Search

ImportanceEarly detection of atrial fibrillation (AF) may help prevent adverse cardiovascular events such as stroke. Deep learning applied to electrocardiograms (ECGs) has been successfully used for early identification of several cardiovascular diseases.ObjectiveTo determine whether deep learning models applied to outpatient ECGs in sinus rhythm can predict AF in a large and diverse patient population.Design, setting, and participantsThis prognostic study was performed on ECGs acquired from January 1, 1987, to December 31, 2022, at 6 US Veterans Affairs (VA) hospital networks and 1 large non-VA academic medical center. Participants included all outpatients with 12-lead ECGs in sinus rhythm.Main outcomes and measuresA convolutional neural network using 12-lead ECGs from 2 US VA hospital networks was trained to predict the presence of AF within 31 days of sinus rhythm ECGs. The model was tested on ECGs held out from training at the 2 VA networks as well as 4 additional VA networks and 1 large non-VA academic medical center.ResultsA total of 907 858 ECGs from patients across 6 VA sites were included in the analysis. These patients had a mean (SD) age of 62.4 (13.5) years, 6.4% were female, and 93.6% were male, with a mean (SD) CHA2DS2-VASc (congestive heart failure, hypertension, age, diabetes mellitus, prior stroke or transient ischemic attack or thromboembolism, vascular disease, age, sex category) score of 1.9 (1.6). A total of 0.2% were American Indian or Alaska Native, 2.7% were Asian, 10.7% were Black, 4.6% were Latinx, 0.7% were Native Hawaiian or Other Pacific Islander, 62.4% were White, 0.4% were of other race or ethnicity (which is not broken down into subcategories in the VA data set), and 18.4% were of unknown race or ethnicity. At the non-VA academic medical center (72 483 ECGs), the mean (SD) age was 59.5 (15.4) years and 52.5% were female, with a mean (SD) CHA2DS2-VASc score of 1.6 (1.4). A total of 0.1% were American Indian or Alaska Native, 7.9% were Asian, 9.4% were Black, 2.9% were Latinx, 0.03% were Native Hawaiian or Other Pacific Islander, 74.8% were White, 0.1% were of other race or ethnicity, and 4.7% were of unknown race or ethnicity. A deep learning model predicted the presence of AF within 31 days of a sinus rhythm ECG on held-out test ECGs at VA sites with an area under the receiver operating characteristic curve (AUROC) of 0.86 (95% CI, 0.85-0.86), accuracy of 0.78 (95% CI, 0.77-0.78), and F1 score of 0.30 (95% CI, 0.30-0.31). At the non-VA site, AUROC was 0.93 (95% CI, 0.93-0.94); accuracy, 0.87 (95% CI, 0.86-0.88); and F1 score, 0.46 (95% CI, 0.44-0.48). The model was well calibrated, with a Brier score of 0.02 across all sites. Among individuals deemed high risk by deep learning, the number needed to screen to detect a positive case of AF was 2.47 individuals for a testing sensitivity of 25% and 11.48 for 75%. Model performance was similar in patients who were Black, female, or younger than 65 years or who had CHA2DS2-VASc scores of 2 or greater.Conclusions and relevanceDeep learning of outpatient sinus rhythm ECGs predicted AF within 31 days in populations with diverse demographics and comorbidities. Similar models could be used in future AF screening efforts to reduce adverse complications associated with this disease.

AIMS: Sex differences in long-term outcomes following hospitalization for heart failure (HF) across ejection fraction (EF) subtypes are not well described. In this study, we evaluated the risk of mortality and rehospitalization among males and females across the spectrum of EF over 5 years of follow-up following an index HF hospitalization event. METHODS AND RESULTS: Patients hospitalized with HF between 1 January 2006 and 31 December 2014 from the American Heart Associations Get With The Guidelines-Heart Failure registry with available 5-year follow-up using Medicare Part A claims data were included. The association between sex and risk of mortality and readmission over a 5-year follow-up period for each HF subtype (HF with reduced EF [HFrEF, EF ≤40%], HF with mildly reduced EF [HFmrEF, EF 41-49%], and HF with preserved EF [HFpEF, EF >50%]) was assessed using adjusted Cox models. The effect modification by the HF subtype for the association between sex and outcomes was assessed by including multiplicative interaction terms in the models. A total of 155 670 patients (median age: 81 years, 53.4% female) were included. Over 5-year follow-up, males and females had comparably poor survival post-discharge; however, females (vs. males) had greater years of survival lost to HF compared with the median age- and sex-matched US population (HFpEF: 17.0 vs. 14.6 years; HFrEF: 17.3 vs. 15.1 years; HFmrEF: 17.7 vs. 14.6 years for age group 65-69 years). In adjusted analysis, females (vs. males) had a lower risk of 5-year mortality (adjusted hazard ratio [aHR] 0.89, 95% confidence interval [CI] 0.87-0.90, p

BackgroundThe burden of heart failure is growing. Guideline-directed medical therapies (GDMT) reduce adverse outcomes in heart failure with reduced ejection fraction (HFrEF). Whether there is geographic variation in HFrEF quality of care is not well described.ObjectivesThis study evaluated variation nationally for prescription of GDMT within the Veterans Health Administration.MethodsA cohort of Veterans with HFrEF had their address linked to hospital referral regions (HRRs). GDMT prescription was defined using pharmacy data between July 1, 2020, and July 1, 2021. Within HRRs, we calculated the percentage of Veterans prescribed GDMT and a composite GDMT z-score. National choropleth maps were created to evaluate prescription variation. Associations between GDMT performance and demographic characteristics were evaluated using linear regression.ResultsMaps demonstrated significant variation in the HRR composite score and GDMT prescriptions. Within HRRs, the prescription of beta-blockers to Veterans was highest with a median of 80% (IQR: 77.3%-82.2%) followed by angiotensin-converting enzyme inhibitor/angiotensin receptor blocker/angiotensin receptor-neprilysin inhibitors (69.3%; IQR: 66.4%-72.1%), sodium-glucose cotransporter 2 inhibitors (10.3%; IQR: 7.7%-12.8%), mineralocorticoid receptor antagonists (29.2%; IQR: 25.8%-33.9%), and angiotensin receptor-neprilysin inhibitors (12.2%; IQR: 8.6%-15.3%). HRR composite GDMT z-scores were inversely associated with the HRR median Gini coefficient (R = -0.13; P = 0.0218) and the percentage of low-income residents (R = -0.117; P = 0.0413).ConclusionsWide geographic differences exist for HFrEF care. Targeted strategies may be required to increase GDMT prescription for Veterans in lower-performing regions, including those affected by income inequality and poverty.

BACKGROUND: Patient-reported outcomes (PROs) may improve care for patients with heart failure. The Kansas City Cardiomyopathy Questionnaire-12 (KCCQ-12) is a patient survey that captures symptom frequency, symptom burden, physical limitations, social limitations, and quality of life. Despite the utility of PROs and the KCCQ-12, the implementation and routine use of these measures can be difficult. We conducted an evaluation of clinician perceptions of the KCCQ-12 to identify barriers and facilitators to implementation into clinical practice. METHODS: We conducted interviews with cardiologists from 4 institutions across the United States and Canada (n=16) and observed clinic visits at 1 institution in Northern California (n=5). Qualitative analysis was conducted in 2 rounds: (1) rapid analysis constructed around major themes related to the aims of the study and (2) content analysis with codes derived from the rapid analysis and implementation science. RESULTS: Most heart failure physicians and advanced practice clinicians reported that the KCCQ-12 was acceptable, appropriate, and useful in clinical care. Clinician engagement efforts, trialability, and the straightforward design of the KCCQ-12 facilitated its use in clinical care. Further opportunities identified to facilitate implementation include more streamlined integration into the electronic health record and comprehensive staff education on PROs. Participants highlighted that the KCCQ-12 was useful in clinic visits to improve the consistency of patient history taking, focus patient-clinician conversations, collect a more accurate account of patient quality of life, track trends in patient well-being over time, and refine clinical decision-making. CONCLUSIONS: In this qualitative study, clinicians reported that the KCCQ-12 enhanced several aspects of heart failure patient care. Use of the KCCQ-12 was facilitated by a robust clinician engagement campaign and the design of the KCCQ-12 itself. Future implementation of PROs in heart failure clinic should focus on streamlining electronic health record integration and providing additional staff education on the value of PROs. REGISTRATION: URL: https://clinicaltrials.gov; Unique identifier: NCT04164004.

Introduction: In patients with heart failure with reduced ejection fraction (HFrEF), use of guideline directed medical therapies (GDMT) reduces mortality. Whether there is geographic variation in GDMT prescribing is not well-characterized. Hypothesis: We assessed the hypothesis that there is wide geographic variation nationally within the Veterans Affairs (VA) healthcare system for the receipt of GDMT. Methods: We linked the primary residence address of a cohort of Veterans with HFrEF receiving care at VA facilities (n=178,856) to hospital referral regions (HRRs). Using VA and non-VA pharmacy data between July 1, 2020 and July 1, 2021, we defined receipt of GDMT. For each HRR, we calculated the percentage of eligible Veterans that were prescribed each class of GDMT and a composite GDMT z-score. Lastly, we constructed national choropleth maps to depict HRR geographic variation in composite z-scores and prescription of each class of GDMT. Results: There was significant variation in individual GDMT class prescription and composite score across HRRs as shown in choropleth maps ( Figure) . Within HRRs, beta-blocker prescription was highest, with a median 80% of Veterans (interquartile range [IQR] 77.3% to 82.2%), afterload-reducing agents including angiotensin converting enzyme inhibitor / angiotensin receptor blocker / angiotensin receptor-neprilysin inhibitors (ARNI) 69.3% (IQR 66.4% to 72.1%), mineralocorticoid receptor antagonists (MRA) 29.2% (IQR 25.8% to 33.9%), sodium-glucose cotransporter 2 inhibitors (SGLT2I) 10.3% (IQR 7.7% to 12.8%), and ARNI 12.2% (IQR 8.6% to 15.3%). HRRs with the highest and lowest composite scores were often found in the same U.S. Census Regions. Conclusions: In conclusion, wide geographic disparities are present for GDMT prescribing to Veterans across HRRs. Veteran receipt of ARNI, SGLT2I, and MRA is low nationwide. Targeted approaches may be necessary to improve GDMT prescription for Veterans in lower performing HRRs.

BackgroundWith rising United States health care expenditure, estimating current spending for patients with heart failure (HF) informs the value of preventative health interventions.ObjectivesThe purpose of this study was to estimate current health care expenditure growth for patients with HF in the United States.MethodsThe authors pooled MEPS (Medical Expenditure Panel Survey) data from 2009-2018 to calculate total HF-related expenditure across clinical settings in the United States. A 2-part model adjusted for demographics, comorbidities, and year was used to estimate annual mean and incremental expenditures associated with HF.ResultsIn the United States, an average of $28,950 (2018 inflation-adjusted dollars) is spent per year for health care-related expenditure for individuals with HF compared with $5,727 for individuals without HF. After adjusting for demographics and comorbidities, a diagnosis of HF was associated with $3,594 in annual incremental expenditure compared with those without HF. HF-related expenditure increased from $26,864 annual per person in 2009-2010 to $32,955 in 2017-2018, representing a 23% rise over 10 years. In comparison, expenditure on myocardial infarction, type 2 diabetes mellitus, and cancer grew by 16%, 28%, and 16%, respectively. Most of the cost was related to hospitalization: $12,569 per year. Outpatient office-based care and prescription medications saw the greatest growth in cost over the period, 41% and 24%, respectively. Estimated incremental national expenditure for HF per year was $22.3 billion; total annual expenditure for adults with HF was $179.5 billion.ConclusionsHF is a costly condition for which expenditure is growing faster than that of other chronic conditions.

The cost of heart failure care is high owing to the cost of hospitalization and chronic treatments. Heart failure treatments vary in their benefit and cost. The cost effectiveness of therapies can be determined by comparing the cost of treatment required to obtain a certain benefit, often defined as an increase in 1 year of life. This review was sponsored by the Heart Failure Society of America and describes the growing economic burden of heart failure for patients and the health care system in the United States. It also provides a summary of the cost effectiveness of drugs, devices, diagnostic tests, hospital care, and transitions of care for patients with heart failure. Many medications that are no longer under patent are inexpensive and highly cost-effective. These include angiotensin-converting enzyme inhibitors, beta-blockers and mineralocorticoid receptor antagonists. In contrast, more recently developed medications and devices, vary in cost effectiveness, and often have high out-of-pocket costs for patients.