Treatment Patterns, Outcomes, and Persistence to Newly Started Heart Failure Medications in Patients with Worsening Heart Failure: A Cohort Study from the United States and Germany.

BACKGROUND: Data are limited regarding guideline-directed medical therapy (GDMT) treatment patterns in patients with worsening heart failure (HF). METHODS: We used administrative claims databases in Germany and the USA to conduct a retrospective cohort study of patients with worsening HF. Two cohorts of patients with prevalent HF and a HF hospitalization (HFH) from 2016 to 2019, alive at discharge (N = 75,140 USA; N = 47,003 Germany) were identified. Index date was the first HFH during the study period. One-year HF rehospitalization and mortality rates were calculated and a composite endpoint of both outcomes assessed using Kaplan-Meier estimation. We evaluated HF medication patterns in the 6 months before and after the index date. New users of a HF medication (at discharge/after index HFH) were followed for 1 year to evaluate persistence (no treatment gaps > 2 months) RESULTS: One-year HF rehospitalization rates were 36.2% (USA) and 47.7% (Germany). One year mortality rates were 30.0% (USA) and 23.0% (Germany), and the composite endpoint (mortality/HF rehospitalization) was reached in 55.1 % (USA) and 56.6% (Germany). Kaplan-Meier plots showed the risk for the composite endpoint was high in the early post discharge period. Comparison of patterns pre- and postindex HFH showed some increase in use of mineralocorticoid receptor antagonists (MRAs), angiotensin receptor-neprilysin inhibitor (ARNI), and triple therapy; use of angiotensin-converting enzyme (ACE) inhibitor/ angiotensin receptor blocker (ARB) plus beta-blockers remained constant/slightly declined; < 20% patients received triple therapy (ACE inhibitor/ARB plus beta-blocker plus MRA). A third of patients were new users; 1 year persistence rates were often low. CONCLUSIONS: Morbidity, mortality, and rehospitalization risk is high among patients with worsening HF; uptake and continuation of GDMT is suboptimal.

Initiation and continuation of pharmacological therapies in patients hospitalized for heart failure in Japan.

Currently, the utilization patterns of medications for heart failure (HF) after worsening HF events remain unelucidated in Japan. Here, we conducted a retrospective cohort study evaluating the changes in HF drug utilization patterns in 6 months before and after hospitalizations for HF. The adherence to newly initiated HF medications was evaluated based on the proportion of days covered (PDC) and persistence as continuous treatment episodes among new users. The study included 9091 patients hospitalized for HF between January 2016 and September 2019, including 2735 (30.1%) patients who were newly prescribed at least one HF medication after hospitalization. Despite increases in the use of foundational HF therapy (beta-blockers, angiotensin-converting-enzyme inhibitors/angiotensin receptor blockers, or mineralocorticoid receptor antagonists), 35.6% and 7.6% of patients were treated with the HF foundational monotherapy or diuretics alone after hospitalization, respectively. The mean PDC of newly initiated HF medications ranged from 0.57 for thiazide diuretics to 0.77 for sodium-glucose cotransporter-2 inhibitors. Continuous use of HF medications during the first year after initiation was observed in 30-60% of patients. The mean PDC and one-year continuous HF medication use were consistently lower in patients aged ≥ 75 years and in patients with a history of HF hospitalization for all HF medication classes except for tolvaptan and digoxin. Despite the guideline recommendations of HF pharmacotherapy, both treatment and adherence were suboptimal after HF hospitalization, especially in vulnerable populations such as older patients and those with prior HF hospitalizations.

Real-world characteristics and use patterns of patients treated with vericiguat: A nationwide longitudinal cohort study in Germany.

PURPOSE: Vericiguat reduced clinical endpoints in patients experiencing worsening heart failure in clinical trials, but its implementation outside trials is unclear. METHODS: This retrospective analysis of longitudinally collected data was based on the IQVIA™ LRx database, which includes ~ 80% of the prescriptions of the 73 million people covered by the German statutory health insurance. RESULTS: Between September 2021 and December 2022, vericiguat was initiated in 2916 adult patients. Their mean age was 73 ± 13 years and 28% were women. While approximately 70% were uptitrated beyond 2.5 mg, only 36% reached 10 mg. Median time to up-titration from 2.5 mg to 5 mg was 17 (quartiles: 11-33) days, and from 2.5 to 10 mg 37 (25-64) days, respectively. In 87% of the patients, adherence to vericiguat was high as indicated by a medication possession ratio of  ≥ 80%, and 67% of the patients persistently used vericiguat during the first year. Women and older patients reached the maximal dose of 10 mg vericiguat less often and received other substance classes of guideline-recommended therapy (GDMT) less frequently. The proportion of patients receiving four pillars of GDMT increased from 29% before vericiguat initiation to 44% afterwards. CONCLUSION: In a real-world setting, despite higher age than in clinical trials, adherence and persistence of vericiguat appeared satisfactory across age categories. Initiation of vericiguat was associated with intensification of concomitant GDMT. Nevertheless, barriers to vericiguat up-titration and implementation of other GDMT, applying in particular to women and elderly patients, need to be investigated further.

Patient Perspectives on the Burden of Heart Failure with Preserved Ejection Fraction in a US Commercially Insured and Medicare Advantage Population: A Survey Study.

Background: Patient-reported health related quality of life (HRQOL) is not routinely assessed in clinical practice. Little is known about health status outcomes reported by patients with heart failure with preserved ejection fraction (HFpEF) in non-clinical trial settings. Purpose: To better understand patient burden of HFpEF in terms of HF-specific functional and symptom status, HRQOL, healthcare resource utilization (HCRU) and costs in a US-based commercial and Medicare Advantage insured population. Patients and methods: We conducted a cross-sectional survey of patients with HFpEF and linked their survey and administrative claims data. Consenting, eligible patients completed a survey that included the 23-item Kansas City Cardiomyopathy Questionnaire (KCCQ-23) and the PROMIS Global Health-10 (GH-10) questionnaire, as well as clinical and demographic questions. HF medication use, HCRU and costs during the 12-month baseline period before the survey were determined from claims data. Generalized linear regression was used to assess the associations between baseline characteristics and the KCCQ-23 overall summary score. Results: Of 598 survey respondents with survey and claims data, 54.7% were female with mean age 74.0 years. The KCCQ-23 overall summary and clinical summary scores were 64.8 and 63.0, respectively, and the GH-10 physical and mental health summary scores were 39.9 and 45.5. Factors related to lower KCCQ-23 overall summary scores were HF treatment and symptom changes during the past 4-weeks before the survey, hospital admission during the past year, low household income, high comorbidity index, and morbid obesity (BMI>40). Total all-cause healthcare costs were $38,243 during the year prior to the survey, of which 42% were HF-related. Conclusion: Patient-reported outcome measure scores indicated impairment due to HF symptoms and physical limitations in this real-world sample of patients with HFpEF, highlighting a need to assess patient-reported outcomes as well as the clinical and economic outcomes traditionally assessed by clinicians, health systems and payers.

Electronic health record characterization and outcomes of heart failure with preserved ejection fraction.

BACKGROUND: Electronic health record (EHR)-based identification of heart failure with preserved ejection fraction (HFpEF) in the clinical setting may facilitate screening for clinical trials by improving the understanding of its epidemiology and outcomes; yet, previous data have yielded variable results. We sought to characterize groups identified with HFpEF by different EHR screening strategies and their associated long-term outcomes across a large and diverse population. METHODS: We retrospectively analyzed 116,499 consecutive patients from an academic referral center who underwent echocardiography, and 9,263 patients who underwent echocardiography within 6 months of right heart catheterization (RHC), between 2008 and 2018. EHR-based screening strategies identified patients with HFpEF using 1) International Classification of Diseases (ICD)-9/10 codes, 2) H2FpEF score ≥6 and ejection fraction (EF) ≥50%, or 3) RHC wedge pressure ≥15 mmHg and EF ≥50%, when available. Primary outcomes were 1) cumulative incident heart failure hospitalization (HFH), and 2) death, over 10 years. RESULTS: There were 33,461 (29%) patients who met either ICD or H2FpEF-HFpEF definition, of whom 5,310 (16%) met both criteria. Compared to ICD-HFpEF, patients with H2FpEF-HFpEF were more likely older (median age 72 vs 67), White (78% vs 64%), and had atrial fibrillation (97% vs 41%). Among those also with RHC, 6,353 (69%) patients met any HFpEF criteria, of whom only 783 (12%) satisfied all three criteria. Female sex was more common among RHC-HFpEF (55%) compared to other methods (H2FpEF-HFpEF, 47%; ICD-HFpEF, 43%). Atrial fibrillation was substantially higher among HFpEF identified by the H2FpEF score (97%) compared to other methods (49% for ICD and 47% for RHC). Across HFpEF screening methods, 10-year cumulative incidence rates for HFH was 32% to 45% for echocardiography only and 43% to 52% for echocardiography and RHC populations; 10-year risk of death was 54% to 56% for echocardiography only and 52% to 57% for echocardiography and RHC populations. CONCLUSIONS: Different EHR-based HFpEF definitions identified cohorts with modest overlap and varying baseline characteristics. Yet, long-term risk for HFH and death were similarly high for cohorts identified among both populations undergoing echocardiography only or echocardiography and RHC. These data aid in identifying relevant subgroups in clinical trials of HFpEF.

Predictive Risk Models to Identify Patients at High-Risk for Severe Clinical Outcomes With Chronic Kidney Disease and Type 2 Diabetes.

INTRODUCTION/OBJECTIVE: Predictive risk models identifying patients at high risk for specific outcomes may provide valuable insights to providers and payers regarding points of intervention and modifiable factors. The goal of our study was to build predictive risk models to identify patients with chronic kidney disease (CKD) and type 2 diabetes (T2D) at high risk for progression to end stage kidney disease (ESKD), mortality, and hospitalization for cardiovascular disease (CVD), cerebrovascular disease (CeVD), and heart failure (HF). METHODS: This was a retrospective observational cohort study utilizing administrative claims data in patients with CKD (stage 3-4) and T2D aged 65 to 89 years enrolled in a Medicare Advantage Drug Prescription plan offered by Humana Inc. between 1/1/2012 and 12/31/2017. Patients were enrolled ≥1 year pre-index and followed for outcomes, including hospitalization for CVD, CeVD and HF, ESKD, and mortality, 2 years post-index. Pre-index characteristics comprising demographic, comorbidities, laboratory values, and treatment (T2D and cardiovascular) were evaluated and included in the models. LASSO technique was used to identify predictors to be retained in the final models followed by logistic regression to generate parameter estimates and model performance statistics. Inverse probability censoring weighting was used to account for varying follow-up time. RESULTS: We identified 169 876 patients for inclusion. Declining estimated glomerular filtration rate (eGFR) increased the risk of hospitalization for CVD (38.6%-61.8%) and HF (2-3 times) for patients with eGFR 15 to 29 mL/min/1.73 m2 compared to patients with eGFR 50 to 59 mL/min/1.73 m2. Patients with urine albumin-to-creatinine ratio (UACR) ≥300 mg/g had greater chance for hospitalization for CVD (2.0 times) and HF (4.9 times), progression to ESKD (2.9 times) and all-cause mortality (2.4 times) than patients with UACR <30 mg/g. Elevated hemoglobin A1c (≥8%) increased the chances for hospitalization for CVD (21.3%), CeVD (45.4%), and death (20.6%). Among comorbidities, history of HF increased the risk for ESKD, mortality, and hospitalization for CVD, CeVD, and HF. CONCLUSIONS: The predictive models developed in this study could potentially be used as decision support tools for physicians and payers, and the risk scores from these models can be applied to future outcomes studies focused on patients with T2D and CKD.

Glomerular filtration rate change and outcomes in type 2 diabetes.

OBJECTIVES: To assess the relationship between relative estimated glomerular filtration rate (eGFR) change and outcomes in patients with type 2 diabetes (T2D). STUDY DESIGN: This retrospective cohort study utilized administrative claims (Humana Research Database) for patients with T2D aged 65 to 89 years, enrolled in a Medicare Advantage plan, with an initial eGFR of 25 to 89 mL/min/1.73m2 in 2008 to 2017, and a second eGFR measurement within 3 to 24 months after the identification date. METHODS: The primary exposure was relative decline in eGFR of 40% or more in a 2-year period. Outcomes included end-stage kidney disease (ESKD) or kidney failure, a composite cardiovascular (CV) outcome, and all-cause mortality assessed with multivariable adjusted survival models. Days out of the home and all-cause total costs were assessed using multivariable adjusted generalized linear models. RESULTS: A total of 288,170 patients were included. The adjusted HR for ESKD or kidney failure was 4.38 (95% CI, 3.99-4.81) in patients with 40% or greater decline versus those with a decline of less than 40%. The adjusted HR was 1.67 (95% CI, 1.53-1.82) for the composite CV outcome and 1.98 (95% CI, 1.87-2.10) for all-cause mortality. Patients with a 40% or greater relative decline had 2.23 times higher all-cause total per patient per month costs ($1910 difference) and 1.82 times higher odds of 7 or more days out of the home versus those with less than 40% relative eGFR decline. CONCLUSIONS: Our results indicate that a relative eGFR decline of 40% or greater is associated with an increased risk of ESKD or kidney failure, CV outcomes and all-cause mortality, and increased health care resource utilization and costs.

Adherence to Chronic Kidney Disease Screening Guidelines Among Patients With Type 2 Diabetes in a US Administrative Claims Database.

OBJECTIVE: To examine the screening rates for kidney damage and function among patients with type 2 diabetes (T2D) and chronic kidney disease stage at diabetes diagnosis using a US administrative claims database. PATIENTS AND METHODS: This cohort study used a claims database enriched with laboratory results data. Patients with T2D (defined as 1 inpatient or 2 outpatient claims for diabetes), aged 18 years or older, and with at least 1 year of follow-up enrollment were identified. Patients with type 1 diabetes, kidney disease, or other related conditions at baseline were excluded. We estimated screening rates using laboratory orders for serum creatinine and estimated glomerular filtration rate (eGFR) measurement and urine albumin to creatinine ratio (UACR). Chronic kidney disease severity was reported using the Kidney Disease: Improving Global Outcomes classification based on laboratory results. RESULTS: A total of 1,881,447 patients with T2D were eligible for analysis. Mean ± SD age was 63.1±13.1 years; 947,150 patients (50.3%) were male. Serum creatinine tests were ordered within 14 days of the index date among 290,722 patients of 622,915 (46.7%) patients with newly-recognized T2D. Overall, 1,595,964 patients (84.8%) had at least one serum creatinine test ordered during the 1-year follow-up period. Fewer patients received a UACR test during follow-up (814,897 [43.3%]). Less than half of all patients with T2D received a laboratory test order for both serum creatinine and urine albumin measurements during the follow-up period. CONCLUSION: Physicians treating patients with diabetes are selectively adhering to chronic kidney disease screening guidelines, as indicated by high rates of eGFR testing, but less frequent UACR testing. Despite recommendations to monitor both eGFR and UACR, less than half of patients were screened for albuminuria during the 1-year follow-up.

Phenotypic clustering of heart failure with preserved ejection fraction reveals different rates of hospitalization.

AIMS: Approximately 50% of patients with heart failure have preserved (≥50%) ejection fraction (HFpEF). Improved understanding of the phenotypic heterogeneity of HFpEF might facilitate development of targeted therapies and interventions. METHODS: This retrospective study characterized a cohort of patients with HFpEF based on similar clinical profiles and evaluated 1-year heart failure related hospitalization. Enrolment, medical and pharmacy data were used to identify patients newly diagnosed with heart failure enrolled in a Medicare Advantage Prescription Drug or commercial healthcare plan. To identify only those patients with HFpEF, we used natural language processing techniques of ejection fraction values abstracted from a linked free-text clinical notes data source. The study population comprised 1515 patients newly identified with HFpEF between 1 January 2011 and 31 December 2015. RESULTS: Using unsupervised machine learning, we identified three distinguishable patient clusters representing different phenotypes: cluster-1 patients had the lowest prevalence of heart failure comorbidities and highest mean age; cluster-2 patients had higher prevalence of metabolic syndrome and pulmonary disease, despite younger mean age; and cluster-3 patients had higher prevalence of cardiac arrhythmia and renal disease. Cluster-3 had the highest 1-year heart failure related hospitalization rates. Within-cluster analysis, prior use of diuretics (cluster-1 and cluster-2) and age (cluster-2 and cluster-3) was associated with 1-year heart failure related hospitalization. Combination therapy was associated with decreased 1-year heart failure related hospitalization in cluster-1. CONCLUSION: This study demonstrated that clustering can be used to characterize subgroups of patients with newly identified HFpEF, assess differences in heart failure related hospitalization rates at 1 year and suggest patient subtypes may respond differently to treatments or interventions.

Real-World Treatment Patterns, Healthcare Resource Utilization, and Costs for Patients with Newly Diagnosed Systolic versus Diastolic Heart Failure.

BACKGROUND: Although the significant burden of heart failure (HF) is well recognized, the relative contributions of systolic HF versus diastolic HF are less defined. OBJECTIVE: To explore the differential burden between patients with systolic and diastolic HF in terms of treatment patterns, healthcare resource utilization (HCRU), costs, and mortality risk. METHODS: This retrospective cohort study used administrative claims data from a large US commercial health insurer integrated with mortality data. Patients newly diagnosed with HF between January 1, 2010, and June 30, 2016, were identified and grouped according to systolic HF or diastolic HF diagnosis and were followed up to 4 years after diagnosis. Treatment patterns, HCRU, costs, and mortality were compared between the 2 groups of patients. RESULTS: Overall, 46,885 patients with systolic HF and 21,854 with diastolic HF were identified and included in the study. Patients with systolic HF had less HCRU than those with diastolic HF during the first year after HF diagnosis, including hospital admissions (70.2% vs 82.4%, respectively; P <.001) and emergency department visits (30.5% vs 39.1%, respectively; P <.001). The average per-patient costs for patients with systolic HF during the 1-year follow-up were higher than for those with diastolic HF ($64,154 vs $59,652, respectively; P <.001), but lower during years 2 through 4 (approximately $23,000-$25,000 annually vs approximately $28,000-$29,000 annually; P <.001). Patients with diastolic HF had a higher adjusted hospitalization risk (odds ratio, 1.62; 95% confidence interval [CI], 1.55-1.69), but comparable adjusted costs (exponentiated estimate, 1.01; 95% CI, 0.99-1.02) and slightly lower mortality risk (hazard ratio, 0.96; 95% CI, 0.93-0.99) versus patients with systolic HF. The number of HF-related medication classes received for other diagnoses during the year preceding an HF diagnosis was associated with lower risks for hospitalization, mortality, and lower costs, with a trend in benefits toward patients with systolic HF. Of note, 21.9% of patients with systolic HF and 25% of patients with diastolic HF filled no HF-related prescriptions in the year after diagnosis. CONCLUSION: This real-world analysis confirms a high disease burden associated with HF and provides insight across the systolic HF and diastolic HF phenotypes. HF-related medication use after diagnosis was suboptimal and underscores a gap in patient care.

Risk Factors for Heart Failure with Preserved or Reduced Ejection Fraction Among Medicare Beneficiaries: Application of Competing Risks Analysis and Gradient Boosted Model.

BACKGROUND: The differential impact of various demographic characteristics and comorbid conditions on development of heart failure (HF) with preserved (pEF) and reduced ejection fraction (rEF) is not well studied among the elderly. METHODS: Using Medicare claims data linked to electronic health records, we conducted an observational cohort study of individuals ≥65 years of age without HF. A Cox proportional hazards model accounting for competing risk of HFrEF and HFpEF incidence was constructed. A gradient-boosted model (GBM) assessed the relative influence (RI) of each predictor in the development of HFrEF and HFpEF. RESULTS: Among 138,388 included individuals, 9701 developed HF (incidence rate = 20.9 per 1000 person-years). Males were more likely to develop HFrEF than HFpEF (HR = 2.07, 95% CI: 1.81-2.37 vs. 1.11, 95% CI: 1.02-1.20, P for heterogeneity <0.01). Atrial fibrillation and pulmonary hypertension had stronger associations with the risk of HFpEF (HR = 2.02, 95% CI: 1.80-2.26 and 1.66, 95% CI: 1.23-2.22) while cardiomyopathy and myocardial infarction were more strongly associated with HFrEF (HR = 4.37, 95% CI: 3.21-5.97 and 1.94, 95% CI: 1.23-3.07). Age was the strongest predictor across all HF subtypes with RI from GBM >35%. Atrial fibrillation was the most influential comorbidity for the development of HFpEF (RI = 8.4%) while cardiomyopathy was the most influential comorbidity for the development of HFrEF (RI = 20.7%). CONCLUSION: These findings of heterogeneous relationships between several important risk factors and heart failure types underline the potential differences in the etiology of HFpEF and HFrEF.

Using Real-World Data to Predict Clinical and Economic Benefits of a Future Drug Based on its Target Product Profile.

INTRODUCTION: For a new drug to be developed, the desired properties are described in a target product profile. OBJECTIVE: We propose a framework for using real-world data to measure the disease-specific costs of the current standard of care and then to project the costs of the proposed new product for early data-driven portfolio decisions to select drug candidates for development. METHODS: We sampled from a cohort of patients representing the current standard of care to generate a hypothetical cohort of patients that fits a given target product profile for a new (hypothetical) treatment. The healthcare costs were determined and compared between standard of care and the new treatment. The approach differed according to the number of outcomes defined in the target product profile, and the cases for one, two, and three outcome variables are described. RESULTS: Based on assumed hypothetical treatment effect, absolute risk and cost reductions were estimated in a worked example. The median costs per day for one patient were estimated to be $10.37 and $8.39 in the original and hypothetical cohorts, respectively. This means that the assumed target product profile would result in cost savings of $1.98 per day and patient-not accounting for any additional drug costs. CONCLUSIONS: We present a simple approach to assess the potential absolute clinical and economic benefit of a new drug based on real-world data and its target product profile. The approach allows for early data-driven portfolio decisions to select drug candidates based on their expected cost savings.

Economic burden of coronary artery disease or peripheral artery disease in patients at high risk of ischemic events in the French setting: a claims database analysis.

Aims: Estimate the direct costs of high-risk patients presenting with coronary artery disease (CAD) or peripheral artery disease (PAD) in France.Materials and methods: This retrospective cohort study used a representative claims database, the "Echantillon Généraliste de Bénéficiaires" (EGB), to identify patients presenting with CAD or PAD between 2011 and 2016. Among those, patients meeting the COMPASS trial selection criteria were selected, as well as controls matched on age and sex. Direct costs (Euros 2016) were estimated in a societal perspective by comparing case and controls.Results: The adult population presenting with CAD or PAD in the EGB in 2016 was estimated at 29,888 individuals, representing a crude prevalence rate of 5.44%. After using the documented selection criteria of the COMPASS study, this population (COMPASS-like) was estimated at 17,369 individuals (58.1% of the CAD and/or PAD total population). Among them, a proportion of 11.5% presented with CAD + PAD. Compared with the original COMPASS population, patients were older (76.5 vs 68.2 years) and with a lower male predominance (60.0% vs 78.2% males). Compared with controls, the COMPASS-like population was characterized by a higher annual mortality (5.9% vs 3.5%) and the presence of more comorbidities on top of CAD and/or PAD. The annual per capita extra direct cost of the COMPASS-like population was estimated at €4,284, with a main contribution from inpatient care (58.9%). This extra cost was higher in the PAD ± CAD sub-group (€5,552) and the CAD + PAD sub-group (€8,067).Limitations: The EGB had limitations about several clinical features defining high-risk patients that may lead to bias in our estimates.Conclusions: Due to the high prevalence of CAD and/or PAD and the associated high unit costs, this population generates a significant economic burden, which is higher among patients with PAD and in those presenting simultaneously with both conditions.

Comparison of Machine Learning Methods With Traditional Models for Use of Administrative Claims With Electronic Medical Records to Predict Heart Failure Outcomes.

Importance: Accurate risk stratification of patients with heart failure (HF) is critical to deploy targeted interventions aimed at improving patients' quality of life and outcomes. Objectives: To compare machine learning approaches with traditional logistic regression in predicting key outcomes in patients with HF and evaluate the added value of augmenting claims-based predictive models with electronic medical record (EMR)-derived information. Design, Setting, and Participants: A prognostic study with a 1-year follow-up period was conducted including 9502 Medicare-enrolled patients with HF from 2 health care provider networks in Boston, Massachusetts ("providers" includes physicians, clinicians, other health care professionals, and their institutions that comprise the networks). The study was performed from January 1, 2007, to December 31, 2014; data were analyzed from January 1 to December 31, 2018. Main Outcomes and Measures: All-cause mortality, HF hospitalization, top cost decile, and home days loss greater than 25% were modeled using logistic regression, least absolute shrinkage and selection operation regression, classification and regression trees, random forests, and gradient-boosted modeling (GBM). All models were trained using data from network 1 and tested in network 2. After selecting the most efficient modeling approach based on discrimination, Brier score, and calibration, area under precision-recall curves (AUPRCs) and net benefit estimates from decision curves were calculated to focus on the differences when using claims-only vs claims + EMR predictors. Results: A total of 9502 patients with HF with a mean (SD) age of 78 (8) years were included: 6113 from network 1 (training set) and 3389 from network 2 (testing set). Gradient-boosted modeling consistently provided the highest discrimination, lowest Brier scores, and good calibration across all 4 outcomes; however, logistic regression had generally similar performance (C statistics for logistic regression based on claims-only predictors: mortality, 0.724; 95% CI, 0.705-0.744; HF hospitalization, 0.707; 95% CI, 0.676-0.737; high cost, 0.734; 95% CI, 0.703-0.764; and home days loss claims only, 0.781; 95% CI, 0.764-0.798; C statistics for GBM: mortality, 0.727; 95% CI, 0.708-0.747; HF hospitalization, 0.745; 95% CI, 0.718-0.772; high cost, 0.733; 95% CI, 0.703-0.763; and home days loss, 0.790; 95% CI, 0.773-0.807). Higher AUPRCs were obtained for claims + EMR vs claims-only GBMs predicting mortality (0.484 vs 0.423), HF hospitalization (0.413 vs 0.403), and home time loss (0.575 vs 0.521) but not cost (0.249 vs 0.252). The net benefit for claims + EMR vs claims-only GBMs was higher at various threshold probabilities for mortality and home time loss outcomes but similar for the other 2 outcomes. Conclusions and Relevance: Machine learning methods offered only limited improvement over traditional logistic regression in predicting key HF outcomes. Inclusion of additional predictors from EMRs to claims-based models appeared to improve prediction for some, but not all, outcomes.

Using Real-World Data to Extrapolate Evidence From Randomized Controlled Trials.

Randomized controlled trials (RCTs) provide evidence for regulatory agencies, shape clinical practice, influence formulary decisions, and have important implications for patients. However, many patient groups that are major consumers of drugs are under-represented in randomized trials. We review three methods to extrapolate evidence from trial participants to different target populations following market approval and discuss how these could be implemented in practice to support regulatory and health technology assessment decisions. Although these methods are not a substitute for less restrictive pre-approval RCTs or rigorous observational studies when sufficient data are available in the post-approval setting, they can help to fill the evidence gap that exists in the early marketing period. Early evidence using real-world data and methods for extrapolating evidence should be reported with clear explanation of assumptions and limitations especially when used to support regulatory and health technology assessment decisions.

Development and Preliminary Validation of a Medicare Claims-Based Model to Predict Left Ventricular Ejection Fraction Class in Patients With Heart Failure.

BACKGROUND: Ejection fraction (EF) class is an important predictor of treatment response in heart failure (HF); however, administrative claims databases lack information on EF, limiting their usefulness in clinical and health services research of HF. METHODS AND RESULTS: We linked Medicare claims data to electronic medical records containing EF measurements for a cohort of 11 073 patients with HF from 2 academic medical centers. A a claims-based model predicting EF class was constructed using data from center 1 ("training sample") and validated using data from center 2 ("testing sample). Linear and logistic regression models with least absolute square shrinkage operator and Bayesian information criteria were developed to select the relevant predictor variables out of the total 57 candidate variables in the training sample. Higher accuracy was noted in the testing sample with models classifying patients into 2 EF classes (reduced EF <0.45) versus preserved EF (≥0.45) when compared with classifying patients into 3 EF classes (reduced, <0.40, moderately reduced, 0.40-0.49, or preserved, ≥0.50). In the testing sample, the most efficient model had 35 predictors and resulted in 83% of patients being correctly classified (95% CI, 82%-84%). The model had positive predictive value of 0.73 (95% CI, 0.68-0.78) and 0.84 (95% CI, 0.83-0.86) and sensitivity of 0.29 (95% CI, 0.25-0.32) and 0.97 (95% CI, 0.97-0.98) for reduced and preserved EF, respectively. In addition to HF-specific diagnosis codes, other factors including age, sex, medication use, and comorbidities, such as myocardial infarction and valve disorders, were important discriminators between EF classes. CONCLUSIONS: The claims-based model developed in this study may be used to identify patient subgroups with specific EF class in studies evaluating the health outcomes, utilization patterns, and cost, of HF patients in routine care when EF measurements are not available.

Real-world evidence of stroke prevention in patients with nonvalvular atrial fibrillation in the United States: the REVISIT-US study.

BACKGROUND: Little data exists regarding the effectiveness and safety of rivaroxaban or apixaban versus warfarin in nonvalvular atrial fibrillation (NVAF) patients treated outside of clinical trials. METHODS: This was a retrospective study using MarketScan claims from January 2012 to October 2014. We included adults, newly initiated on rivaroxaban, apixaban or warfarin, with a baseline CHA2DS2-VASc score ≥2, ≥2 diagnosis codes for NVAF and ≥180 days of continuous medical and prescription benefits. Patients with a prior stroke, systemic embolism or intracranial hemorrhage (ICH) were excluded. Eligible rivaroxaban or apixaban users were 1:1 propensity-score matched individually to warfarin users. Cox regression was performed to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for rivaroxaban and apixaban versus warfarin for the combined endpoint of ischemic stroke or ICH and each endpoint individually. RESULTS: Upon matching 11,411 rivaroxaban to 11,411 warfarin users, rivaroxaban was associated with a significant reduction of the combined endpoint of ischemic stroke or ICH versus warfarin (HR = 0.61, 95% CI = 0.45-0.82). ICH was significantly (HR = 0.53, 95% CI = 0.35-0.79) and ischemic stroke nonsignificantly reduced (HR = 0.71, 95% CI = 0.47-1.07) by rivaroxaban versus warfarin. After matching 4083 apixaban and 4083 warfarin users, apixaban was found to nonsignificantly reduce the combined endpoint of ischemic stroke or ICH versus warfarin (HR = 0.63, 95% CI = 0.35-1.12) and to reduce ICH risk (HR = 0.38, 95% CI = 0.17-0.88). Ischemic stroke risk was nonsignificantly increased with apixaban (HR = 1.13, 95% CI = 0.49-2.63) versus warfarin. LIMITATIONS: Sample size and number of combined events observed were relatively small. Residual confounding could not be ruled out. CONCLUSIONS: Rivaroxaban and apixaban were associated with less ICH than warfarin and both are likely associated with reductions in the combined endpoint. Further investigation to validate the numerically higher rate of ischemic stroke with apixaban versus warfarin is required.

Impact of Switching From a Vitamin K Antagonist to Rivaroxaban on Satisfaction With Anticoagulation Therapy: The XANTUS-ACTS Substudy.

BACKGROUND: The efficacy, safety, and ease of use of rivaroxaban may reduce anticoagulation-treatment burden and improve nonvalvular atrial fibrillation (NVAF) patient satisfaction compared with vitamin K antagonists (VKAs). HYPOTHESIS: Transitioning from a VKA to rivaroxaban improves treatment satisfaction in routine practice. METHODS: Xarelto for Prevention of Stroke in Patients With Atrial Fibrillation (XANTUS) is a prospective, noninterventional study in patients with NVAF prescribed rivaroxaban for prevention of stroke in routine practice. Patients receiving a VKA 4 weeks prior to the initial XANTUS study visit and switched to rivaroxaban were asked to complete the Anti-Clot Treatment Scale (ACTS). Changes from the initial visit to the first follow-up visit at ∼ 3 months (corresponding to a comparison of rivaroxaban vs prior VKA) for ACTS burden and benefit scores were calculated using and reported as least squared mean differences (LSMDs) with 95% confidence intervals (CIs). RESULTS: The study included 1291 NVAF patients with prior VKA treatment. The mean baseline ACTS burden and benefit scores were 50.51 ± 8.42 and 10.30 ± 2.70, respectively. After ∼ 3 months of rivaroxaban treatment, LSMDs were 4.38 points (95% CI: 2.53-6.22, P < 0.0001) for the burden and 1.01 points (95% CI: 0.27-1.75, P = 0.0075) for the benefit score. Fifty-four percent and 48% of patients reported experiencing at least a minimally important clinical difference in burden and benefit scores, respectively. CONCLUSIONS: Within this XANTUS cohort, switching from a VKA to rivaroxaban yielded statistically and clinically significant improvements in ACT burden and benefit scores.

Treatment Persistence and Discontinuation with Rivaroxaban, Dabigatran, and Warfarin for Stroke Prevention in Patients with Non-Valvular Atrial Fibrillation in the United States.

A retrospective cohort analysis of the US MarketScan claims databases was performed to compare persistence and discontinuation rates between the vitamin K antagonist warfarin and the non-vitamin K antagonist oral anticoagulants rivaroxaban and dabigatran in patients with non-valvular atrial fibrillation. The analysis included adult patients with non-valvular atrial fibrillation newly initiated on rivaroxaban, dabigatran, or warfarin between November 1, 2011 and December 31, 2013, with a baseline CHA2DS2-VASc score ≥2, two or more non-valvular atrial fibrillation diagnosis codes (427.31), and ≥6 months' continuous medical and pharmacy benefit enrollment before oral anticoagulant initiation. Propensity score matching was performed to match patients receiving rivaroxaban with those receiving dabigatran (1:1) and warfarin (1:1). Patients were followed until the first event of inpatient death, end of continuous enrollment, or end of study period. Medication persistence was defined as absence of a refill gap of >60 days. Discontinuation was defined as no additional refill for >90 days and through to end of follow-up. Hazard ratios (HRs) of oral anticoagulant persistence and discontinuation were estimated using Cox proportional hazard models. In total, 3,2634 patients were included (n = 10878/oral anticoagulant group). Rivaroxaban was associated with better persistence than both dabigatran (HR 0.64, 95% confidence interval [CI] 0.62-0.67) and warfarin (HR 0.62, 95% CI 0.59-0.64) and lower discontinuation than dabigatran (HR 0.61, 95% CI 0.58-0.64) and warfarin (HR 0.65, 95% CI 0.62-0.68). Real-world analysis of oral anticoagulant use may reveal whether the relatively high persistence/low discontinuation demonstrated for rivaroxaban translates into lower rates of stroke.

Clinical and economic impact of rivaroxaban on the burden of atrial fibrillation: The case study of Japan.

OBJECTIVES: Atrial fibrillation (AF) affects an estimated 1.5 million individuals in Japan, increasing their stroke risk and imposing considerable costs on the Japanese healthcare system. To reduce stroke incidence, guidelines recommend using anticoagulants in moderate-to-high risk non-valvular AF (NVAF) patients; however, many patients receive no treatment, aspirin only, or remain poorly-controlled on vitamin K antagonists (VKAs) due to high VKA discontinuation rates and non-adherence to guidelines. A prevalence-based Markov model was developed to estimate the clinical and budgetary impact of treating these patients with Xarelto(TM) (rivaroxaban, Bayer AG) in Japan. METHODS: Population, baseline risk of events, and associated management costs were estimated using data from Japanese publications where available. Treatment efficacy and safety were derived from published data and the J-ROCKET AF trial. Drug and physician visit costs were based on data from the Ministry of Health, Labor, and Welfare, the J-ROCKET AF trial, and Japanese clinical guidelines. RESULTS: This model demonstrates that increased use of rivaroxaban in inadequately-managed NVAF patients could avoid 456 081 non-fatal ischemic strokes (IS) and 76 975 cardiovascular deaths over 10 years in Japan. This clinical benefit offsets the increased incidence of myocardial infarctions and anticoagulant-related bleeding. Decreased event costs could lead to a ¥188.4 billion decrease in net spending over the analysis time horizon. CONCLUSIONS: Introducing rivaroxaban may decrease the burden of NVAF in Japanese society. From a clinical perspective, the reduction in IS and embolic events outweighs the increased risk of anticoagulant-related bleeding; from an economic perspective, reduced event costs offset drug and physician visit costs, resulting in cost savings.