Advancing Value-Based Models for Heart Failure: A Call to Action From the Value in Healthcare Initiative's Value-Based Models Learning Collaborative

Neurologic Alternative Payment Models-Bridging the Gap

This Viewpoint discusses alternative payments models and the responsibilities of health care professionals.

Feasibility of a Multi-Component Strengths-Building Intervention for Caregivers of Persons With Heart Failure

Caregivers of persons with heart failure (HF) navigate complex care plans, yet support strategies often focus solely on meeting the needs of patients. We conducted a randomized waitlist control trial (N = 38) to test the feasibility and gauge initial effect size of the Caregiver Support intervention on quality of life, caregiver burden, and self-efficacy among HF caregivers. The intervention includes up to five remote, nurse-facilitated sessions. Components address: holistic caregiver assessment, life purpose, action planning, resources, and future planning. Caregivers were 93.3% female, 60% White, and 63.3% spouses. Average age was 59.4. Participants who completed the intervention reported high satisfaction and acceptability of activities. Between-group effect sizes at 16 and 32 weeks suggest improvement in quality of life (mental health) (.88; 1.08), caregiver burden (.31; .37), and self-efficacy (.63; .74). Caregivers found Caregiver Support acceptable and feasible. Findings contribute evidence that this intervention can enhance caregiver outcomes. Clinicaltrials.gov Identifier NCT04090749.

Value-based Reimbursement as a Mechanism to Achieve Social and Financial Impact in the Healthcare System

Value-based reimbursement strategies have been considered in the continuous search for establishing a sustainable healthcare system. For models that have been already implemented, success is demonstrated according to specific details of the patients' consumption profile based on their clinical condition and the risk balance among all the stakeholders. From fee-for-service to value-based bundled payment strategies, the manner in which accurate patient-level cost and outcome information are used varies, resulting in different risk agreements between stakeholders. A thorough understanding of value-based reimbursement agreements that views such agreements as a mechanism for risk management is critical to the task of ensuring that the healthcare system generates social impacts while ensuring financial sustainability. This perspective article focuses on a critical analysis of the impact of value-based reimbursement strategies on the healthcare system from a social and financial perspective. A critical analysis of the literature about value-based reimbursement was used to identify how these strategies impact healthcare systems. The literature analysis was followed by the conceptual description of value-based reimbursement agreements as mechanisms for achieving social and financial impacts on the healthcare system. There is no single successful path toward payment reform. Payment reform is used as a strategy to re-engineer the way in which the system is organized to provide care to patients, and its successful implementation leads to cultural, social, and financial changes. Stakeholders have reached consensus regarding the claim that the use of value reimbursement strategies and business models could increase efficiency and generate social impact by reducing healthcare inequity and improving population health. However, the successful implementation of such new strategies involves financial and social risks that require better management by all the stakeholders. The use of cutting-edge technologies are essential advances to manage these risks and must be paired with strong leadership focusing on the directive to improve population health and, consequently, value. Payment reform is used as a mechanism to re-engineer how the system is organized to deliver care to patients, and its successful implementation is expected to result in social and financial modifications to the healthcare system.

Variation in Ischemic Stroke Payments in the USA: A Medicare Beneficiary Study

INTRODUCTION

The growing cost of stroke care has created the need for outcome-oriented and cost-saving payment models. Identifying imbalances in the current reimbursement model is an essential step toward designing impactful value-based reimbursement strategies. This study describes the variation in reimbursement fees for ischemic stroke management across the USA.

METHODS

This Medicare Fee-For-Service claims study examines USA beneficiaries who suffered an ischemic stroke from 2021Q1 to 2022Q2 identified using the Medicare-Severity Diagnosis-Related Groups (MS-DRGs). Demographic national and regional US data were extracted from the Census Bureau. The MS-DRG codes were grouped into four categories according to treatment modality and clinical complexity. Our primary outcome of interest was payments made across individual USA and US geographic regions, assessed by computing the mean incremental payment in cases of comparable complexity. Differences between states for each MS-DRG were statistically evaluated using a linear regression model of the logarithmic transformed payments.

RESULTS

227,273 ischemic stroke cases were included in our analysis. Significant variations were observed among all DRGs defined by medical complexity, treatment modality, and states (p < 0.001). Differences in mean payment per case with the same MS-DRG vary by as high as 500% among individual states. Although higher payment rates were observed in MS-DRG codes with major comorbidities or complexity (MCC), the variation was more expressive for codes without MCC. It was not possible to identify a standard mean incremental fee at a state level. At a regional level, the Northeast registered the highest fees, followed by the West, Midwest, and South, which correlate with poverty rates and median household income in the regions.

CONCLUSIONS

The payment variability observed across USA suggests that the current reimbursement system needs to be aligned with stroke treatment costs. Future studies may go one step further to evaluate accurate stroke management costs to guide policymakers in introducing health policies that promote better care for stroke patients.

Value-Based Payment for Clinicians Treating Cardiovascular Disease: A Policy Statement From the American Heart Association

Clinician payment is transitioning from fee-for-service to value-based payment, with reimbursement tied to health care quality and cost. However, the overarching goals of value-based payment-to improve health care quality, lower costs, or both-have been largely unmet. This policy statement reviews the current state of value-based payment and provides recommended best practices for future design and implementation. The policy statement is divided into sections that detail different aspects of value-based payment: (1) key program design features (patient population, quality measurement, cost measurement, and risk adjustment), (2) the role of equity during design and evaluation, (3) adjustment of payment, and (4) program implementation and evaluation. Each section introduces the topic, describes important considerations, and lists examples from existing programs. Each section includes recommended best practices for future program design. The policy statement highlights 4 key themes for successful value-based payment. First, programs should carefully weigh the incentives between lowering cost and improving quality of care and ensure that there is adequate focus on quality of care. Second, the expansion of value-based payment should be a tool for improving equity, which is central to quality of care and should be a focal point of program design and evaluation. Third, value-based payment should continue to move away from fee for service toward more flexible funding that allows clinicians to focus resources on the interventions that best help patients. Last, successful programs should find ways to channel clinicians' intrinsic motivation to improve their performance and the care for their patients. These principles should guide the future development of clinician value-based payment models.

Value elements and methods of value-based pricing for drugs in Japan: a systematic review

INTRODUCTION

Value-based pricing (VBP) can be a promising tool for optimizing drug prices. However, there is no consensus on the specific value elements and pricing method that should be used for VBP.

AREAS COVERED

We performed a systematic review and narrative synthesis to investigate the value elements and pricing method for VBP. The main inclusion criterion was that value elements, VBP method, and estimated prices for actual drugs were reported. We performed a search in MEDLINE and ICHUSHI Web. Eight articles met the selection criteria. Four studies adopted the cost-effectiveness analysis (CEA) approach and the others used different approaches. The CEA approach included the value elements of productivity, value of hope, real option value, disease severity, insurance value in addition to costs and quality-adjusted life years. The other approaches used efficacy, toxicity, novelty, rarity, research and development costs, prognosis, population health burden, unmet needs, and effectiveness. Each study used individual methods to quantify these broader value elements.

EXPERT OPINION

Both conventional and broader value elements are used for VBP. To allow VBP to be widely applied to various diseases, a simple, versatile method is preferable. Further research is needed to establish VBP method which enables to incorporate broader values.

Policy-Driven Post-Acute Care Program Lowers Mortality Rate and Medical Expenditures After Hospitalization for Acute Heart Failure: A Nationwide Propensity Score-Matched Study

OBJECTIVE

The beneficial effects of multidisciplinary disease management programs have been demonstrated. The present study investigated the effects of a policy-driven, health insurance-reimbursed, heart failure (HF) post-acute care (PAC) program on mortality, health care service utilization, and readmission expenses for patients following hospitalization for HF.

DESIGN

This was a retrospective propensity score-matched cohort study using the Taiwan National Health Insurance Research Database.

SETTING AND PARTICIPANTS

In total, 4346 patients (2173 receiving HF-PAC and 2173 controls) with left ventricular ejection fraction of ≤40% who were discharged following hospitalization for HF were included for analysis.

METHODS

All patients were followed up after discharge for all-cause mortality, emergency visits within 30 days, and length of stay and medical expenses for readmission within 180 days after discharge.

RESULTS

After propensity score matching, baseline characteristics of the HF-PAC and control groups were similar. During a mean follow-up period of 1.59 ± 0.92 years, according to the Cox multivariable analysis, HF-PAC reduced mortality by 48% compared with the control group, independent of traditional risk factors (hazard ratio = 0.520, 95% CI = 0.452-0.597, P < .001). Kaplan-Meier curves revealed that HF-PAC was associated with a higher cumulative survival rate (log-rank = 96.43, P < .001). HF-PAC also decreased the frequency of emergency visits after discharge by 23% in the 30 days post discharge and decreased length of stay and medical expenses related to readmission by 61% and 63%, respectively, in the 180 days post discharge (all P < .001).

CONCLUSIONS AND IMPLICATIONS

HF-PAC reduces short-term all-cause emergency visits, length of stay, and medical expenses for all-cause readmission and all-cause mortality in patients discharged following hospitalization for HF. Our findings suggest that PAC should include care continuity, optimal adaptation of transitional care components, and HF cardiologist engagement with multidisciplinary coordination.

Wearable Biosensors in Congenital Heart Disease: Needs to Advance the Field

Traditional measures of clinical status and physiology have generally been based in health care settings, episodic, short in duration, and performed at rest. Wearable biosensors provide an opportunity to obtain continuous non-invasive physiologic data from patients with congenital heart disease (CHD) in the real-world setting, over longer durations, and across varying levels of activity. However, there are significant technical limitations to the use of wearable biosensors in CHD. Here, we review current applications of wearable biosensors in CHD; how clinical and research uses of wearable biosensors must consider various CHD physiologies; the technical challenges in developing wearable biosensors for CHD; and special considerations for digital biomarkers in CHD.

Trends in Medicare Reimbursement for Adult Cardiothoracic Surgery Procedures: 2007 to 2020

Background. Cardiovascular disease has been the leading cause of death in the US for decades. Over half a million cardiothoracic surgery procedures are performed per year, with an increasingly aging population and rising healthcare costs. The purpose of this study was to evaluate trends in Medicare reimbursement rates from 2007 to 2020 for various cardiothoracic surgery procedures. Methods. The Centers for Medicare & Medicaid Services Physician Fee Schedule Look-Up Tool was queried for common procedural terminology codes for 119 common cardiothoracic surgery procedures to obtain reimbursement data by year. Procedures were organized into cardiac, CABG, and thoracic subgroups. All monetary data were adjusted for inflation to 2020 US dollars. Adjusted data were analyzed to calculate compound annual growth rates (CAGR), average annual change, and total percent change for each procedure. Results. After adjusting for inflation, the reimbursement rates for cardiothoracic surgery procedures decreased by 10.20% on average. Reimbursement rates for cardiac, CABG, and thoracic surgical procedures decreased by 8.74%, 14.46%, and 10.94%, respectively. The mean annual change overall was −$14.47, and the CAGR was 0.82%. CABG procedures had the greatest decrease in CAGR (−1.11%), annual change (−$30.30), and total percent change (−14.46%). Conclusions. Medicare reimbursements for cardiothoracic surgery procedures steadily decreased from 2007 to 2020, with CABG procedures experiencing the highest percentage of decline. Dissemination of these findings is crucial to raising awareness for healthcare administrators, surgeons, insurance companies, and policymakers to ensure the accessibility of these procedures for high-quality cardiothoracic surgery care in the United States.

Public policy for healthy living: How COVID-19 has changed the landscape

The coronavirus disease 2019 (COVID-19) pandemic had a transformational impact on public policy as governments played a leading role, working alongside and coordinating with business/industry, healthcare, public health, education, transportation, researchers, non-governmental organizations, philanthropy, and media/communications. This paper summarizes the impact of the pandemic on different areas of public policy affecting healthy living and cardiovascular health including prevention (i.e., nutrition, physical activity, air quality, tobacco use), risk factors for chronic disease (hypertension, diabetes, obesity, substance abuse), access to health care, care delivery and payment reform, telehealth and digital health, research, and employment policy. The paper underscores where public policy is evolving and where there are needs for future evidence base to inform policy development, and the intersections between the public and private sectors across the policy continuum. There is a continued need for global multi-sector coordination to optimize population health.

Incidence of Timely Outpatient Follow-Up Care After Emergency Department Encounters for Acute Heart Failure

BACKGROUND

Patients who are discharged from the emergency department (ED) after an encounter for acute heart failure are at high risk for return hospitalization. These patients may benefit from timely outpatient follow-up care to reassess volume status, adjust medications, and reinforce self-care strategies. This study examines the incidence of outpatient follow-up care after ED encounters for acute heart failure and describes patient characteristics associated with obtaining timely follow-up care.

METHODS

We conducted a retrospective cohort study using an administrative claims database for a large US commercial insurer, from January 1, 2012 to June 30, 2019. Participants included adult patients discharged from the ED with principal diagnosis of acute heart failure. The primary outcome was obtaining an in-person outpatient clinic visit for heart failure within 30 days. We also examined the competing risk of all-cause hospitalization within 30 days and without an intervening outpatient clinic visit. We estimated competing risk regression models to identify patient characteristics associated with obtaining outpatient follow-up and report cause-specific hazard ratios.

RESULTS

The cohort included 52 732 patients, with mean age of 73.9 years (95% CI, 73.8-74.0) and 27 395 (52.0% [95% CI, 51.5-52.4]) female patients. Within 30 days of the ED encounter, 12 279 (23.2%) patients attended an outpatient clinic visit for heart failure, with 8382 (15.9%) patients hospitalized before they could obtain an outpatient clinic visit. In the adjusted analysis, patients that were younger, women, reporting non-Hispanic Black race, and had fewer previous clinic visits were less likely to obtain outpatient follow-up care.

CONCLUSIONS

Few patients obtain timely outpatient follow-up after ED visits for heart failure, although nearly 20% require hospitalization within 30 days. Improved transitions following discharge from the ED may represent an opportunity to improve outcomes for patients with acute heart failure.

Blood Pressure and Glycemic Control Among Ambulatory US Adults With Heart Failure: National Health and Nutrition Examination Survey 2001 to 2018

BACKGROUND

Multisociety guidelines recommend a goal systolic blood pressure (BP) <130 mm Hg and a hemoglobin A1c (HbA1c) <8% in patients with heart failure (HF), regardless of ejection fraction. Few studies have described BP and glycemic control in ambulatory patients with HF and racial and ethnic disparities in this subset of the population.

METHODS

We evaluated prevalence of uncontrolled BP and HbA1c in non-Hispanic Black, non-Hispanic White, and Mexican American adults aged ≥20 years with self-reported HF (National Health and Nutrition Examination Surveys: 2001-2018). Prevalence ratios (95% CI) for uncontrolled BP and HbA1c were calculated by race and ethnicity and adjusted for sex, age, treatment, and socioeconomic status. In secondary analyses, we examined trends in the prevalence of uncontrolled BP and HbA1c.

RESULTS

Uncontrolled BP was present in 48% (95% CI, 49%-56%) of adults with HF (representing 2.3 million people). Non-Hispanic Black participants had a higher prevalence of uncontrolled BP compared with non-Hispanic White participants (53% [48%-58%] compared with 47% [43%-51%], P<0.05). In adjusted models, non-Hispanic Black participants were 1.19 (1.02-1.39) times more likely to have uncontrolled BP than non-Hispanic White participants. Overall, uncontrolled HbA1c was found in 8% (6%, 10%) with no differences by race and ethnicity. Prevalence of uncontrolled BP improved over time but uncontrolled risk factors remained high-2017 to 2018: 41% (36%, 47%) and 7% (5%, 12%) had uncontrolled BP and HbA1c, respectively.

CONCLUSIONS

We document an unacceptably high prevalence of uncontrolled BP and HbA1c in a nationally representative, ambulatory HF sample with significant differences in BP control by race and ethnicity.

Rationale and design of a cluster-randomized pragmatic trial aimed at improving use of guideline directed medical therapy in outpatients with heart failure: PRagmatic trial of messaging to providers about treatment of heart failure (PROMPT-HF)

Heart failure with reduced ejection fraction (HFrEF) is one of the most common chronic illnesses in the United States and carries significant risk of morbidity and mortality. Use of guideline-directed medical therapy (GDMT) for patients with HFrEF has been shown to dramatically improve outcomes, but adoption of these treatments remains generally low. Possible explanations for poor GDMT uptake include lack of knowledge about recommended management strategies and provider reluctance due to uncertainties regarding application of said guidelines to real-world practice. One way to overcome these barriers is by harnessing the electronic health record (EHR) to create patient-centered "best practice alerts" (BPAs) that can guide clinicians to prescribe appropriate medical therapies. If found to be effective, these low-cost interventions can be rapidly applied across large integrated healthcare systems. The PRagmatic Trial Of Messaging to Providers about Treatment of Heart Failure (PROMPT-HF) trial is a pragmatic, cluster randomized controlled trial designed to test the hypothesis that tailored and timely alerting of recommended GDMT in heart failure (HF) will result in greater adherence to guidelines when compared with usual care. PROMPT-HF has completed enrollment of 1,310 ambulatory patients with HFrEF cared for by 100 providers who were randomized to receive a BPA vs usual care. The BPA alerted providers to GDMT recommended for their patients and displayed current left ventricular ejection fraction (LVEF) along with the most recent blood pressure, heart rate, serum potassium and creatinine levels, and estimated glomerular filtration rate. It also linked to an order set customized to the patient that suggests medications within each GDMT class not already prescribed. Our goal is to examine whether tailored EHR-based alerting for outpatients with HFrEF will lead to higher rates of GDMT at 30 days post randomization when compared with usual care. Additionally, we are assessing clinical outcomes such as hospital readmissions and death between the alert versus usual care group. Trial Registration: Clinicaltrials.gov NCT04514458.

Hospitalization Rates in Patients with Heart Failure and Reduced Ejection Fraction Initiating Sacubitril/Valsartan or Angiotensin-Converting Enzyme Inhibitors/Angiotensin Receptor Blockers: A Retrospective Cohort Study

Introduction

The angiotensin receptor neprilysin inhibitor (ARNI) sacubitril/valsartan (SAC/VAL) has shown benefit in patients with symptomatic heart failure (HF), including those naïve to renin–angiotensin–aldosterone system inhibitor (RAASi) therapy, and is considered the preferred RAASi for chronic HF. Real-world data on ARNI, specifically in RAASi-naïve patients, are limited. This study compared real-world outcomes of ARNI (SAC/VAL) vs. angiotensin-converting enzyme inhibitor (ACEi) or angiotensin receptor blocker (ARB) therapy in RAASi-naïve patients with HF and reduced ejection fraction (HFrEF).

Methods

This retrospective cohort study included de-identified data on RAASi-naïve patients with HFrEF (left ventricular ejection fraction ≤ 40%) who had newly initiated SAC/VAL or ACEi/ARB between July 1, 2015, and March 31, 2019, from the Optum® Electronic Health Records database in the US. New SAC/VAL users were propensity score matched 1:2 with new ACEi/ARB users by pre-selected characteristics. One-year post-index rates of all-cause, HF, and cardiovascular hospitalizations and the composite of HF hospitalization or emergency room (ER) visits were measured using negative binomial regression. Time to first all-cause hospitalization, HF hospitalization, and composite of HF hospitalization or ER visits was measured using a subdistribution hazards model.

Results

The matched sample included 3059 new SAC/VAL and 6118 new ACEi/ARB users. Rates of all-cause hospitalization and composite of HF hospitalization or ER visits were significantly lower with SAC/VAL compared with ACEi/ARB (incidence rate ratio [95% confidence interval]: 0.87 [0.81–0.93] and 0.87 [0.81–0.94], respectively), whereas rates of HF hospitalizations and cardiovascular hospitalizations were similar (1.00 [0.91–1.11] and 0.94 [0.87–1.02], respectively). Time-to-event analyses also showed a similar trend.

Conclusions

In real-world clinical practice, RAASi-naïve patients with HFrEF initiating SAC/VAL were less likely to be hospitalized than those initiating ACEi/ARB, suggesting a potential for a reduced clinical and economic burden in these patients.

Effects of Septal Reduction Therapy on Acute Cardiovascular Events and All-Cause Mortality in Patients with Hypertrophic Cardiomyopathy

Septal reduction therapy (SRT) -i.e. septal myectomy and alcohol septal ablation-has been performed to treat medically refractory hypertrophic cardiomyopathy (HCM) for decades. However, it is largely unknown whether SRT prevents HCM-related cardiovascular events or death. The objective was to examine the effects of SRT on acute cardiovascular events and all-cause mortality in HCM. We performed a propensity score (PS) -matched study using databases that capture all hospitalizations and outpatient visits in New York state. We identified patients with HCM who underwent SRT between 2007 and 2014 (i.e. the SRT group) and those who had never had SRT but had at least one hospitalization for HCM during the same period (i.e. the control group). We performed PS matching at a 1:1 ratio. The primary outcome was a composite of acute cardiovascular events and all-cause mortality during 0-180 days and 181-360 days. The secondary outcome was 180- and 360-day all-cause mortality. We included 846 patients with HCM (423 PS-matched pairs). Patients who underwent SRT had a lower risk of the primary outcome event (0-180 days: odds ratio [OR], 0.54; 95% confidence intervals (CI), 0.37-0.80; P = 0.002 and 181-360 days: OR, 0.33; 95% CI, 0.22-0.51; P < 0.0001). Furthermore, the risk of all-cause mortality was lower at 180 days (OR, 0.37; 95% CI, 0.22-0.63; P = 0.0003) and 360 days post-SRT (OR, 0.32; 95% CI, 0.20-0.51; P < 0.0001). In conclusion, our PS-matched study using population-based datasets demonstrated that SRT was associated with a reduced risk of a composite of acute cardiovascular events and all-cause mortality in HCM during the first post-SRT year.

Policy and Payment Challenges in the Postpandemic Treatment of Heart Failure: Value-Based Care and Telehealth

Increasing patient and therapeutic complexity have created both challenges and opportunities for heart failure care. Within this background, the coronavirus disease-2019 pandemic has disrupted care as usual, accelerating the need for transition from volume-based to value-based care, and demanding a rapid expansion of telehealth and remote care for heart failure. Patients, clinicians, health systems, and payors have by necessity become more invested in these issues. Herein we review recent changes in health care policy related to the movement from volume to value-based payment and from in-person to remote care delivery.

Effect of a Hospital and Postdischarge Quality Improvement Intervention on Clinical Outcomes and Quality of Care for Patients With Heart Failure With Reduced Ejection Fraction: The CONNECT-HF Randomized Clinical Trial

IMPORTANCE

Adoption of guideline-directed medical therapy for patients with heart failure is variable. Interventions to improve guideline-directed medical therapy have failed to consistently achieve target metrics, and limited data exist to inform efforts to improve heart failure quality of care.

OBJECTIVE

To evaluate the effect of a hospital and postdischarge quality improvement intervention compared with usual care on heart failure outcomes and care.

DESIGN, SETTING, AND PARTICIPANTS

This cluster randomized clinical trial was conducted at 161 US hospitals and included 5647 patients (2675 intervention vs 2972 usual care) followed up after a hospital discharge for acute heart failure with reduced ejection fraction (HFrEF). The trial was performed from 2017 to 2020, and the date of final follow-up was August 31, 2020.

INTERVENTIONS

Hospitals (n = 82) randomized to a hospital and postdischarge quality improvement intervention received regular education of clinicians by a trained group of heart failure and quality improvement experts and audit and feedback on heart failure process measures (eg, use of guideline-directed medical therapy for HFrEF) and outcomes. Hospitals (n = 79) randomized to usual care received access to a generalized heart failure education website.

MAIN OUTCOMES AND MEASURES

The coprimary outcomes were a composite of first heart failure rehospitalization or all-cause mortality and change in an opportunity-based composite score for heart failure quality (percentage of recommendations followed).

RESULTS

Among 5647 patients (mean age, 63 years; 33% women; 38% Black; 87% chronic heart failure; 49% recent heart failure hospitalization), vital status was known for 5636 (99.8%). Heart failure rehospitalization or all-cause mortality occurred in 38.6% in the intervention group vs 39.2% in usual care (adjusted hazard ratio, 0.92 [95% CI, 0.81 to 1.05). The baseline quality-of-care score was 42.1% vs 45.5%, respectively, and the change from baseline to follow-up was 2.3% vs -1.0% (difference, 3.3% [95% CI, -0.8% to 7.3%]), with no significant difference between the 2 groups in the odds of achieving a higher composite quality score at last follow-up (adjusted odds ratio, 1.06 [95% CI, 0.93 to 1.21]).

CONCLUSIONS AND RELEVANCE

Among patients with HFrEF in hospitals randomized to a hospital and postdischarge quality improvement intervention vs usual care, there was no significant difference in time to first heart failure rehospitalization or death, or in change in a composite heart failure quality-of-care score.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT03035474.

Machine ​learning algorithms for claims data-based prediction of in-hospital mortality in patients with heart failure

Aims

Models predicting mortality in heart failure (HF) patients are often limited with regard to performance and applicability. The aim of this study was to develop a reliable algorithm to compute expected in‐hospital mortality rates in HF cohorts on a population level based on administrative data comparing regression analysis with different machine learning (ML) models.

Methods and results

Inpatient cases with primary International Statistical Classification of Diseases and Related Health Problems (ICD‐10) encoded discharge diagnosis of HF non‐electively admitted to 86 German Helios hospitals between 1 January 2016 and 31 December 2018 were identified. The dataset was randomly split 75%/25% for model development and testing. Highly unbalanced variables were removed. Four ML algorithms were applied, and all algorithms were tuned using a grid search with multiple repetitions. Model performance was evaluated by computing receiver operating characteristic areas under the curve. In total, 59 125 cases (69.8% aged 75 years or older, 51.9% female) were investigated, and in‐hospital mortality was 6.20%. Areas under the curve of all ML algorithms outperformed regression analysis in the testing dataset with values of 0.829 [95% confidence interval (CI) 0.814–0.843] for logistic regression, 0.875 (95% CI 0.863–0.886) for random forest, 0.882 (95% CI 0.871–0.893) for gradient boosting machine, 0.866 (95% CI 0.854–0.878) for single‐layer neural networks, and 0.882 (95% CI 0.872–0.893) for extreme gradient boosting. Brier scores demonstrated a good calibration especially of the latter three models.

Conclusions

We introduced reliable models to calculate expected in‐hospital mortality based only on administrative routine data using ML algorithms. A broad application could supplement quality measurement programs and therefore improve future HF patient care.

Value in Healthcare Initiative: Summary and Key Recommendations

In spring 2018, the American Heart Association convened the Value in Healthcare Summit to begin an important conversation about the challenges patients with cardiovascular disease face in accessing and deriving quality and value from the healthcare system. Following the summit and recognizing the collective momentum it created, the American Heart Association, in collaboration with the Robert J. Margolis Center for Health Policy at Duke University, launched the Value in Healthcare Initiative-Transforming Cardiovascular Care. Four areas of focus were identified, and learning collaboratives were established and proceeded to conduct concrete, actionable problem solving in 4 high-impact areas in cardiovascular care: Value-Based Models, Partnering with Regulators, Predict and Prevent, and Prior Authorization. The deliverables from these groups are being disseminated in 4 stand-alone articles, and their publication will initiate further work to test and evaluate each of these promising areas of reform. This article provides an overview of the initiative's findings and highlights key cross-cutting themes for consideration as the initiative moves forward.

Frontiers of Upstream Stroke Prevention and Reduced Stroke Inequity Through Predicting, Preventing, and Managing Hypertension and Atrial Fibrillation: A Call to Action From the Value in Healthcare Initiative's Predict & Prevent Learning Collaborative

Stroke is one of the leading causes of morbidity and mortality in the United States. While age-adjusted stroke mortality was falling, it has leveled off in recent years due in part to advances in medical technology, health care options, and population health interventions. In addition to adverse trends in stroke-related morbidity and mortality across the broader population, there are sociodemographic inequities in stroke risk. These challenges can be addressed by focusing on predicting and preventing modifiable upstream risk factors associated with stroke, but there is a need to develop a practical framework that health care organizations can use to accomplish this task across diverse settings. Accordingly, this article describes the efforts and vision of the multi-stakeholder Predict & Prevent Learning Collaborative of the Value in Healthcare Initiative, a collaboration of the American Heart Association and the Robert J. Margolis, MD, Center for Health Policy at Duke University. This article presents a framework of a potential upstream stroke prevention program with evidence-based implementation strategies for predicting, preventing, and managing stroke risk factors. It is meant to complement existing primary stroke prevention guidelines by identifying frontier strategies that can address gaps in knowledge or implementation. After considering a variety of upstream medical or behavioral risk factors, the group identified 2 risk factors with substantial direct links to stroke for focusing the framework: hypertension and atrial fibrillation. This article also highlights barriers to implementing program components into clinical practice and presents implementation strategies to overcome those barriers. A particular focus was identifying those strategies that could be implemented across many settings, especially lower-resource practices and community-based enterprises representing broad social, economic, and geographic diversity. The practical framework is designed to provide clinicians and health systems with effective upstream stroke prevention strategies that encourage scalability while allowing customization for their local context.