A Natural Language Processing-Based Approach for Identifying Hospitalizations for Worsening Heart Failure Within an Integrated Health Care Delivery System

Applying natural language processing to identify emergency department and observation encounters for worsening heart failure

AIMS

Worsening heart failure (WHF) events occurring in non-inpatient settings are becoming increasingly recognized, with implications for prognostication. We evaluate the performance of a natural language processing (NLP)-based approach compared with traditional diagnostic coding for non-inpatient clinical encounters and left ventricular ejection fraction (LVEF).

METHODS AND RESULTS

We compared characteristics for encounters that did vs. did not meet WHF criteria, stratified by care setting [i.e. emergency department (ED) and observation stay]. Overall, 8407 (22%) encounters met NLP-based criteria for WHF (3909 ED visits and 4498 observation stays). The use of an NLP-derived definition adjudicated 3983 (12%) of non-primary HF diagnoses as meeting consensus definitions for WHF. The most common diagnosis indicated in these encounters was dyspnoea. Results were primarily driven by observation stays, in which 2205 (23%) encounters with a secondary HF diagnosis met the WHF definition by NLP.

CONCLUSIONS

The use of standard claims-based adjudication for primary diagnosis in the non-inpatient setting may lead to misclassification of WHF events in the ED and overestimate observation stays. Primary diagnoses alone may underestimate the burden of WHF in non-hospitalized settings.

Natural language processing of clinical notes enables early inborn error of immunity risk ascertainment

Background

There are now approximately 450 discrete inborn errors of immunity (IEI) described; however, diagnostic rates remain suboptimal. Use of structured health record data has proven useful for patient detection but may be augmented by natural language processing (NLP). Here we present a machine learning model that can distinguish patients from controls significantly in advance of ultimate diagnosis date.

Objective

We sought to create an NLP machine learning algorithm that could identify IEI patients early during the disease course and shorten the diagnostic odyssey.

Methods

Our approach involved extracting a large corpus of IEI patient clinical-note text from a major referral center's electronic health record (EHR) system and a matched control corpus for comparison. We built text classifiers with simple machine learning methods and trained them on progressively longer time epochs before date of diagnosis.

Results

The top performing NLP algorithm effectively distinguished cases from controls robustly 36 months before ultimate clinical diagnosis (area under precision recall curve > 0.95). Corpus analysis demonstrated that statistically enriched, IEI-relevant terms were evident 24+ months before diagnosis, validating that clinical notes can provide a signal for early prediction of IEI.

Conclusion

Mining EHR notes with NLP holds promise for improving early IEI patient detection.

Heart failure with mildly reduced and preserved ejection fraction: A review of disease burden and remaining unmet medical needs within a new treatment landscape

This review provides a comprehensive overview of heart failure with mildly reduced and preserved ejection fraction (HFmrEF/HFpEF), including its definition, diagnosis, and epidemiology; clinical, humanistic, and economic burdens; current pharmacologic landscape in key pharmaceutical markets; and unmet needs to identify key knowledge gaps. We conducted a targeted literature review in electronic databases and prioritized articles with valuable insights into HFmrEF/HFpEF. Overall, 27 randomized controlled trials (RCTs), 66 real-world evidence studies, 18 clinical practice guidelines, and 25 additional publications were included. Although recent heart failure (HF) guidelines set left ventricular ejection fraction thresholds to differentiate categories, characterization and diagnosis criteria vary because of the incomplete disease understanding. Recent epidemiological data are limited and diverse. Approximately 50% of symptomatic HF patients have HFpEF, more common than HFmrEF. Prevalence varies with country because of differing definitions and study characteristics, making prevalence interpretation challenging. HFmrEF/HFpEF has considerable mortality risk, and the mortality rate varies with study and patient characteristics and treatments. HFmrEF/HFpEF is associated with considerable morbidity, poor patient outcomes, and common comorbidities. Patients require frequent hospitalizations; therefore, early intervention is crucial to prevent disease burden. Recent RCTs show promising results like risk reduction of composite cardiovascular death or HF hospitalization. Costs data are scarce, but the economic burden is increasing. Despite new drugs, unmet medical needs requiring new treatments remain. Thus, HFmrEF/HFpEF is a growing global healthcare concern. With improving yet incomplete understanding of this disease and its promising treatments, further research is required for better patient outcomes.

Sex-Based Differences in the Epidemiology, Clinical Characteristics, and Outcomes Associated with Worsening Heart Failure Events in a Learning Health System

BACKGROUND

Differences in demographics, risk factors, and clinical characteristics may contribute to variation in men and women in terms of prevalence, clinical setting, and outcomes associated with worsening heart failure (WHF) events.

OBJECTIVES

To describe sex-based differences in the epidemiology, clinical characteristics, and outcomes associated with WHF events across clinical settings.

METHODS

We examined adults diagnosed with HF from 2010-2019 within a large, integrated healthcare delivery system. Electronic health record data were accessed for hospitalizations, emergency department (ED) visits/observation stays, and outpatient encounters. WHF was identified using validated natural language processing algorithms and defined as ≥1 symptom, ≥2 objective findings (including ≥1 sign), and ≥1 change in HF-related therapy. Incidence rates and associated outcomes for WHF were compared across care setting by sex.

RESULTS

We identified 1,122,368 unique clinical encounters with a diagnosis code for HF, with 124,479 meeting WHF criteria. These WHF encounters existed among 102,116 patients, of which 48,543 (47.5%) were women and 53,573 (52.5%) were men. Women experiencing WHF were older and more likely to have HF with preserved ejection fraction compared to men. The clinical settings of WHF were similar among women and men: hospitalizations (36.8% vs. 37.7%), ED visits or observation stays (11.8% vs. 13.4%), and outpatient encounters (4.4% vs. 4.9%). Women had lower odds of 30-day mortality following an index hospitalization (adjusted odds ratio [aOR] 0.88, 95% confidence interval [CI] 0.83-0.93) or ED visit/observation stay (aOR 0.86, 95% 0.75-0.98) for WHF.

CONCLUSION

Women and men contribute similarly to WHF events across diverse clinical settings despite marked differences in age and left ventricular ejection fraction.

Contemporary Decongestion Strategies in Patients Hospitalized for Heart Failure: A National Community-Based Cohort Study

BACKGROUND

Heart failure (HF) is a leading cause of hospitalization in the United States. Decongestion remains a central goal of inpatient management, but contemporary decongestion practices and associated weight loss have not been well characterized nationally.

OBJECTIVES

This study aimed to describe contemporary inpatient diuretic practices and clinical predictors of weight loss in patients hospitalized for HF.

METHODS

The authors identified HF hospitalizations from 2015 to 2022 in a U.S. national database aggregating deidentified patient-level electronic health record data across 31 geographically diverse community-based health systems. The authors report patient characteristics and inpatient weight change as a primary indicator of decongestion. Predictors of weight loss were evaluated using multivariable models. Temporal trends in inpatient diuretic practices, including augmented diuresis strategies such as adjunctive thiazides and continuous diuretic infusions, were assessed.

RESULTS

The study cohort included 262,673 HF admissions across 165,482 unique patients. The median inpatient weight loss was 5.3 pounds (Q1-Q3: 0.0-12.8 pounds) or 2.4 kg (Q1-Q3: 0.0-5.8 kg). Discharge weight was higher than admission weight in 20% of encounters. An increase of ≥0.3 mg/dL in serum creatinine from admission to inpatient peak occurred in >30% of hospitalizations and was associated with less weight loss. Adjunctive diuretic agents were utilized in <20% of encounters but were associated with greater weight loss.

CONCLUSIONS

In a large-scale U.S. community-based cohort study of HF hospitalizations, estimated weight loss from inpatient decongestion remains highly variable, with weight gain observed across many admissions. Augmented diuresis strategies were infrequently used. Comparative effectiveness trials are needed to establish optimal strategies for inpatient decongestion for acute HF.

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.

Uptake of Sodium-Glucose Cotransporter-2 Inhibitors in Hospitalized Patients With Heart Failure: Insights From the Veterans Affairs Healthcare System

BACKGROUND

The use of sodium-glucose cotransporter-2 inhibitors (SGLT2is) in Veterans Affairs (VA) patients hospitalized with heart failure (HF) has not been reported previously.

METHODS

VA electronic health record data were used to identify patients hospitalized for HF (primary or secondary diagnosis) from 01/2019-11/2022. Patients with SGLT2i allergy, advanced/end-stage chronic kidney disease (CKD) or advanced HF therapies were excluded. We identified factors associated with discharge SGLT2i prescriptions for patients hospitalized due to HF in 2022. We also compared SGLT2i and angiotensin receptor-neprilysin inhibitor (ARNI) prescription rates. Hospital-level variations in SGLT2i prescriptions were assessed via the median odds ratio.

RESULTS

A total of 69,680 patients were hospitalized due to HF; 10.3% were prescribed SGLT2i at discharge (4.4% newly prescribed, 5.9% continued preadmission therapy). SGLT2i prescription increased over time and was higher in patients with HFrEF and primary HF. Among 15,762 patients hospitalized in 2022, SGLT2i prescription was more likely in patients with diabetes (adjusted odds ratio [aOR] 2.27; 95% confidence interval [CI]: 2.09-2.47) and ischemic heart disease (aOR 1.14; 95% CI: 1.03-1.26). Patients with increased age (aOR 0.77 per 10 years; 95% CI: 0.73-0.80) and lower systolic blood pressure (aOR 0.94 per 10 mmHg; 95% CI: 0.92-0.96) were less likely to be prescribed SGLT2i, and SGLT2i prescription was not more likely in patients with CKD (aOR 1.07; 95% CI 0.98-1.16). The adjusted median odds ratio suggested a 1.8-fold variation in the likelihood that similar patients at 2 random VA sites were prescribed SGLT2i (range 0-21.0%). In patients with EF ≤ 40%, 30.9% were prescribed SGLT2i while 26.9% were prescribed ARNI (P < 0.01).

CONCLUSION

One-tenth of VA patients hospitalized for HF were prescribed SGLT2i at discharge. Opportunities exist to reduce variation in SGLT2i prescription rates across hospitals and to promote its use in patients with CKD and older age.

Endpoint adjudication in cardiovascular clinical trials

Endpoint adjudication (EA) is a common feature of contemporary randomized controlled trials (RCTs) in cardiovascular medicine. Endpoint adjudication refers to a process wherein a group of expert reviewers, known as the clinical endpoint committee (CEC), verify potential endpoints identified by site investigators. Events that are determined by the CEC to meet pre-specified trial definitions are then utilized for analysis. The rationale behind the use of EA is that it may lessen the potential misclassification of clinical events, thereby reducing statistical noise and bias. However, it has been questioned whether this is universally true, especially given that EA significantly increases the time, effort, and resources required to conduct a trial. Herein, we compare the summary estimates obtained using adjudicated vs. non-adjudicated site designated endpoints in major cardiovascular RCTs in which both were reported. Based on these data, we lay out a framework to determine which trials may warrant EA and where it may be redundant. The value of EA is likely greater when cardiovascular trials have nuanced primary endpoints, endpoint definitions that align poorly with practice, sub-optimal data completeness, greater operator variability, and lack of blinding. EA may not be needed if the primary endpoint is all-cause death or all-cause hospitalization. In contrast, EA is likely merited for more nuanced endpoints such as myocardial infarction, bleeding, worsening heart failure as an outpatient, unstable angina, or transient ischaemic attack. A risk-based approach to adjudication can potentially allow compromise between costs and accuracy. This would involve adjudication of a small proportion of events, with further adjudication done if inconsistencies are detected.

Association Between Participation in a Heart Failure Telemonitoring Program and Health Care Utilization and Death Within an Integrated Health Care Delivery System

BACKGROUND

The clinical usefulness of remote telemonitoring to reduce postdischarge health care use and death in adults with heart failure (HF) remains controversial.

METHODS AND RESULTS

Within a large integrated health care delivery system, we matched patients enrolled in a postdischarge telemonitoring intervention from 2015 to 2019 to patients not receiving telemonitoring at up to a 1:4 ratio on age, sex, and calipers of a propensity score. Primary outcomes were readmissions for worsening HF and all-cause death within 30, 90, and 365 days of the index discharge; secondary outcomes were all-cause readmissions and any outpatient diuretic dose adjustments. We matched 726 patients receiving telemonitoring to 1985 controls not receiving telemonitoring, with a mean age of 75 ± 11 years and 45% female. Patients receiving telemonitoring did not have a significant reduction in worsening HF hospitalizations (adjusted rate ratio [aRR] 0.95, 95% confidence interval [CI] 0.68-1.33), all-cause death (adjusted hazard ratio 0.60, 95% CI 0.33-1.08), or all-cause hospitalization (aRR 0.82, 95% CI 0.65-1.05) at 30 days, but did have an increase in outpatient diuretic dose adjustments (aRR 1.84, 95% CI 1.44-2.36). All associations were similar at 90 and 365 days postdischarge.

CONCLUSIONS

A postdischarge HF telemonitoring intervention was associated with more diuretic dose adjustments but was not significantly associated with HF-related morbidity and mortality.

Developing Clinical Risk Prediction Models for Worsening Heart Failure Events and Death by Left Ventricular Ejection Fraction

Background There is a need to develop electronic health record-based predictive models for worsening heart failure (WHF) events across clinical settings and across the spectrum of left ventricular ejection fraction (LVEF). Methods and Results We studied adults with heart failure (HF) from 2011 to 2019 within an integrated health care delivery system. WHF encounters were ascertained using natural language processing and structured data. We conducted boosted decision tree ensemble models to predict 1-year hospitalizations, emergency department visits/observation stays, and outpatient encounters for WHF and all-cause death within each LVEF category: HF with reduced ejection fraction (EF) (LVEF <40%), HF with mildly reduced EF (LVEF 40%-49%), and HF with preserved EF (LVEF ≥50%). Model discrimination was evaluated using area under the curve and calibration using mean squared error. We identified 338 426 adults with HF: 61 045 (18.0%) had HF with reduced EF, 49 618 (14.7%) had HF with mildly reduced EF, and 227 763 (67.3%) had HF with preserved EF. The 1-year risks of any WHF event and death were, respectively, 22.3% and 13.0% for HF with reduced EF, 17.0% and 10.1% for HF with mildly reduced EF, and 16.3% and 10.3% for HF with preserved EF. The WHF model displayed an area under the curve of 0.76 and mean squared error of 0.13, whereas the model for death displayed an area under the curve of 0.83 and mean squared error of 0.076. Performance and predictors were similar across WHF encounter types and LVEF categories. Conclusions We developed risk prediction models for 1-year WHF events and death across the LVEF spectrum using structured and unstructured electronic health record data and observed no substantial differences in model performance or predictors except for death, despite differences in underlying HF cause.

Cardiovascular Care Innovation through Data-Driven Discoveries in the Electronic Health Record

The electronic health record (EHR) represents a rich source of patient information, increasingly being leveraged for cardiovascular research. Although its primary use remains the seamless delivery of health care, the various longitudinally aggregated structured and unstructured data elements for each patient within the EHR can define the computational phenotypes of disease and care signatures and their association with outcomes. Although structured data elements, such as demographic characteristics, laboratory measurements, problem lists, and medications, are easily extracted, unstructured data are underused. The latter include free text in clinical narratives, documentation of procedures, and reports of imaging and pathology. Rapid scaling up of data storage and rapid innovation in natural language processing and computer vision can power insights from unstructured data streams. However, despite an array of opportunities for research using the EHR, specific expertise is necessary to adequately address confidentiality, accuracy, completeness, and heterogeneity challenges in EHR-based research. These often require methodological innovation and best practices to design and conduct successful research studies. Our review discusses these challenges and their proposed solutions. In addition, we highlight the ongoing innovations in federated learning in the EHR through a greater focus on common data models and discuss ongoing work that defines such an approach to large-scale, multicenter, federated studies. Such parallel improvements in technology and research methods enable innovative care and optimization of patient outcomes.

Natural Language Processing for Adjudication of Heart Failure Hospitalizations in a Multi-Center Clinical Trial

Background

The gold standard for outcome adjudication in clinical trials is chart review by a physician clinical events committee (CEC), which requires substantial time and expertise. Automated adjudication by natural language processing (NLP) may offer a more resource-efficient alternative. We previously showed that the Community Care Cohort Project (C3PO) NLP model adjudicates heart failure (HF) hospitalizations accurately within one healthcare system.

Methods

This study externally validated the C3PO NLP model against CEC adjudication in the INVESTED trial. INVESTED compared influenza vaccination formulations in 5260 patients with cardiovascular disease at 157 North American sites. A central CEC adjudicated the cause of hospitalizations from medical records. We applied the C3PO NLP model to medical records from 4060 INVESTED hospitalizations and evaluated agreement between the NLP and final consensus CEC HF adjudications. We then fine-tuned the C3PO NLP model (C3PO+INVESTED) and trained a de novo model using half the INVESTED hospitalizations, and evaluated these models in the other half. NLP performance was benchmarked to CEC reviewer inter-rater reproducibility.

Results

1074 hospitalizations (26%) were adjudicated as HF by the CEC. There was high agreement between the C3PO NLP and CEC HF adjudications (agreement 87%, kappa statistic 0.69). C3PO NLP model sensitivity was 94% and specificity was 84%. The fine-tuned C3PO and de novo NLP models demonstrated agreement of 93% and kappa of 0.82 and 0.83, respectively. CEC reviewer inter-rater reproducibility was 94% (kappa 0.85).

Conclusion

Our NLP model developed within a single healthcare system accurately identified HF events relative to the gold-standard CEC in an external multi-center clinical trial. Fine-tuning the model improved agreement and approximated human reproducibility. NLP may improve the efficiency of future multi-center clinical trials by accurately identifying clinical events at scale.

Effect of Sotagliflozin on Early Mortality and Heart Failure-Related Events: A Post Hoc Analysis of SOLOIST-WHF

BACKGROUND

Approximately 25% of patients admitted to hospitals for worsening heart failure (WHF) are readmitted within 30 days.

OBJECTIVES

The authors conducted a post hoc analysis of the SOLOIST-WHF (Effect of Sotagliflozin on Cardiovascular Events in Patients With Type 2 Diabetes Post-WHF) trial to evaluate the efficacy of sotagliflozin versus placebo to decrease mortality and HF-related events among patients who began study treatment on or before discharge from their index hospitalization.

METHODS

The main endpoint of interest was cardiovascular death or HF-related event (HF hospitalization or urgent care visit) occurring within 90 and 30 days after discharge for the index WHF hospitalization. Treatment comparisons were by proportional hazards models, generating HRs, 95% CIs, and P values.

RESULTS

Of 1,222 randomized patients, 596 received study drug on or before their date of discharge. Sotagliflozin reduced the main endpoint at 90 days after discharge (HR: 0.54 [95% CI: 0.35-0.82]; P = 0.004) and at 30 days (HR: 0.49 [95% CI: 0.27-0.91]; P = 0.023) and all-cause mortality at 90 days (HR: 0.39 [95% CI: 0.17-0.88]; P = 0.024). In subgroup analyses, sotagliflozin reduced the 90-day main endpoint regardless of sex, age, estimated glomerular filtration rate, N-terminal pro-B-type natriuretic peptide, left ventricular ejection fraction, or mineralocorticoid receptor agonist use. Sotagliflozin was well-tolerated but with slightly higher rates of diarrhea and volume-related events than placebo.

CONCLUSIONS

Starting sotagliflozin before discharge in patients with type 2 diabetes hospitalized for WHF significantly decreased cardiovascular deaths and HF events through 30 and 90 days after discharge, emphasizing the importance of beginning sodium glucose cotransporter treatment before discharge.

Leveraging electronic health record data for endometriosis research

Endometriosis is a chronic, complex disease for which there are vast disparities in diagnosis and treatment between sociodemographic groups. Clinical presentation of endometriosis can vary from asymptomatic disease-often identified during (in)fertility consultations-to dysmenorrhea and debilitating pelvic pain. Because of this complexity, delayed diagnosis (mean time to diagnosis is 1.7-3.6 years) and misdiagnosis is common. Early and accurate diagnosis of endometriosis remains a research priority for patient advocates and healthcare providers. Electronic health records (EHRs) have been widely adopted as a data source in biomedical research. However, they remain a largely untapped source of data for endometriosis research. EHRs capture diverse, real-world patient populations and care trajectories and can be used to learn patterns of underlying risk factors for endometriosis which, in turn, can be used to inform screening guidelines to help clinicians efficiently and effectively recognize and diagnose the disease in all patient populations reducing inequities in care. Here, we provide an overview of the advantages and limitations of using EHR data to study endometriosis. We describe the prevalence of endometriosis observed in diverse populations from multiple healthcare institutions, examples of variables that can be extracted from EHRs to enhance the accuracy of endometriosis prediction, and opportunities to leverage longitudinal EHR data to improve our understanding of long-term health consequences for all patients.

Leveraging electronic health records to streamline the conduct of cardiovascular clinical trials

Conventional randomized controlled trials (RCTs) can be expensive, time intensive, and complex to conduct. Trial recruitment, participation, and data collection can burden participants and research personnel. In the past two decades, there have been rapid technological advances and an exponential growth in digitized healthcare data. Embedding RCTs, including cardiovascular outcome trials, into electronic health record systems or registries may streamline screening, consent, randomization, follow-up visits, and outcome adjudication. Moreover, wearable sensors (i.e. health and fitness trackers) provide an opportunity to collect data on cardiovascular health and risk factors in unprecedented detail and scale, while growing internet connectivity supports the collection of patient-reported outcomes. There is a pressing need to develop robust mechanisms that facilitate data capture from diverse databases and guidance to standardize data definitions. Importantly, the data collection infrastructure should be reusable to support multiple cardiovascular RCTs over time. Systems, processes, and policies will need to have sufficient flexibility to allow interoperability between different sources of data acquisition. Clinical research guidelines, ethics oversight, and regulatory requirements also need to evolve. This review highlights recent progress towards the use of routinely generated data to conduct RCTs and discusses potential solutions for ongoing barriers. There is a particular focus on methods to utilize routinely generated data for trials while complying with regional data protection laws. The discussion is supported with examples of cardiovascular outcome trials that have successfully leveraged the electronic health record, web-enabled devices or administrative databases to conduct randomized trials.

Heart disease risk factors detection from electronic health records using advanced NLP and deep learning techniques

Heart disease remains the major cause of death, despite recent improvements in prediction and prevention. Risk factor identification is the main step in diagnosing and preventing heart disease. Automatically detecting risk factors for heart disease in clinical notes can help with disease progression modeling and clinical decision-making. Many studies have attempted to detect risk factors for heart disease, but none have identified all risk factors. These studies have proposed hybrid systems that combine knowledge-driven and data-driven techniques, based on dictionaries, rules, and machine learning methods that require significant human effort. The National Center for Informatics for Integrating Biology and Beyond (i2b2) proposed a clinical natural language processing (NLP) challenge in 2014, with a track (track2) focused on detecting risk factors for heart disease risk factors in clinical notes over time. Clinical narratives provide a wealth of information that can be extracted using NLP and Deep Learning techniques. The objective of this paper is to improve on previous work in this area as part of the 2014 i2b2 challenge by identifying tags and attributes relevant to disease diagnosis, risk factors, and medications by providing advanced techniques of using stacked word embeddings. The i2b2 heart disease risk factors challenge dataset has shown significant improvement by using the approach of stacking embeddings, which combines various embeddings. Our model achieved an F1 score of 93.66% by using BERT and character embeddings (CHARACTER-BERT Embedding) stacking. The proposed model has significant results compared to all other models and systems that we developed for the 2014 i2b2 challenge.

Virtual Care Team Guided Management of Patients With Heart Failure During Hospitalization

BACKGROUND

Scalable and safe approaches for heart failure guideline-directed medical therapy (GDMT) optimization are needed.

OBJECTIVES

The authors assessed the safety and effectiveness of a virtual care team guided strategy on GDMT optimization in hospitalized patients with heart failure with reduced ejection fraction (HFrEF).

METHODS

In a multicenter implementation trial, we allocated 252 hospital encounters in patients with left ventricular ejection fraction ≤40% to a virtual care team guided strategy (107 encounters among 83 patients) or usual care (145 encounters among 115 patients) across 3 centers in an integrated health system. In the virtual care team group, clinicians received up to 1 daily GDMT optimization suggestion from a physician-pharmacist team. The primary effectiveness outcome was in-hospital change in GDMT optimization score (+2 initiations, +1 dose up-titrations, -1 dose down-titrations, -2 discontinuations summed across classes). In-hospital safety outcomes were adjudicated by an independent clinical events committee.

RESULTS

Among 252 encounters, the mean age was 69 ± 14 years, 85 (34%) were women, 35 (14%) were Black, and 43 (17%) were Hispanic. The virtual care team strategy significantly improved GDMT optimization scores vs usual care (adjusted difference: +1.2; 95% CI: 0.7-1.8; P < 0.001). New initiations (44% vs 23%; absolute difference: +21%; P = 0.001) and net intensifications (44% vs 24%; absolute difference: +20%; P = 0.002) during hospitalization were higher in the virtual care team group, translating to a number needed to intervene of 5 encounters. Overall, 23 (21%) in the virtual care team group and 40 (28%) in usual care experienced 1 or more adverse events (P = 0.30). Acute kidney injury, bradycardia, hypotension, hyperkalemia, and hospital length of stay were similar between groups.

CONCLUSIONS

Among patients hospitalized with HFrEF, a virtual care team guided strategy for GDMT optimization was safe and improved GDMT across multiple hospitals in an integrated health system. Virtual teams represent a centralized and scalable approach to optimize GDMT.

Effect of Medically Tailored Meals on Clinical Outcomes in Recently Hospitalized High-Risk Adults

BACKGROUND

Inability to adhere to nutritional recommendations is common and linked to worse outcomes in patients with nutrition-sensitive conditions.

OBJECTIVES

The purpose of this study is to evaluate whether medically tailored meals (MTMs) improve outcomes in recently discharged adults with nutrition-sensitive conditions compared with usual care.

RESEARCH DESIGN

Remote pragmatic randomized trial.

SUBJECTS

Adults with heart failure, diabetes, or chronic kidney disease being discharged home between April 27, 2020, and June 9, 2021, from 5 hospitals within an integrated health care delivery system.

MEASURES

Participants were prerandomized to 10 weeks of MTMs (with or without virtual nutritional counseling) compared with usual care. The primary outcome was all-cause hospitalization within 90 days after discharge. Exploratory outcomes included all-cause and cause-specific health care utilization and all-cause death within 90 days after discharge.

RESULTS

A total of 1977 participants (MTMs: n=993, with 497 assigned to also receive virtual nutritional counseling; usual care: n=984) were enrolled. Compared with usual care, MTMs did not reduce all-cause hospitalization at 90 days after discharge [adjusted hazard ratio, aHR: 1.02, 95% confidence interval (CI), 0.86-1.21]. In exploratory analyses, MTMs were associated with lower mortality (aHR: 0.65, 95% CI, 0.43-0.98) and fewer hospitalizations for heart failure (aHR: 0.53, 95% CI, 0.33-0.88), but not for any emergency department visits (aHR: 0.95, 95% CI, 0.78-1.15) or diabetes-related hospitalizations (aHR: 0.75, 95% CI, 0.31-1.82). No additional benefit was observed with virtual nutritional counseling.

CONCLUSIONS

Provision of MTMs after discharge did not reduce risk of all-cause hospitalization in adults with nutrition-sensitive conditions. Additional large-scale randomized controlled trials are needed to definitively determine the impact of MTMs on survival and cause-specific health care utilization in at-risk individuals.

Contemporary Applications of Machine Learning for Device Therapy in Heart Failure

Despite a better understanding of the underlying pathogenesis of heart failure (HF), pharmacotherapy, surgical, and percutaneous interventions do not prevent disease progression in all patients, and a significant proportion of patients end up requiring advanced therapies. Machine learning (ML) is gaining wider acceptance in cardiovascular medicine because of its ability to incorporate large, complex, and multidimensional data and to potentially facilitate the creation of predictive models not constrained by many of the limitations of traditional statistical approaches. With the coexistence of "big data" and novel advanced analytic techniques using ML, there is ever-increasing research into applying ML in the context of HF with the goal of improving patient outcomes. Through this review, the authors describe the basics of ML and summarize the existing published reports regarding contemporary applications of ML in device therapy for HF while highlighting the limitations to widespread implementation and its future promises.

Contemporary outpatient management of patients with worsening heart failure with reduced ejection fraction: Rationale and design of the CHART-HF study

BACKGROUND

Patients with heart failure with reduced ejection fraction (HFrEF) and worsening HF events (WHFE) represent a distinct subset of patients with a substantial comorbidity burden, greater potential for intolerance to medical therapy, and high risk of subsequent death, hospitalization and excessive healthcare costs. Although multiple therapies have been shown to be efficacious and safe in this high-risk population, there are limited real-world data regarding factors that impact clinical decision-making when initiating or modifying therapy. Likewise, prior analyses of US clinical practice support major gaps in medical therapy for HFrEF and few medication changes during longitudinal follow-up, yet granular data on reasons why clinicians do not initiate or up-titrate guideline-directed medication are lacking.

METHODS

We designed the CHART-HF study, an observational study of approximately 1,500 patients comparing patients with and without WHFE (WHFE defined as receipt of intravenous diuretics in the inpatient, outpatient, or emergency department setting) who had an index outpatient visit in the US between 2017 and 2019. Patient-level data on clinical characteristics, clinical outcomes, and therapy will be collected from 2 data sources: a single integrated health system, and a national panel of cardiologists. Furthermore, clinician-reported rationale for treatment decisions and the factors prioritized with selection and optimization of therapies in real-world practice will be obtained. To characterize elements of clinician decision-making not documented in the medical record, the panel of cardiologists will review records of patients seen under their care to explicitly note their primary reason for initiating, discontinuing, and titrating medications specific medications, as well as the reason for not making changes to each medication during the outpatient visit.

CONCLUSIONS

Results from CHART-HF have the potential to detail real-world US practice patterns regarding care of patients with HFrEF with versus without a recent WHFE, to examine clinician-reported reasons for use and non-use of guideline-directed medical therapy, and to characterize the magnitude and nature of clinical inertia toward evidence-based medication changes for HFrEF.