fib) concentrations, but the evidence is not definitive. Therefore, the aim of this meta-analysis of head-to-head randomized trials was to compare the efficacy of statins and fibrates on plasma Fib concentrations.The literature search included Medline, Scopus, and Web of Science up to February 1st, 2015, to identify head-to-head comparative randomized trials investigating the efficacy of fibrates vs statins on plasma Fib concentrations.In total 22 trials with 2762 participants were included to the meta-analysis. Random-effect meta-analysis suggested a significantly greater effect of fibrates vs statins in lowering plasma Fib concentrations (weighted mean difference [WMD]: -40.7mg/dL, 95% confidence interval [CI]: -55.2, -26.3, p0.001). When the analysis was stratified according to the type of fibrate administered, there were significant Fib-lowering effects with both bezafibrate (n=8 treatment arms; WMD: -23.7mg/dL, 95% CI: -41.8, -5.7, p=0.01) and fenofibrate (n=15 treatment arms; WMD: -43.7mg/dL, 95% CI: -61.3, -26.2, p0.001). Overall, there was a numerically greater effect in the subgroup of trials with ≥12 weeks duration (n=17 treatment arms; WMD: -42.7mg/dL, 95% CI: -60.3, -25.1, p0.001) compared with the subgroup of trials lasting12 weeks (n=7 treatment arms; WMD: -36.7mg/dL, 95% CI: -52.0, -21.4, p0.001).Monotherapy with either fibrates or statins suggested a significantly greater effect of fibrates in lowering plasma Fib concentrations. According to these findings, mechanisms associated with fibrinogen metabolism might be responsible for the distinct effects of statins and fibrates in reducing cardiovascular endpoints.

BackgroundComorbid diabetes and substance use diagnoses (SUD) represent a hazardous combination, both in terms of healthcare cost and morbidity. To date, there is limited information about the association of SUD and related mental disorders with type 2 diabetes mellitus (T2DM).MethodsWe examined the associations between T2DM and multiple psychiatric diagnosis categories, with a focus on SUD and related psychiatric comorbidities among adults with T2DM. We analyzed electronic health record (EHR) data on 170,853 unique adults aged ≥18 years from the EHR warehouse of a large academic healthcare system. Logistic regression analyses were conducted to estimate the strength of an association for comorbidities.ResultsOverall, 9% of adults (n=16,243) had T2DM. Blacks, Hispanics, Asians, and Native Americans had greater odds of having T2DM than whites. All 10 psychiatric diagnosis categories were more prevalent among adults with T2DM than among those without T2DM. Prevalent diagnoses among adults with T2MD were mood (21.22%), SUD (17.02%: tobacco 13.25%, alcohol 4.00%, drugs 4.22%), and anxiety diagnoses (13.98%). Among adults with T2DM, SUD was positively associated with mood, anxiety, personality, somatic, and schizophrenia diagnoses.ConclusionsWe examined a large diverse sample of individuals and found clinical evidence of SUD and psychiatric comorbidities among adults with T2DM. These results highlight the need to identify feasible collaborative care models for adults with T2DM and SUD related psychiatric comorbidities, particularly in primary care settings, that will improve behavioral health and reduce health risk.

A new paradigm for preventing atherosclerotic cardiovascular disease (ASCVD) is needed. The most recent US data show the long‐term decline in cardiovascular deaths has stopped, and has started to increase in the most at‐risk populations.1 Indeed, rising rates of obesity and diabetes mellitus in the setting of suboptimal risk factor control have resulted in a similar number of cardiovascular events occurring in those aged

Prevention of cardiovascular disease today is as different from Geoffrey Rose's day as day is to night. In his time, net benefit from medical therapy was uncertain because there were no therapies that had been proved to be safe, effective, and affordable. However, the evidence that therapies that lower blood pressure and cholesterol prevent cardiovascular events is now incontrovertible with the result that guidelines recommend medical treatment of a large segment of the population to prevent cardiovascular disease. Population prevention medicine is not something that may occur tomorrow. Population prevention medicine is being practiced today. Yet, much remains to be done. The medical treatments of hypertension and hyperlipidemia both take place within a medical model that was not designed to deal with the numbers that could benefit from preventive medical therapy, and we need to refine our present system of care to take the realities of population medicine into account.