A Robust e-Epidemiology Tool in Phenotyping Heart Failure with Differentiation for Preserved and Reduced Ejection Fraction: the Electronic Medical Records and Genomics (eMERGE) Network

Volumetric brain MRI signatures of heart failure with preserved ejection fraction in the setting of dementia

Heart failure with preserved ejection fraction (HFpEF) is an important, emerging risk factor for dementia, but it is not clear whether HFpEF contributes to a specific pattern of neuroanatomical changes in dementia. A major challenge to studying this is the relative paucity of datasets of patients with dementia, with/without HFpEF, and relevant neuroimaging. We sought to demonstrate the feasibility of using modern data mining tools to create and analyze clinical imaging datasets and identify the neuroanatomical signature of HFpEF-associated dementia. We leveraged the bioinformatics tools at Vanderbilt University Medical Center to identify patients with a diagnosis of dementia with and without comorbid HFpEF using the electronic health record. We identified high resolution, clinically-acquired neuroimaging data on 30 dementia patients with HFpEF (age 76.9 ± 8.12 years, 61% female) as well as 301 age- and sex-matched patients with dementia but without HFpEF to serve as comparators (age 76.2 ± 8.52 years, 60% female). We used automated image processing pipelines to parcellate the brain into 132 structures and quantify their volume. We found six regions with significant atrophy associated with HFpEF: accumbens area, amygdala, posterior insula, anterior orbital gyrus, angular gyrus, and cerebellar white matter. There were no regions with atrophy inversely associated with HFpEF. Patients with dementia and HFpEF have a distinct neuroimaging signature compared to patients with dementia only. Five of the six regions identified in are in the temporo-parietal region of the brain. Future studies should investigate mechanisms of injury associated with cerebrovascular disease leading to subsequent brain atrophy.

Natural Language Processing for Adjudication of Heart Failure in a Multicenter Clinical Trial: A Secondary Analysis of a Randomized Clinical Trial

Importance

The gold standard for outcome adjudication in clinical trials is medical record review by a physician clinical events committee (CEC), which requires substantial time and expertise. Automated adjudication of medical records by natural language processing (NLP) may offer a more resource-efficient alternative but this approach has not been validated in a multicenter setting.

Objective

To externally validate the Community Care Cohort Project (C3PO) NLP model for heart failure (HF) hospitalization adjudication, which was previously developed and tested within one health care system, compared to gold-standard CEC adjudication in a multicenter clinical trial.

Design, Setting, and Participants

This was a retrospective analysis of the Influenza Vaccine to Effectively Stop Cardio Thoracic Events and Decompensated Heart Failure (INVESTED) trial, which compared 2 influenza vaccines in 5260 participants with cardiovascular disease at 157 sites in the US and Canada between September 2016 and January 2019. Analysis was performed from November 2022 to October 2023.

Exposures

Individual sites submitted medical records for each hospitalization. The central INVESTED CEC and the C3PO NLP model independently adjudicated whether the cause of hospitalization was HF using the prepared hospitalization dossier. The C3PO NLP model was fine-tuned (C3PO + INVESTED) and a de novo NLP model was trained using half the INVESTED hospitalizations.

Main Outcomes and Measures

Concordance between the C3PO NLP model HF adjudication and the gold-standard INVESTED CEC adjudication was measured by raw agreement, κ, sensitivity, and specificity. The fine-tuned and de novo INVESTED NLP models were evaluated in an internal validation cohort not used for training.

Results

Among 4060 hospitalizations in 1973 patients (mean [SD] age, 66.4 [13.2] years; 514 [27.4%] female and 1432 [72.6%] male]), 1074 hospitalizations (26%) were adjudicated as HF by the CEC. There was good agreement between the C3PO NLP and CEC HF adjudications (raw agreement, 87% [95% CI, 86-88]; κ, 0.69 [95% CI, 0.66-0.72]). C3PO NLP model sensitivity was 94% (95% CI, 92-95) and specificity was 84% (95% CI, 83-85). The fine-tuned C3PO and de novo NLP models demonstrated agreement of 93% (95% CI, 92-94) and κ of 0.82 (95% CI, 0.77-0.86) and 0.83 (95% CI, 0.79-0.87), respectively, vs the CEC. CEC reviewer interrater reproducibility was 94% (95% CI, 93-95; κ, 0.85 [95% CI, 0.80-0.89]).

Conclusions and Relevance

The C3PO NLP model developed within 1 health care system identified HF events with good agreement relative to the gold-standard CEC in an external multicenter clinical trial. Fine-tuning the model improved agreement and approximated human reproducibility. Further study is needed to determine whether NLP will improve the efficiency of future multicenter clinical trials by identifying clinical events at scale.

Automated Identification of Heart Failure with Reduced Ejection Fraction using Deep Learning-based Natural Language Processing

Background

The lack of automated tools for measuring care quality has limited the implementation of a national program to assess and improve guideline-directed care in heart failure with reduced ejection fraction (HFrEF). A key challenge for constructing such a tool has been an accurate, accessible approach for identifying patients with HFrEF at hospital discharge, an opportunity to evaluate and improve the quality of care.

Methods

We developed a novel deep learning-based language model for identifying patients with HFrEF from discharge summaries using a semi-supervised learning framework. For this purpose, hospitalizations with heart failure at Yale New Haven Hospital (YNHH) between 2015 to 2019 were labeled as HFrEF if the left ventricular ejection fraction was under 40% on antecedent echocardiography. The model was internally validated with model-based net reclassification improvement (NRI) assessed against chart-based diagnosis codes. We externally validated the model on discharge summaries from hospitalizations with heart failure at Northwestern Medicine, community hospitals of Yale New Haven Health in Connecticut and Rhode Island, and the publicly accessible MIMIC-III database, confirmed with chart abstraction.

Results

A total of 13,251 notes from 5,392 unique individuals (mean age 73 ± 14 years, 48% female), including 2,487 patients with HFrEF (46.1%), were used for model development (train/held-out test: 70/30%). The deep learning model achieved an area under receiving operating characteristic (AUROC) of 0.97 and an area under precision-recall curve (AUPRC) of 0.97 in detecting HFrEF on the held-out set. In external validation, the model had high performance in identifying HFrEF from discharge summaries with AUROC 0.94 and AUPRC 0.91 on 19,242 notes from Northwestern Medicine, AUROC 0.95 and AUPRC 0.96 on 139 manually abstracted notes from Yale community hospitals, and AUROC 0.91 and AUPRC 0.92 on 146 manually reviewed notes at MIMIC-III. Model-based prediction of HFrEF corresponded to an overall NRI of 60.2 ± 1.9% compared with the chart diagnosis codes (p-value < 0.001) and an increase in AUROC from 0.61 [95% CI: 060-0.63] to 0.91 [95% CI 0.90-0.92].

Conclusions

We developed and externally validated a deep learning language model that automatically identifies HFrEF from clinical notes with high precision and accuracy, representing a key element in automating quality assessment and improvement for individuals with HFrEF.

Retrospective comparison of traditional and artificial intelligence-based heart failure phenotyping in a US health system to enable real-world evidence

OBJECTIVE

Quantitatively evaluate the quality of data underlying real-world evidence (RWE) in heart failure (HF).

DESIGN

Retrospective comparison of accuracy in identifying patients with HF and phenotypic information was made using traditional (ie, structured query language applied to structured electronic health record (EHR) data) and advanced (ie, artificial intelligence (AI) applied to unstructured EHR data) RWE approaches. The performance of each approach was measured by the harmonic mean of precision and recall (F1 score) using manual annotation of medical records as a reference standard.

SETTING

EHR data from a large academic healthcare system in North America between 2015 and 2019, with an expected catchment of approximately 5 00 000 patients.

POPULATION

4288 encounters for 1155 patients aged 18-85 years, with 472 patients identified as having HF.

OUTCOME MEASURES

HF and associated concepts, such as comorbidities, left ventricular ejection fraction, and selected medications.

RESULTS

The average F1 scores across 19 HF-specific concepts were 49.0% and 94.1% for the traditional and advanced approaches, respectively (p<0.001 for all concepts with available data). The absolute difference in F1 score between approaches was 45.1% (98.1% relative increase in F1 score using the advanced approach). The advanced approach achieved superior F1 scores for HF presence, phenotype and associated comorbidities. Some phenotypes, such as HF with preserved ejection fraction, revealed dramatic differences in extraction accuracy based on technology applied, with a 4.9% F1 score when using natural language processing (NLP) alone and a 91.0% F1 score when using NLP plus AI-based inference.

CONCLUSIONS

A traditional RWE generation approach resulted in low data quality in patients with HF. While an advanced approach demonstrated high accuracy, the results varied dramatically based on extraction techniques. For future studies, advanced approaches and accuracy measurement may be required to ensure data are fit-for-purpose.

Mendelian randomization and the association of fibroblast growth factor-23 with heart failure with preserved ejection fraction

PURPOSE OF REVIEW

Observational data provide compelling evidence for elevated fibroblast growth factor-23 (FGF23) as a risk factor for heart failure (HF), particularly heart failure with preserved ejection fraction (HFpEF). Given the limitations of observational studies, uncertainties persist regarding the causal role of FGF23 in the pathogenesis of HF and HFpEF. Recently, Mendelian randomization (MR) studies have been performed to examine causal associations between FGF23 and HF phenotypes.

RECENT FINDINGS

The current review describes the methodological basis of the MR techniques used to examine the causal role of FGF23 on HF phenotypes, highlighting the importance of large-scale multiomics data. The findings from most of the MR studies indicate an absence of evidence of a causal effect of FGF23 on the risk of HF in general population settings. However, analysis using individual-level data showed a strong association between genetically-predicted FGF23 and HFpEF in individuals with a genetic predisposition to low estimated glomerular filtration (eGFR).

SUMMARY

Evidence from MR analysis suggests a causal role of FGF23 in the pathogenesis of HFpEF in low eGFR settings - a finding supported by experimental, clinical, and epidemiological data. While future MR studies of FGF23 and HFpEF could provide further evidence, randomized trials of FGF23-lowering agents could provide the most definitive answers on the association in chronic kidney disease populations.

Diagnostic signature for heart failure with preserved ejection fraction (HFpEF): a machine learning approach using multi-modality electronic health record data

BACKGROUND

Heart failure with preserved ejection fraction (HFpEF) is thought to be highly prevalent yet remains underdiagnosed. Evidence-based treatments are available that increase quality of life and decrease hospitalization. We sought to develop a data-driven diagnostic model to predict from electronic health records (EHR) the likelihood of HFpEF among patients with unexplained dyspnea and preserved left ventricular EF.

METHODS AND RESULTS

The derivation cohort comprised patients with dyspnea and echocardiography results. Structured and unstructured data were extracted using an automated informatics pipeline. Patients were retrospectively diagnosed as HFpEF (cases), non-HF (control cohort I), or HF with reduced EF (HFrEF; control cohort II). The ability of clinical parameters and investigations to discriminate cases from controls was evaluated by extreme gradient boosting. A likelihood scoring system was developed and validated in a separate test cohort. The derivation cohort included 1585 consecutive patients: 133 cases of HFpEF (9%), 194 non-HF cases (Control cohort I) and 1258 HFrEF cases (Control cohort II). Two HFpEF diagnostic signatures were derived, comprising symptoms, diagnoses and investigation results. A final prediction model was generated based on the averaged likelihood scores from these two models. In a validation cohort consisting of 269 consecutive patients [with 66 HFpEF cases (24.5%)], the diagnostic power of detecting HFpEF had an AUROC of 90% (P < 0.001) and average precision of 74%.

CONCLUSION

This diagnostic signature enables discrimination of HFpEF from non-cardiac dyspnea or HFrEF from EHR and can assist in the diagnostic evaluation in patients with unexplained dyspnea. This approach will enable identification of HFpEF patients who may then benefit from new evidence-based therapies.

A multicenter program for electronic health record screening for patients with heart failure with preserved ejection fraction: Lessons from the DELIVER-EHR initiative

Efficiency in clinical trial recruitment and enrollment remains a major challenge in many areas of clinical medicine. In particular, despite the prevalence of heart failure with preserved ejection fraction (HFpEF), identifying patients with HFpEF for clinical trials has proven to be especially challenging. In this manuscript, we review strategies for contemporary clinical trial recruitment and present insights from the results of the DELIVER Electronic Health Record (EHR) Screening Initiative. The DELIVER trial was designed to evaluate the effects of dapagliflozin on clinical outcomes in patients with HFpEF. Within this trial, the multicenter DELIVER EHR Screening Initiative utilized EHR-based techniques in order to improve recruitment at selected sites in the United States. For this initiative, we developed and deployed a computable phenotype from the trial's eligibility criteria along with additional EHR tools at interested sites. Sites were then surveyed at the end of the program regarding lessons learned. Six sites were recruited, trained, and supported to utilize the EHR methodology and computable phenotype. Sites found the initiative to be helpful in identifying eligible patients and cited the individualized expert technical support as a critical factor in utilizing the program effectively. We found that the major challenge of implementation was the process of converting traditional inclusion/exclusion criteria into a computable phenotype within an established and ongoing trial. Other significant challenges noted by sites were the following: impact of the COVID-19 pandemic, engagement/support by local institutions, and limited availability of internal EHR experts/resources to execute programming. The study represents a proof-of-concept in the ability to utilize EHR-based tools in clinical trial recruitment for patients with HFpEF and provides important lessons for future initiatives. ClinicalTrials.gov Identifier: NCT03619213.

Association of Genetically Predicted Fibroblast Growth Factor-23 with Heart Failure: A Mendelian Randomization Study

BACKGROUND AND OBJECTIVES

Elevated fibroblast growth factor-23 (FGF23) has been consistently associated with heart failure, particularly heart failure with preserved ejection fraction, among patients with CKD and in the general population. FGF23 may directly induce cardiac remodeling and heart failure. However, biases affecting observational studies impede robust causal inferences. Mendelian randomization leverages genetic determinants of a risk factor to examine causality. We performed a two-sample Mendelian randomization to assess causal associations between FGF23 and heart failure.

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS

Genetic instruments were genome-wide significant genetic variants associated with FGF23, including variants near PIP5K1B, RGS14, LINC01229, and CYP24A1. We analyzed data from the Heart Failure Molecular Epidemiology for Therapeutic Targets and BioVU biobanks to examine associations of the four variants with overall heart failure, heart failure with preserved ejection fraction, and heart failure with reduced and mid-range ejection fraction. We developed an eGFR polygenic risk score using summary statistics from the Chronic Kidney Disease Genetics Consortium (CKDGen) genome-wide association study of eGFR in >1 million individuals and performed stratified analyses across eGFR polygenic risk score strata.

RESULTS

Genetically determined FGF23 was not associated with overall heart failure in the Heart Failure Molecular Epidemiology for Therapeutic Targets consortium (odds ratio, 1.13; 95% confidence interval, 0.89 to 1.42 per unit higher genetically predicted log FGF23) and the full BioVU sample (odds ratio, 1.32; 95% confidence interval, 0.95 to 1.84). In stratified analyses in BioVU, higher FGF23 was associated with overall heart failure (odds ratio, 3.09; 95% confidence interval, 1.38 to 6.91) among individuals with low eGFR-polygenic risk score (<1 SD below the mean), but not those with high eGFR-polygenic risk score (P interaction = 0.02). Higher FGF23 was also associated with heart failure with preserved ejection fraction among all BioVU participants (odds ratio, 1.47; 95% confidence interval, 1.01 to 2.14) and individuals with low eGFR-polygenic risk score (odds ratio, 7.20; 95% confidence interval, 2.80 to 18.49), but not those high eGFR-polygenic risk score (P interaction = 2.25 × 10^-4). No significant associations were observed with heart failure with reduced and midrange ejection fraction.

CONCLUSION

We found no association between genetically predicted FGF23 and heart failure in the Heart Failure Molecular Epidemiology for Therapeutic Targets consortium. In BioVU, genetically elevated FGF23 was associated with higher heart failure risk, specifically heart failure with preserved ejection fraction, particularly among individuals with low genetically predicted eGFR.

PODCAST

This article contains a podcast at https://www.asn-online.org/media/podcast/CJASN/2022_07_28_CJN00960122.mp3.

Deep Learning Analysis of Polish Electronic Health Records for Diagnosis Prediction in Patients with Cardiovascular Diseases

Electronic health records naturally contain most of the medical information in the form of doctor’s notes as unstructured or semi-structured texts. Current deep learning text analysis approaches allow researchers to reveal the inner semantics of text information and even identify hidden consequences that can offer extra decision support to doctors. In the presented article, we offer a new automated analysis of Polish summary texts of patient hospitalizations. The presented models were found to be able to predict the final diagnosis with almost 70% accuracy based just on the patient’s medical history (only 132 words on average), with possible accuracy increases when adding further sentences from hospitalization results; even one sentence was found to improve the results by 4%, and the best accuracy of 78% was achieved with five extra sentences. In addition to detailed descriptions of the data and methodology, we present an evaluation of the analysis using more than 50,000 Polish cardiology patient texts and dive into a detailed error analysis of the approach. The results indicate that the deep analysis of just the medical history summary can suggest the direction of diagnosis with a high probability that can be further increased just by supplementing the records with further examination results.

Systematic review of current natural language processing methods and applications in cardiology

Natural language processing (NLP) is a set of automated methods to organise and evaluate the information contained in unstructured clinical notes, which are a rich source of real-world data from clinical care that may be used to improve outcomes and understanding of disease in cardiology. The purpose of this systematic review is to provide an understanding of NLP, review how it has been used to date within cardiology and illustrate the opportunities that this approach provides for both research and clinical care. We systematically searched six scholarly databases (ACM Digital Library, Arxiv, Embase, IEEE Explore, PubMed and Scopus) for studies published in 2015-2020 describing the development or application of NLP methods for clinical text focused on cardiac disease. Studies not published in English, lacking a description of NLP methods, non-cardiac focused and duplicates were excluded. Two independent reviewers extracted general study information, clinical details and NLP details and appraised quality using a checklist of quality indicators for NLP studies. We identified 37 studies developing and applying NLP in heart failure, imaging, coronary artery disease, electrophysiology, general cardiology and valvular heart disease. Most studies used NLP to identify patients with a specific diagnosis and extract disease severity using rule-based NLP methods. Some used NLP algorithms to predict clinical outcomes. A major limitation is the inability to aggregate findings across studies due to vastly different NLP methods, evaluation and reporting. This review reveals numerous opportunities for future NLP work in cardiology with more diverse patient samples, cardiac diseases, datasets, methods and applications.

Heart failure clinical care analysis uncovers risk reduction opportunities for preserved ejection fraction subtype

Heart failure (HF) has no cure and, for HF with preserved ejection fraction (HFpEF), no life-extending treatments. Defining the clinical epidemiology of HF could facilitate earlier identification of high-risk individuals. We define the clinical epidemiology of HF subtypes (HFpEF and HF with reduced ejection fraction [HFrEF]), identified among 2.7 million individuals receiving routine clinical care. Differences in patterns and rates of accumulation of comorbidities, frequency of hospitalization, use of specialty care, were defined for each HF subtype. Among 28,156 HF cases, 8322 (30%) were HFpEF and 11,677 (42%) were HFrEF. HFpEF was the more prevalent subtype among older women. 177 Phenotypes differentially associated with HFpEF versus HFrEF. HFrEF was more frequently associated with diagnoses related to ischemic cardiac injury while HFpEF was associated more with non-cardiac comorbidities and HF symptoms. These comorbidity patterns were frequently present 3 years prior to a HFpEF diagnosis. HF subtypes demonstrated distinct patterns of clinical co-morbidities and disease progression. For HFpEF, these comorbidities were often non-cardiac and manifested prior to the onset of a HF diagnosis. Recognizing these comorbidity patterns, along the care continuum, may present a window of opportunity to identify individuals at risk for developing incident HFpEF.

FIBER: enabling flexible retrieval of electronic health records data for clinical predictive modeling

OBJECTIVES

The development of clinical predictive models hinges upon the availability of comprehensive clinical data. Tapping into such resources requires considerable effort from clinicians, data scientists, and engineers. Specifically, these efforts are focused on data extraction and preprocessing steps required prior to modeling, including complex database queries. A handful of software libraries exist that can reduce this complexity by building upon data standards. However, a gap remains concerning electronic health records (EHRs) stored in star schema clinical data warehouses, an approach often adopted in practice. In this article, we introduce the FlexIBle EHR Retrieval (FIBER) tool: a Python library built on top of a star schema (i2b2) clinical data warehouse that enables flexible generation of modeling-ready cohorts as data frames.

MATERIALS AND METHODS

FIBER was developed on top of a large-scale star schema EHR database which contains data from 8 million patients and over 120 million encounters. To illustrate FIBER's capabilities, we present its application by building a heart surgery patient cohort with subsequent prediction of acute kidney injury (AKI) with various machine learning models.

RESULTS

Using FIBER, we were able to build the heart surgery cohort (n = 12 061), identify the patients that developed AKI (n = 1005), and automatically extract relevant features (n = 774). Finally, we trained machine learning models that achieved area under the curve values of up to 0.77 for this exemplary use case.

CONCLUSION

FIBER is an open-source Python library developed for extracting information from star schema clinical data warehouses and reduces time-to-modeling, helping to streamline the clinical modeling process.

Invasive Hemodynamics in Heart Failure with Preserved Ejection Fraction: Importance of Detecting Pulmonary Vascular Remodeling and Right Heart Function

Heart failure (HF) is an ongoing crisis reaching epidemic proportions worldwide. About 50% of HF patients have a preserved ejection fraction. Invasive hemodynamics have shown varied results in patients who have HF with preserved ejection fraction (HFpEF). This article attempts to summarize the importance of detecting pulmonary vascular remodeling in HFpEF using invasive hemodynamics. Incorporating newer invasive hemodynamic parameters such as diastolic pulmonary gradient, pulmonary arterial compliance, pulmonary vascular resistance, and pulmonary arterial pulsatility index may improve patient selection for studies used in defining advanced therapies and clinical outcomes. Profiling of patients using invasive hemodynamic parameters may lead to better patient selection for clinical research.

Unconventional Natural Gas Development and Hospitalization for Heart Failure in Pennsylvania

BACKGROUND

Growing literature linking unconventional natural gas development (UNGD) to adverse health has implicated air pollution and stress pathways. Persons with heart failure (HF) are susceptible to these stressors.

OBJECTIVES

This study sought to evaluate associations between UNGD activity and hospitalization among HF patients, stratified by both ejection fraction (EF) status (reduced [HFrEF], preserved [HFpEF], not classifiable) and HF severity.

METHODS

We evaluated the odds of hospitalization among patients with HF seen at Geisinger from 2008 to 2015 using electronic health records. We assigned metrics of UNGD activity by phase (pad preparation, drilling, stimulation, and production) 30 days before hospitalization or a frequency-matched control selection date. We assigned phenotype status using a validated algorithm.

RESULTS

We identified 9,054 patients with HF with 5,839 hospitalizations (mean age 71.1 ± 12.7 years; 47.7% female). Comparing 4th to 1st quartiles, adjusted odds ratios (95% confidence interval) for hospitalization were 1.70 (1.35 to 2.13), 0.97 (0.75 to 1.27), 1.80 (1.35 to 2.40), and 1.62 (1.07 to 2.45) for pad preparation, drilling, stimulation, and production metrics, respectively. We did not find effect modification by HFrEF or HFpEF status. Associations of most UNGD metrics with hospitalization were stronger among those with more severe HF at baseline.

CONCLUSIONS

Three of 4 phases of UNGD activity were associated with hospitalization for HF in a large sample of patients with HF in an area of active UNGD, with similar findings by HFrEF versus HFpEF status. Older patients with HF seem particularly vulnerable to adverse health impacts from UNGD activity.

Electronic Medical Record-Based Case Phenotyping for the Charlson Conditions: Scoping Review

Background

Electronic medical records (EMRs) contain large amounts of rich clinical information. Developing EMR-based case definitions, also known as EMR phenotyping, is an active area of research that has implications for epidemiology, clinical care, and health services research.

Objective

This review aims to describe and assess the present landscape of EMR-based case phenotyping for the Charlson conditions.

Methods

A scoping review of EMR-based algorithms for defining the Charlson comorbidity index conditions was completed. This study covered articles published between January 2000 and April 2020, both inclusive. Embase (Excerpta Medica database) and MEDLINE (Medical Literature Analysis and Retrieval System Online) were searched using keywords developed in the following 3 domains: terms related to EMR, terms related to case finding, and disease-specific terms. The manuscript follows the Preferred Reporting Items for Systematic reviews and Meta-analyses extension for Scoping Reviews (PRISMA) guidelines.

Results

A total of 274 articles representing 299 algorithms were assessed and summarized. Most studies were undertaken in the United States (181/299, 60.5%), followed by the United Kingdom (42/299, 14.0%) and Canada (15/299, 5.0%). These algorithms were mostly developed either in primary care (103/299, 34.4%) or inpatient (168/299, 56.2%) settings. Diabetes, congestive heart failure, myocardial infarction, and rheumatology had the highest number of developed algorithms. Data-driven and clinical rule–based approaches have been identified. EMR-based phenotype and algorithm development reflect the data access allowed by respective health systems, and algorithms vary in their performance.

Conclusions

Recognizing similarities and differences in health systems, data collection strategies, extraction, data release protocols, and existing clinical pathways is critical to algorithm development strategies. Several strategies to assist with phenotype-based case definitions have been proposed.

Bi-directional association between depression and HF: An electronic health records-based cohort study

Objective:

To determine whether a bi-directional relationship exists between depression and HF within a single population of individuals receiving primary care services, using longitudinal electronic health records (EHRs).

Methods:

This retrospective cohort study utilized EHRs for adults who received primary care services within a large healthcare system in 2006. Validated EHR-based algorithms identified 10,649 people with depression (depression cohort) and 5,911 people with HF (HF cohort) between January 1, 2006 and December 31, 2018. Each person with depression or HF was matched 1:1 with an unaffected referent on age, sex, and outpatient service use. Each cohort (with their matched referents) was followed up electronically to identify newly diagnosed HF (in the depression cohort) and depression (in the HF cohort) that occurred after the index diagnosis of depression or HF, respectively. The risks of these outcomes were compared (vs. referents) using marginal Cox proportional hazard models adjusted for 16 comorbid chronic conditions.

Results:

2,024 occurrences of newly diagnosed HF were observed in the depression cohort and 944 occurrences of newly diagnosed depression were observed in the HF cohort over approximately 4–6 years of follow-up. People with depression had significantly increased risk for developing newly diagnosed HF (HR 2.08, 95% CI 1.89–2.28) and people with HF had a significantly increased risk of newly diagnosed depression (HR 1.34, 95% CI 1.17–1.54) after adjusting for all 16 comorbid chronic conditions.

Conclusion:

These results provide evidence of a bi-directional relationship between depression and HF independently of age, sex, and multimorbidity from chronic illnesses.

Association of Genetic Variants at TRPC6 With Chemotherapy-Related Heart Failure

Background: Our previous GWAS identified genetic variants at six novel loci that were associated with a decline in left ventricular ejection fraction (LVEF), p < 1 × 10⁻⁵ in 1,191 early breast cancer patients from the N9831 clinical trial of chemotherapy plus trastuzumab. In this study we sought replication of these loci.

Methods: We tested the top loci from the GWAS for association with chemotherapy-related heart failure (CRHF) using 26 CRHF cases from N9831 and 984 patients from the Mayo Clinic Biobank which included CRHF cases (N = 12) and control groups of patients treated with anthracycline +/– trastuzumab without HF (N = 282) and patients with HF that were never treated with anthracycline or trastuzumab (N = 690). We further examined associated loci in the context of gene expression and rare coding variants using a TWAS approach in heart left ventricle and Sanger sequencing, respectively. Doxorubicin-induced apoptosis and cardiomyopathy was modeled in human iPSC-derived cardiomyocytes and endothelial cells and a mouse model, respectively, that were pre-treated with GsMTx-4, an inhibitor of TRPC6.

Results:TRPC6 5′ flanking variant rs57242572-T was significantly more frequent in cases compared to controls, p = 0.031, and rs61918162-T showed a trend for association, p = 0.065. The rs61918162 T-allele was associated with higher TRPC6 expression in the heart left ventricle. We identified a single TRPC6 rare missense variant (rs767086724, N338S, prevalence 0.0025% in GnomAD) in one of 38 patients (2.6%) with CRHF. Pre-treatment of cardiomyocytes and endothelial cells with GsMTx4 significantly reduced doxorubicin-induced apoptosis. Similarly, mice treated with GsMTx4 had significantly improved doxorubicin-induced cardiac dysfunction.

Conclusions: Genetic variants that are associated with increased TRPC6 expression in the heart and rare TRPC6 missense variants may be clinically useful as risk factors for CRHF. GsMTx-4 may be a cardioprotective agent in patients with TRPC6 risk variants. Replication of the genetic associations in larger well-characterized samples and functional studies are required.

Identification of Patients with Heart Failure in Large Datasets

Large registries, administrative data, and the electronic health record (EHR) offer opportunities to identify patients with heart failure, which can be used for research purposes, process improvement, and optimal care delivery. Identification of cases is challenging because of the heterogeneous nature of the disease, which encompasses various phenotypes that may respond differently to treatment. The increasing availability of both structured and unstructured data in the EHR has expanded opportunities for cohort construction. This article reviews the current literature on approaches to identification of heart failure, and looks toward the future of machine learning, big data, and phenomapping.

Natural Language Processing for the Evaluation of Methodological Standards and Best Practices of EHR-based Clinical Research

The effective use of EHR data for clinical research is challenged by the lack of methodologic standards, transparency, and reproducibility. For example, our empirical analysis on clinical research ontologies and reporting standards found little-to-no informatics-related standards. To address these issues, our study aims to leverage natural language processing techniques to discover the reporting patterns and data abstraction methodologies for EHR-based clinical research. We conducted a case study using a collection of full articles of EHR-based population studies published using the Rochester Epidemiology Project infrastructure. Our investigation discovered an upward trend of reporting EHR-related research methodologies, good practice, and the use of informatics related methods. For example, among 1279 articles, 24.0% reported training for data abstraction, 6% reported the abstractors were blinded, 4.5% tested the inter-observer agreement, 5% reported the use of a screening/data collection protocol, 1.5% reported that team meetings were organized for consensus building, and 0.8% mentioned supervision activities by senior researchers. Despite that, the overall ratio of reporting/adoption of methodologic standards was still low. There was also a high variation regarding clinical research reporting. Thus, continuously developing process frameworks, ontologies, and reporting guidelines for promoting good data practice in EHR-based clinical research are recommended.

The effect of number of healthcare visits on study sample selection in electronic health record data

Introduction

Few studies have addressed how to select a study sample when using electronic health record (EHR) data.

Objective

To examine how changing criterion for number of visits in EHR data required for inclusion in a study sample would impact one basic epidemiologic measure: estimates of disease period prevalence.

Methods

Year 2016 EHR data from three Midwestern health systems (Northwestern Medicine in Illinois, University of Iowa Health Care, and Froedtert & the Medical College of Wisconsin, all regional tertiary health care systems including hospitals and clinics) was used to examine how alternate definitions of the study sample, based on number of healthcare visits in one year, affected measures of disease period prevalence. In 2016, each of these health systems saw between 160,000 and 420,000 unique patients. Curated collections of ICD-9, ICD-10, and SNOMED codes (from CMS-approved electronic clinical quality measures) were used to define three diseases: acute myocardial infarction, asthma, and diabetic nephropathy).

Results

Across all health systems, increasing the minimum required number of visits to be included in the study sample monotonically increased crude period prevalence estimates. The rate at which prevalence estimates increased with number of visits varied across sites and across diseases.

Conclusion

In addition to providing thorough descriptions of case definitions, when using EHR data authors must carefully describe how a study sample is identified and report data for a range of sample definitions, including minimum number of visits, so that others can assess the sensitivity of reported results to sample definition in EHR data.

Key words

Electronic Health Records, Sampling Studies, Prevalence, Methods

Characteristics and utilisation of the Mayo Clinic Biobank, a clinic-based prospective collection in the USA: cohort profile

PURPOSE

The Mayo Clinic Biobank was established to provide a large group of patients from which comparison groups (ie, controls) could be selected for case-control studies, to create a prospective cohort with sufficient power for common outcomes and to support electronic health record (EHR) studies.

PARTICIPANTS

A total of 56 862 participants enrolled (21% response rate) into the Mayo Clinic Biobank from Rochester, Minnesota (77%, n=43 836), Jacksonville, Florida (18%, n=10 368) and La Crosse, Wisconsin (5%, n=2658). Participants were all Mayo Clinic patients, 18 years of age or older and US residents.

FINDINGS TO DATE

Overall, 43% of participants were 65 years of age or older and female participants were more frequent (59%) than males at all sites. Most participants resided in the Upper Midwest regions of the USA (Minnesota, Iowa, Illinois or Wisconsin), Florida or Georgia. Self-reported race among Biobank participants was 90% white. Here we provide examples of the types of studies that have successfully utilised the resource, including (1) investigations of the population itself, (2) provision of controls for case-control studies, (3) genotype-driven research, (4) EHR-based research and (5) prospective recruitment to other studies. Over 270 projects have been approved to date to access Biobank data and/or samples; over 200 000 sample aliquots have been approved for distribution.

FUTURE PLANS

The data and samples in the Mayo Clinic Biobank can be used for various types of epidemiological and clinical studies, especially in the setting of case-control studies for which the Biobank samples serve as control samples. We are planning cohort studies with additional follow-up and acquisition of genetic information on a large scale.

Cohort Selection for Clinical Trials From Longitudinal Patient Records: Text Mining Approach

Background

Clinical trials are an important step in introducing new interventions into clinical practice by generating data on their safety and efficacy. Clinical trials need to ensure that participants are similar so that the findings can be attributed to the interventions studied and not to some other factors. Therefore, each clinical trial defines eligibility criteria, which describe characteristics that must be shared by the participants. Unfortunately, the complexities of eligibility criteria may not allow them to be translated directly into readily executable database queries. Instead, they may require careful analysis of the narrative sections of medical records. Manual screening of medical records is time consuming, thus negatively affecting the timeliness of the recruitment process.

Objective

Track 1 of the 2018 National Natural Language Processing Clinical Challenge focused on the task of cohort selection for clinical trials, aiming to answer the following question: Can natural language processing be applied to narrative medical records to identify patients who meet eligibility criteria for clinical trials? The task required the participating systems to analyze longitudinal patient records to determine if the corresponding patients met the given eligibility criteria. We aimed to describe a system developed to address this task.

Methods

Our system consisted of 13 classifiers, one for each eligibility criterion. All classifiers used a bag-of-words document representation model. To prevent the loss of relevant contextual information associated with such representation, a pattern-matching approach was used to extract context-sensitive features. They were embedded back into the text as lexically distinguishable tokens, which were consequently featured in the bag-of-words representation. Supervised machine learning was chosen wherever a sufficient number of both positive and negative instances was available to learn from. A rule-based approach focusing on a small set of relevant features was chosen for the remaining criteria.

Results

The system was evaluated using microaveraged F measure. Overall, 4 machine algorithms, including support vector machine, logistic regression, naïve Bayesian classifier, and gradient tree boosting (GTB), were evaluated on the training data using 10–fold cross-validation. Overall, GTB demonstrated the most consistent performance. Its performance peaked when oversampling was used to balance the training data. The final evaluation was performed on previously unseen test data. On average, the F measure of 89.04% was comparable to 3 of the top ranked performances in the shared task (91.11%, 90.28%, and 90.21%). With an F measure of 88.14%, we significantly outperformed these systems (81.03%, 78.50%, and 70.81%) in identifying patients with advanced coronary artery disease.

Conclusions

The holdout evaluation provides evidence that our system was able to identify eligible patients for the given clinical trial with high accuracy. Our approach demonstrates how rule-based knowledge infusion can improve the performance of machine learning algorithms even when trained on a relatively small dataset.

Identifying heart failure using EMR-based algorithms

BACKGROUND

Heart failure (HF) is a major clinical and public health problem, the management of which will benefit from large-scale pragmatic research that leverages electronic medical records (EMR). Requisite to using EMRs for HF research is the development of reliable algorithms to identify HF patients. We aimed to develop and validate computable phenotype algorithms to identify patients with HF using standardized data elements defined by the Patient Centered Outcomes Research Network (PCORnet) Common Data Model (CDM).

METHODS

We built HF computable phenotypes utilizing the data domains of HF diagnosis codes, prescribed HF-related medications and N-terminal B-type natriuretic peptide (NT-proBNP). Algorithms were validated in a cohort (n = 76,254) drawn from Olmsted County, MN between 2010-2012 a sample of whose records were manually reviewed to confirm HF according to Framingham criteria.

RESULTS

The different algorithms we tested provided different tradeoffs between sensitivity and positive predictive value (PPV). The highest sensitivity (78.7%) algorithm utilized one HF diagnosis code and had the lowest PPV (68.5%). The addition of more algorithm components, such as additional HF diagnosis codes, HF medications or elevated NT-proBNP, improved the PPV while reducing sensitivity. When added to a diagnostic code, the addition of NT-proBNP (>450 pg/mL) had a similar impact compared to additional HF medication criteria, increasing PPV by ∼3-4% and decreasing sensitivity by ∼7-10%.

CONCLUSIONS

Algorithms derived from PCORnet CDM elements can be used to identify patients with HF without manual adjudication with reasonable sensitivity and PPV. Algorithm choice should be driven by the goal of the research.

The eMERGE genotype set of 83,717 subjects imputed to ~40 million variants genome wide and association with the herpes zoster medical record phenotype

The Electronic Medical Records and Genomics (eMERGE) network is a network of medical centers with electronic medical records linked to existing biorepository samples for genomic discovery and genomic medicine research. The network sought to unify the genetic results from 78 Illumina and Affymetrix genotype array batches from 12 contributing medical centers for joint association analysis of 83,717 human participants. In this report, we describe the imputation of eMERGE results and methods to create the unified imputed merged set of genome‐wide variant genotype data. We imputed the data using the Michigan Imputation Server, which provides a missing single‐nucleotide variant genotype imputation service using the minimac3 imputation algorithm with the Haplotype Reference Consortium genotype reference set. We describe the quality control and filtering steps used in the generation of this data set and suggest generalizable quality thresholds for imputation and phenotype association studies. To test the merged imputed genotype set, we replicated a previously reported chromosome 6 HLA‐B herpes zoster (shingles) association and discovered a novel zoster‐associated loci in an epigenetic binding site near the terminus of chromosome 3 (3p29).

Challenges of evaluating chronic heart failure and acute heart failure events in research studies using large health care databases

Epidemiological studies on heart failure (HF) using large health care databases are becoming increasingly frequent, as they represent an invaluable opportunity to characterize the importance and risk factors of HF from a population perspective. Nevertheless, because of its complex diagnosis and natural history, the heterogeneous use of the relevant terminology in routine clinical practice, and the limitations of some disease coding systems, HF can be a challenging condition to assess using large health care databases as the main source of information. In this narrative review, we discuss some of the challenges that researchers may face, with a special focus on the identification and validation of chronic HF cases and acute HF decompensations. For each of these challenges, we present some potential solutions inspired by the literature and/or based on our research experience, aimed at increasing the internal validity of research and at informing its interpretation. We also discuss future directions on the field, presenting constructive recommendations aimed at facilitating the conduct of valid epidemiological studies on HF in the coming years.

Development and validation of a heart failure with preserved ejection fraction cohort using electronic medical records

Background

Heart failure (HF) with preserved ejection fraction (HFpEF) comprises nearly half of prevalent HF, yet is challenging to curate in a large database of electronic medical records (EMR) since it requires both accurate HF diagnosis and left ventricular ejection fraction (EF) values to be consistently ≥50%.

Methods

We used the national Veterans Affairs EMR to curate a cohort of HFpEF patients from 2002 to 2014. EF values were extracted from clinical documents utilizing natural language processing and an iterative approach was used to refine the algorithm for verification of clinical HFpEF. The final algorithm utilized the following inclusion criteria: any International Classification of Diseases-9 (ICD-9) code of HF (428.xx); all recorded EF ≥50%; and either B-type natriuretic peptide (BNP) or aminoterminal pro-BNP (NT-proBNP) values recorded OR diuretic use within one month of diagnosis of HF. Validation of the algorithm was performed by 3 independent reviewers doing manual chart review of 100 HFpEF cases and 100 controls.

Results

We established a HFpEF cohort of 80,248 patients (out of a total 1,155,376 patients with the ICD-9 diagnosis of HF). Mean age was 72 years; 96% were males and 12% were African-Americans. Validation analysis of the HFpEF algorithm had a sensitivity of 88%, specificity of 96%, positive predictive value of 96%, and a negative predictive value of 87% to identify HFpEF cases.

Conclusion

We developed a sensitive, highly specific algorithm for detecting HFpEF in a large national database. This approach may be applicable to other large EMR databases to identify HFpEF patients.

Association between an individual housing-based socioeconomic index and inconsistent self-reporting of health conditions: a prospective cohort study in the Mayo Clinic Biobank

OBJECTIVE

Using surveys to collect self-reported information on health and disease is commonly used in clinical practice and epidemiological research. However, the inconsistency of self-reported information collected longitudinally in repeated surveys is not well investigated. We aimed to investigate whether a socioeconomic status based on current housing characteristics, HOUsing-based SocioEconomic Status (HOUSES) index linking current address information to real estate property data, is associated with inconsistent self-reporting.

STUDY SETTING AND PARTICIPANTS

We performed a prospective cohort study using the Mayo Clinic Biobank (MCB) participants who resided in Olmsted County, Minnesota, USA, at the time of enrolment between 2009 and 2013, and were invited for a 4-year follow-up survey (n=11 717).

PRIMARY AND SECONDARY OUTCOME MEASURES

Using repeated survey data collected at the baseline and 4 years later, the primary outcome was the inconsistency in survey results when reporting prevalent diseases, defined by reporting to have 'ever' been diagnosed with a given disease in the baseline survey but reported 'never' in the follow-up survey. Secondary outcome was the response rate for the 4-year follow-up survey.

RESULTS

Among the MCB participants invited for the 4-year follow-up survey, 8508/11 717 (73%) responded to the survey. Forty-three per cent had at least one inconsistent self-reported disease. Lower HOUSES was associated with higher inconsistency rates, and the association remained significant after pertinent characteristics such as age and perceived general health (OR=1.46; 95% CI 1.17 to 1.84 for the lowest compared with the highest HOUSES decile). HOUSES was also associated with lower response rate for the follow-up survey (56% vs 77% for the lowest vs the highest HOUSES decile).

CONCLUSION

This study demonstrates the importance of using the HOUSES index that reflects current SES when using self-reporting through repeated surveys, as the HOUSES index at baseline survey was inversely associated with inconsistent self-report and the response rate for the follow-up survey.

Validity of Cardiovascular Data From Electronic Sources: The Multi-Ethnic Study of Atherosclerosis and HealthLNK

BACKGROUND

Understanding the validity of data from electronic data research networks is critical to national research initiatives and learning healthcare systems for cardiovascular care. Our goal was to evaluate the degree of agreement of electronic data research networks in comparison with data collected by standardized research approaches in a cohort study.

METHODS

We linked individual-level data from MESA (Multi-Ethnic Study of Atherosclerosis), a community-based cohort, with HealthLNK, a 2006 to 2012 database of electronic health records from 6 Chicago health systems. To evaluate the correlation and agreement of blood pressure in HealthLNK in comparison with in-person MESA examinations, and body mass index in HealthLNK in comparison with MESA, we used Pearson correlation coefficients and Bland-Altman plots. Using diagnoses in MESA as the criterion standard, we calculated the performance of HealthLNK for hypertension, obesity, and diabetes mellitus diagnosis by using International Classification of Diseases, Ninth Revision codes and clinical data. We also identified potential myocardial infarctions, strokes, and heart failure events in HealthLNK and compared them with adjudicated events in MESA.

RESULTS

Of the 1164 MESA participants enrolled at the Chicago Field Center, 802 (68.9%) participants had data in HealthLNK. The correlation was low for systolic blood pressure (0.39; P<0.0001). In comparison with MESA, HealthLNK overestimated systolic blood pressure by 6.5 mm Hg (95% confidence interval, 4.2-7.8). There was a high correlation between body mass index in MESA and HealthLNK (0.94; P<0.0001). HealthLNK underestimated body mass index by 0.3 kg/m² (95% confidence interval, -0.4 to -0.1). With the use of International Classification of Diseases, Ninth Revision codes and clinical data, the sensitivity and specificity of HealthLNK queries for hypertension were 82.4% and 59.4%, for obesity were 73.0% and 89.8%, and for diabetes mellitus were 79.8% and 93.3%. In comparison with adjudicated cardiovascular events in MESA, the concordance rates for myocardial infarction, stroke, and heart failure were, respectively, 41.7% (5/12), 61.5% (8/13), and 62.5% (10/16).

CONCLUSIONS

These findings illustrate the limitations and strengths of electronic data repositories in comparison with information collected by traditional standardized epidemiological approaches for the ascertainment of cardiovascular risk factors and events.

Epidemiology of heart failure with preserved ejection fraction

Heart failure (HF) with preserved ejection fraction (HFpEF) is a clinical syndrome associated with poor quality of life, substantial health-care resource utilization, and premature mortality. We summarize the current knowledge regarding the epidemiology of HFpEF with a focus on community-based studies relevant to quantifying the population burden of HFpEF. Current data regarding the prevalence and incidence of HFpEF in the community as well as associated conditions and risk factors, risk of morbidity and mortality after diagnosis, and quality of life are presented. In the community, approximately 50% of patients with HF have HFpEF. Although the age-specific incidence of HF is decreasing, this trend is less dramatic for HFpEF than for HF with reduced ejection fraction (HFrEF). The risk of HFpEF increases sharply with age, but hypertension, obesity, and coronary artery disease are additional risk factors. After adjusting for age and other risk factors, the risk of HFpEF is fairly similar in men and women, whereas the risk of HFrEF is much lower in women. Multimorbidity is common in both types of HF, but slightly more severe in HFpEF. A majority of deaths in patients with HFpEF are cardiovascular, but the proportion of noncardiovascular deaths is higher in HFpEF than HFrEF.

A comparison between physicians and computer algorithms for form CMS-2728 data reporting

INTRODUCTION

CMS-2728 form (Medical Evidence Report) assesses 23 comorbidities chosen to reflect poor outcomes and increased mortality risk. Previous studies questioned the validity of physician reporting on forms CMS-2728. We hypothesize that reporting of comorbidities by computer algorithms identifies more comorbidities than physician completion, and, therefore, is more reflective of underlying disease burden.

METHODS

We collected data from CMS-2728 forms for all 296 patients who had incident ESRD diagnosis and received chronic dialysis from 2005 through 2014 at Indiana University outpatient dialysis centers. We analyzed patients' data from electronic medical records systems that collated information from multiple health care sources. Previously utilized algorithms or natural language processing was used to extract data on 10 comorbidities for a period of up to 10 years prior to ESRD incidence. These algorithms incorporate billing codes, prescriptions, and other relevant elements. We compared the presence or unchecked status of these comorbidities on the forms to the presence or absence according to the algorithms.

FINDINGS

Computer algorithms had higher reporting of comorbidities compared to forms completion by physicians. This remained true when decreasing data span to one year and using only a single health center source. The algorithms determination was well accepted by a physician panel. Importantly, algorithms use significantly increased the expected deaths and lowered the standardized mortality ratios.

DISCUSSION

Using computer algorithms showed superior identification of comorbidities for form CMS-2728 and altered standardized mortality ratios. Adapting similar algorithms in available EMR systems may offer more thorough evaluation of comorbidities and improve quality reporting.

Improvement in Cardiovascular Risk Prediction with Electronic Health Records

The aim of this study was to compare the QRISKII, an electronic health data-based risk score, to the Framingham Risk Score (FRS) and atherosclerotic cardiovascular disease (ASCVD) score. Risk estimates were calculated for a cohort of 8783 patients, and the patients were followed up from November 29, 2012, through June 1, 2015, for a cardiovascular disease (CVD) event. During follow-up, 246 men and 247 women had a CVD event. Cohen's kappa statistic for the comparison of the QRISKII and FRS was 0.22 for men and 0.23 for women, with the QRISKII classifying more patients in the higher-risk groups. The QRISKII and ASCVD were more similar with kappa statistics of 0.49 for men and 0.51 for women. The QRISKII shows increased discrimination with area under the curve (AUC) statistics of 0.65 and 0.71, respectively, compared to the FRS (0.59 and 0.66) and ASCVD (0.63 and 0.69). These results demonstrate that incorporating additional data from the electronic health record (EHR) may improve CVD risk stratification.