Dataset Information

Regional Validation and Recalibration of Clinical Predictive Models for Patients With Acute Heart Failure.

ABSTRACT: Heart failure clinical practice guidelines recommend applying validated clinical predictive models (CPMs) to support decision making. While CPMs are now widely available, the generalizability of heart failure CPMs is largely unknown.We identified CPMs derived in North America that predict mortality for patients with acute heart failure and validated these models in different world regions to assess performance in a contemporary international clinical trial (N=4133) of patients with acute heart failure treated with guideline-directed medical therapy. We performed independent external validations of 3 CPMs predicting in-hospital mortality, 60-day mortality, and 1-year mortality, respectively. CPM discrimination decreased in all regional validation cohorts. The median change in area under the receiver operating curve was -0.09 (range -0.05 to -0.23). Regional calibration was highly variable (90th percentile of absolute difference between smoothed observed and predicted values range <1% to >50%). Calibration remained poor after global recalibrations; however, region-specific recalibration procedures significantly improved regional performance (recalibrated 90th percentile of absolute difference range <1% to 5% across all regions and all models).Acute heart failure CPM discrimination and calibration vary substantially across different world regions; region-specific (as opposed to global) recalibration techniques are needed to improve CPM calibration.

SUBMITTER: Wessler BS

PROVIDER: S-EPMC5721739 | biostudies-literature | 2017 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Regional Validation and Recalibration of Clinical Predictive Models for Patients With Acute Heart Failure.

Wessler Benjamin S BS Ruthazer Robin R Udelson James E JE Gheorghiade Mihai M Zannad Faiez F Maggioni Aldo A Konstam Marvin A MA Kent David M DM

Journal of the American Heart Association 20171118 11

<h4>Background</h4>Heart failure clinical practice guidelines recommend applying validated clinical predictive models (CPMs) to support decision making. While CPMs are now widely available, the generalizability of heart failure CPMs is largely unknown.<h4>Methods and results</h4>We identified CPMs derived in North America that predict mortality for patients with acute heart failure and validated these models in different world regions to assess performance in a contemporary international clinica ...[more]

PMID: 29151026

Similar Datasets

Project description:ObjectivesWe sought to describe the demographics, clinical characteristics, management and outcomes of patients in Oman with acute heart failure (AHF) as part of the Gulf aCute heArt failuRe rEgistry (CARE) project.MethodsData were analyzed from 988 consecutive patients admitted with AHF to 12 hospitals in Oman between 14 February and 14 November 2012.ResultsThe mean age of our patients was 63±12 years. Over half (57%) were male and 95% were Omani citizens. Fifty-seven percent of patients presented with acute decompensated chronic heart failure (ADCHF) while 43% had new-onset AHF. The primary comorbid conditions were hypertension (72%), coronary artery disease (55%), and diabetes mellitus (53%). Ischemic heart disease (IHD), hypertensive heart disease, and idiopathic cardiomyopathy were the most common etiologies of AHF in Oman. The median left ventricular ejection fraction of the cohort was 36% (27-45%) with 56% of the patients having heart failure with reduced ejection fraction (< 40%). Atrial fibrillation was seen in 15% of patients. Acute coronary syndrome (ACS) and non-compliance with medications were the most common precipitating factors. At discharge, angiotensin converting enzyme inhibitors and beta-blockers were prescribed adequately, but aldosterone antagonists were under prescribed. Within 12-months follow-up, one in two patients were rehospitalized for AHF. In-hospital mortality was 7.1%, which doubled to 15.7% at three months and reached 26.4% at one-year post discharge.ConclusionsOman CARE was the first prospective multicenter registry of AHF in Oman and showed that heart failure (HF) patients present at a younger age with recurrent ADCHF and HF with reduced ejection fraction. IHD was the most common etiology of HF with a low prevalence of AHF, but a high prevalence of acute coronary syndrome and non-compliance with medications precipitating HF. A quarter of patients died at one-year follow-up even though at discharge medical therapy was nearly optimal. Our study indicates an urgent need for prevention, early diagnosis, and treatment of AHF in Oman.

Project description:BackgroundMorbidity and mortality rates associated with heart failure remain high. A wide variety of demographic and clinical factors as well as biomarkers are associated with increased mortality rates. Despite this, most multivariate predictive models for heart failure mortality have predictive accuracies characterized by a C-statistic (area under the receiver operating curve) of ≈0.74.Methods and resultsWe analyzed data on 963 patients enrolled in the Vesnarinone Evaluation of Survival Trial (VEST), including circulating levels of 2 cytokines (tumor necrosis factor and interleukin-6) and their receptors sampled at baseline and at 8, 16, and 24 weeks. We built multivariate logistic regression models by using standard clinical variables and time-series of cytokine and cytokine receptor levels, using independent components analysis to handle collinearity among cytokine measurements, and L2-penalized stepwise regression for variable selection. We also built ensemble models with these data, using gentle boosting. Our multivariate logistic regression model using time-series cytokine measurements predicts 1-year mortality rates significantly better (P=0.001) than the baseline model, with a C-statistic of 0.81±0.03. Without the cytokines, the baseline model has a C-statistic of 0.73±0.03, and, with only baseline cytokine and cytokine receptor levels added, the model has a C-statistic of 0.74±0.04. An ensemble model of 100 decision stumps with serial cytokine measurements has a significantly better (P=0.04) C-statistic of 0.84±0.02. An ensemble model with baseline cytokine data and without the serial measurements has a C-statistic of 0.74±0.04.ConclusionsSignificant gains in accuracy of one year mortality prediction in chronic heart failure can be obtained by using logistic regression models that incorporate serial measurements of biomarkers such as cytokine and cytokine receptor levels. Ensemble models capture inherent variability in large patient populations, and boost predictive accuracy through the use of time-series measurements.

Project description:BackgroundWith higher adoption of electronic health records at health-care centers, electronic search algorithms (computable phenotype) for identifying acute decompensated heart failure (ADHF) among hospitalized patients can be an invaluable tool to enhance data abstraction accuracy and efficacy in order to improve clinical research accrual and patient centered outcomes. We aimed to derive and validate a computable phenotype for ADHF in hospitalized patients.MethodsWe screened 256, 443 eligible (age > 18 years and with prior research authorization) individuals who were admitted to Mayo Clinic Hospital in Rochester, MN, from January 1, 2006, through December 31, 2014. Using a randomly selected derivation cohort of 938 patients, several iterations of a free-text electronic search were developed and refined. The computable phenotype was subsequently validated in an independent cohort 100 patients. The sensitivity and specificity of the computable phenotype were compared to the gold standard (expert review of charts) and International Classification of Diseases-9 (ICD-9) codes for Acute Heart Failure.ResultsIn the derivation cohort, the computable phenotype achieved a sensitivity of 97.5%, and specificity of 100%, whereas ICD-9 codes for Acute Heart Failure achieved a sensitivity of 47.5% and specificity of 96.7%. When all Heart Failure codes (ICD-9) were used, sensitivity and specificity were 97.5 and 86.6%, respectively. In the validation cohort, the sensitivity and specificity of the computable phenotype were 100 and 98.5%. The sensitivity and specificity for the ICD-9 codes (Acute Heart Failure) were 42 and 98.5%. Upon use of all Heart Failure codes (ICD-9), sensitivity and specificity were 96.8 and 91.3%.ConclusionsOur results suggest that using computable phenotype to ascertain ADHF from the clinical notes contained within the electronic medical record are feasible and reliable. Our computable phenotype outperformed ICD-9 codes for the detection of ADHF.

Dataset Information

Regional Validation and Recalibration of Clinical Predictive Models for Patients With Acute Heart Failure.

Publications

Regional Validation and Recalibration of Clinical Predictive Models for Patients With Acute Heart Failure.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets