Optimizing utilization of sacubitril/valsartan for treatment of heart failure could improve provider performance in the Bundled Payments for Care Improvement initiative and the Medicare Shared Savings Program.
Objectives: The burden of heart failure (HF) in Medicare beneficiaries and the shift toward value-based care is well documented. This study evaluates the association between sacubitril/valsartan (SAC/VAL) use and medical expenditures for beneficiaries with HF with reduced ejection fraction (HFrEF) within the framework of the Bundled Payments for Care Improvement (BPCI) initiative or the Medicare Shared Savings Program (MSSP).
Study Design: Retrospective cohort study.
Methods: We utilized 100% Medicare enrollment and Parts A/B/D claims to identify beneficiaries with HFrEF during (1) replicated BPCI episodes between 2016 and 2018 or (2) calendar year 2018 and managed by a MSSP participant. Use of SAC/VAL or angiotensin-converting enzyme inhibitors (ACEIs)/angiotensin II receptor blockers (ARBs) and Medicare Parts A/B expenditures were examined in propensity score (PS)–matched treatment groups in both programs.
Results: For the BPCI initiative, 13,785 episodes with SAC/VAL were PS matched to 41,355 with ACEIs/ARBs; 12,003 episodes with SAC/VAL were matched to 36,009 with neither treatment. Parts A/B expenditures for patients treated with SAC/VAL were not statistically different compared with those treated with ACEIs/ARBs (mean difference, –$447; 95% CI, –$1227 to $333) and $10,249 less vs patients receiving neither treatment (95% CI, –$11,335 to –$9163). For the MSSP, 18,004 patients utilizing SAC/VAL were PS matched to 54,012 patients receiving ACEIs/ARBs; 7186 patients utilizing SAC/VAL were matched to 21,558 receiving neither treatment. The SAC/VAL cohort had 2018 annual Parts A/B expenditures that were $875 less vs the ACEI/ARB cohort (95% CI, –$1650 to –$101) and $21,467 less vs neither treatment (95% CI, –$11,376 to –$7361).
Conclusions: Optimizing utilization of SAC/VAL for beneficiaries with HFrEF could reduce medical expenditures and improve provider performance in the context of alternative payment models.
Am J Accountable Care. 2023;11(1):5-17. https://doi.org/10.37765/ajac.2023.89339
There are more than 6 million patients with heart failure (HF) in the United States,1-4 resulting in approximately 1 million hospitalizations each year. Half of these hospitalizations are among patients with HF with reduced ejection fraction (HFrEF),1,5-7 and research shows that the hospitalizations are expensive to manage.8-13 In 2021, the American College of Cardiology (ACC) released its expert consensus decision pathway recommending use of an angiotensin receptor-neprilysin inhibitor (ARNI) as the preferred renin angiotensin antagonist for HFrEF.14 Sacubitril/valsartan (SAC/VAL), an ARNI and guideline-directed drug therapy, has been found to reduce HF-related hospitalizations and rehospitalizations compared with the standard of care, angiotensin-converting enzyme inhibitors (ACEIs), in 2 landmark clinical trials: the 2014 PARADIGM-HF trial (NCT01035255) and the 2018 PIONEER-HF trial (NCT02554890).15-25
Within the Medicare program, in which 7.1% of beneficiaries have HF,8 health care providers continue to identify ways to balance quality and cost of care.26 Under a traditional fee-for-service (FFS) system, providers and facilities are reimbursed based on the individual services rendered, placing an emphasis on the quantity of care. This emphasis does not include incentives for providers to focus on quality, efficiency, and outcomes. In contrast, incentives for value-based care participants in alternative payment models (APMs) encourage care coordination and high-quality, lower-cost services.27-31
CMS’ Center for Medicare & Medicaid Innovation is at the forefront of APM development and testing in the health care system. As of 2020, more than 40 APMs were being tested, and additional models will be rolled out in the coming years.32 In bundled payment models, participants are reimbursed a single payment that covers all services provided during an episode of care.33 Payments to a specific clinician are adjusted based on the total payment, because the clinician is not responsible for providing all care during that episode. In shared savings models, medical expenditures are compared with a risk-adjusted benchmark, and participating providers may share a percentage of savings or losses depending on their quality performance.34,35 APMs in the cardiovascular space have seen mixed success, and there is limited information on the association between use of guideline-directed drug therapy and APM outcomes.36
This retrospective cohort study replicated 90-day postdischarge episodes of care and a 365-day performance period of 2 Medicare APMs—the Bundled Payments for Care Improvement (BPCI) initiative and the Medicare Shared Savings Program (MSSP)—for Medicare beneficiaries with HFrEF to assess the association between utilization of SAC/VAL and medical expenditures.
This study utilized the 100% sample of Medicare FFS enrollment and Parts A/B/D claims from 2 time periods: (1) January 1, 2015, through December 31, 2018, for the BPCI analysis and (2) January 1, 2017, through December 31, 2018, for the MSSP analysis. Public use files were used to identify BPCI and MSSP participants.37,38
A retrospective cohort study design was employed to examine BPCI Model 2 episodes from 2016 to 2018 (the BPCI initiative has 4 models of care, with Model 2 focused on the acute and postacute care episode) and MSSP participant annual data from calendar year (CY) 2018. Effectively, we constructed these episodes and performance years within the data based on triggering events (inpatient hospitalization for BPCI and start of CY for MSSP).
Unlike traditional longitudinal claims analyses seeking to understand the total cost of care from a payer perspective, this analysis was structured to inform conceptually how SAC/VAL utilization may affect provider performance, measured as total Part A and Part B expenditures, within the framework of the BPCI Model 2 and MSSP. BPCI Model 2 was examined because it includes both acute and postacute care as opposed to other models that focus on one or the other.
The BPCI Model 2 analysis replicated HF postdischarge 90-day episodes of care for patients with HFrEF. An episode indexed on the initial acute inpatient stay with a Medicare Severity Diagnosis Related Group for congestive HF (291, 292, 293) and included all Parts A/B covered medical services from the hospital stay and during the 90 days post hospital discharge. A 12-month look-back period, based on episode initiation, was utilized to capture beneficiaries’ baseline demographic and clinical characteristics. Patients were required to have at least 2 medical claims between January 1, 2016, and December 31, 2018, with an International Classification of Diseases, Tenth Revision (ICD-10) code for systolic HF (I5020, I5021, I5022, I5023, I5040, I5041, I5042, I5043; code set available in eAppendix Table 1 [eAppendix available at ajmc.com]) during the 12-month baseline period and/or 90-day episode window to confirm a systolic classification and be continuously enrolled in Medicare Parts A/B/D as the primary payer for the entire episode. Systolic HF diagnoses were allowed during the 90-day episode because the hospitalization triggers the start of the episode and not the diagnosis. Episodes were excluded if they overlapped with previous episodes or if the patients had end-stage renal disease (ESRD), had initial hospital stays longer than 365 days, or died before the end of the episode.
The MSSP analysis examined patients with HFrEF managed by accountable care organizations participating in the MSSP in 2018 per CMS assignment. Beneficiaries were followed longitudinally from the beginning of CY 2018 (ie, January 1, 2018, the index date) through the end of the year. A 12-month look-back period, from January 1, 2017, to December 31, 2017, was utilized to capture the beneficiaries’ demographic and clinical characteristics. The MSSP analysis included beneficiaries with HFrEF, indicated by at least 2 medical claims with an ICD-10 code for systolic HF during the 12-month baseline period and/or 12-month episode window, who did not die between January 1, 2017, and December 31, 2018; were continuously enrolled in Medicare Parts A/B/D; were assigned to a MSSP participant throughout CY 2018 (performance year) and CY 2017 (baseline period); and had no other primary payer or ESRD. MSSP also excludes patients who died during the performance year as part of its shared savings calculation.
BPCI Model 2 and MSSP episodes were placed into treatment cohorts based on patient utilization of specific HF medications during the 90-day episode (BPCI Model 2) or calendar year (MSSP) defined by the presence of 1 or more prescription claim with a National Drug Code for the drug of interest. Cohorts were mutually exclusive and established based on whether the beneficiary utilized (1) SAC/VAL, (2) an ACEI/ARB and not SAC/VAL (referred to as the ACEI/ARB treatment group), and (3) neither an ACEI/ARB nor SAC/VAL in the respective periods (ie, 90 days post discharge for BPCI and CY 2018 for MSSP).
Within the BPCI Model 2 sample, the SAC/VAL and ACEI/ARB treatment groups were restricted for generalized linear modeling to individuals who had no more than a 30-day gap in days covered, allowing for a small gap in claims, because patients may be discharged from the hospital with a filled script of medication. In the MSSP modeling analysis, the SAC/VAL and ACEI/ARB treatment groups were restricted for the generalized linear modeling to individuals who were adherent to treatment during the performance year, defined as proportion of days covered of at least 0.8.
The primary outcomes were total Medicare expenditures for Parts A/B medical services during the episode and performance periods. For the BPCI Model 2 replication, this included all medical services from the date of the triggering HF-related hospital admission up to 90 days post discharge, and for the MSSP replication, this included all medical services for CY 2018, reflecting the measure for provider performance in these 2 APMs, respectively. Because Part D drug costs are not included in the BPCI and MSSP and this analysis specifically aimed to mirror the design of these programs, they were not included as part of the expenditure outcomes nor was Part D utilization measured. All outcomes were reported by treatment cohort. Utilization and expenditures for each covered Medicare Parts A/B service were measured as secondary outcomes, including hospitalizations/rehospitalizations, emergency department visits, physician office visits, skilled nursing facility (SNF) admissions, and home health services. Each of these outcomes was measured as both all-cause and HF-related (indicated by an International Classification of Diseases, Ninth Revision or ICD-10 code for HF in the primary position on the claim) expenditures and rates. All expenditures were reported in US$ for the year they were recorded.
Baseline demographic characteristics, clinical characteristics, and medication utilization were measured at baseline. The Medicare Hierarchical Condition Category (HCC) composite risk score was calculated using the CMS algorithm39 as an estimator of future health care expenditures. Medical and pharmacy resource utilization were also assessed at baseline.
All analyses were conducted using SAS version 9.4 (SAS Institute). Unadjusted rates of health care resource utilization (HCRU) were calculated as the number of events per 10,000 episodes (BPCI Model 2) or beneficiaries (MSSP) and stratified by treatment cohort.
The SAC/VAL cohort was separately propensity score (PS) matched40 to the ACEI/ARB cohort and the cohort receiving neither treatment using a 1:3 matching ratio to increase study precision and appropriately capture the variation in the large control groups while balancing matching rate. Pairwise comparisons were made to simplify comparisons and focus on the effect of ARNI treatment compared with each of the other treatment cohorts. The treatment cohorts were matched on age, sex, race, US Census region, dual eligibility for Medicare and Medicaid, eligibility for Part D low-income subsidy, baseline comorbidities, baseline medication utilization, HCC risk score,39 baseline HCRU, and baseline health care expenditures. Prescribing provider type for HF medications was utilized to match the SAC/VAL group to the ACEI/ARB group, but the treatment group that utilized neither did not have a prescribing provider by definition and thus was not included when matching the SAC/VAL cohort to that cohort. Among the BPCI sample, treatment cohorts were also matched on episode index year and a count variable of the episode number for the contributing patient.
Propensity scores were constructed through a logistic regression with baseline variables included as independent variables. The dependent variable of the propensity score model, which was specified as a multiple variable logistic regression, was the likelihood of receiving SAC/VAL. Patients in the SAC/VAL treatment group were matched to patients from the remaining comparator sample based on a layering nearest neighbor approach or similar greedy matching algorithm. This method maximized the match and retained the sample while also ensuring balance among the treatment groups.
Descriptive statistics for all variables across both programs by treatment cohort were assessed before and after matching, and the standardized mean difference was used to assess whether the PS-matched treatment cohorts were balanced in baseline characteristics.41 Generalized linear models with a gamma distribution were constructed to estimate the adjusted differences in the expenditure outcomes between the PS-matched treatment cohorts. Statistical significance was determined at an α level of 0.05.
BPCI Model 2
A total of 412,781 HFrEF episodes under the replicated BPCI program between 2016 and 2018 were identified and further stratified by treatment exposure (Figure). Patients received SAC/VAL in 17,444 (4%), ACEIs/ARBs in 190,184 (46%), and neither in 205,153 (50%) of the BPCI Model 2 episodes.
A total of 13,785 episodes with patients who utilized SAC/VAL were PS matched to 41,355 episodes with patients who utilized ACEIs/ARBs. A total of 12,003 episodes with patients who utilized SAC/VAL were matched to 36,009 episodes with patients who utilized neither treatment. The baseline demographic and clinical characteristics, expenditures, and HCRU were well balanced between the PS-matched SAC/VAL and ACEI/ARB treatment cohorts and the SAC/VAL cohort and the treatment cohort who received neither treatment (Table 1 and eAppendix Table 2). Descriptive statistics outlining the sample’s baseline characteristics prior to matching are available in the eAppendix (eAppendix Table 3).
Prior to matching, the unadjusted mean total Parts A/B 90-day episode expenditures for the SAC/VAL group ($19,858) were similar to those for the ACEI/ARB group ($19,650), but they were lower compared with those treated with neither SAC/VAL nor ACEIs/ARBs ($25,757).
After controlling for baseline characteristics and duration of therapy, BPCI HF episode Parts A/B expenditures for patients with HFrEF treated with SAC/VAL were not significantly different from those treated with ACEIs/ARBs (mean difference, –$447; 95% CI, –$1227 to $333; P = .26) and $10,249 less than patients receiving neither treatment (95% CI, –$11,335 to –$9163; P < .01). Mean episode rehospitalization and SNF expenditures for patients receiving SAC/VAL were $741 (95% CI, –$1371 to –$111; P < .02) and $298 (95% CI, –$423 to –$172; P < .01) less, respectively, than those for patients treated with ACEIs/ARBs and $4783 (95% CI, –$5629 to –$3937; P < .01) and $3731 (95% CI, –$4017to –$3446; P < .01) less than those receiving neither therapy. There was not a significant difference in the HF-related rehospitalization expenditures between the SAC/VAL and ACEI/ARB groups (95% CI, –$340 to $560; P = .63), but HF-related rehospitalization expenditures in the SAC/VAL group were $1396 less than in the cohort who received neither treatment (95% CI, –$2023 to –$770; P < .01) (Table 2). Results describing HCRU rates in the PS-matched cohorts are available in the eAppendix (eAppendix Table 4).
After applying the inclusion and exclusion criteria, the analytic sample included 312,611 beneficiaries with HFrEF in the MSSP CY 2018 (Figure). Among the beneficiaries included in the MSSP performance year, 20,619 (6%) patients received SAC/VAL, 188,818 (60%) received ACEIs/ARBs, and 103,174 (34%) received neither treatment.
A total of 18,004 patients who utilized SAC/VAL were PS matched to 54,012 patients who utilized ACEIs/ARBs. A total of 7186 patients who utilized SAC/VAL were matched to 21,558 patients who utilized neither treatment. The baseline demographic and clinical characteristics, expenditures, and HCRU were balanced between the PS-matched SAC/VAL and ACEI/ARB treatment cohorts, as well as between the SAC/VAL cohort and the treatment cohort who received neither (Table 3 and eAppendix Table 5). Descriptive statistics highlighting baseline characteristics prior to matching are available in the eAppendix (eAppendix Table 3).
Prior to matching, the mean unadjusted total Parts A/B annual expenditures for the performance period were lower for patients with HFrEF treated with SAC/VAL ($28,085) compared with those for patients treated with ACEIs/ARBs ($29,021) and with those for the cohort who received neither treatment ($31,626).
After controlling for baseline characteristics and treatment adherence, the SAC/VAL cohort had total annual Medicare Parts A/B expenditures that were $875 less than those of the ACEI/ARB cohort (95% CI, –$1650 to –$100; P = .03) and $9369 less than those of the cohort who received neither treatment (95% CI, –$11,376 to –$7361; P < .01). Mean annual hospitalization and SNF expenditures for patients receiving SAC/VAL were $1778 (95% CI, –$2316 to –$1240; P < .01) and $259 (95% CI, –$383 to –$135; P < .01) less, respectively, than those of the patients treated with ACEIs/ARBs and $7049 (95% CI, –$8432 to –$5666; P < .01) and $2276 (95% CI, –$2757 to –$1796; P < .01) less than those of the patients receiving neither therapy. HF-related total Parts A/B expenditures in the SAC/VAL group were $726 higher compared with the ACEIs/ARBs group (95% CI, $476-$979; P < .01), but HF-related hospitalization expenditures were not statistically different from the cohort who received neither treatment (95% CI, –$35 to $399; P = .10) (Table 4). Results describing HCRU rates in the PS-matched cohorts are available in the eAppendix (eAppendix Table 6).
The health care system is moving toward value-based care, which incentivizes high-quality and lower-cost services. Approximately 40% of 2018 payments were through traditional Medicare FFS payments (not tied to quality), whereas approximately 25% were FFS with a link to quality and/or value, 31% were through APMs built on FFS (eg, bundled payments, shared savings), and 5% were population-based payments (eg, per member per month and global budgets), which highlights the growing presence of APMs and value-based care.42
This retrospective study evaluated the association between utilization of SAC/VAL and expenditures for patients with HFrEF within the BPCI Model 2 90-day episode and MSSP performance year frameworks. Patients with HFrEF who receive SAC/VAL have significantly lower expenditures across a number of care settings compared with patients who receive ACEIs/ARBs or neither treatment. All-cause rehospitalization expenditures were the largest cost driver for patients with HFrEF. Past research found that decreased revenue from reduced hospitalizations for patients with HF taking SAC/VAL was more than offset by the bonuses and shared savings of the APM.16,17 In the MSSP CY 2018 analysis, total mean HF-related Parts A/B expenditures were higher among the SAC/VAL cohort compared with the ACEIs/ARBs group. This may be due to inconsistency in coding practices for patients treated with SAC/VAL vs non-SAC/VAL therapies. Previous research found that hospital inpatient claims are more likely to be coded with an HF diagnosis for patients treated with SAC/VAL than ACEIs/ARBs.43 It is important to note that Part D prescription drug costs are out of scope of the MSSP and BPCI and thus are not evaluated in the present study. Prior research has already examined the impact of SAC/VAL utilization on total cost of care from the payer and societal perspectives.19,44 In contrast, our study is designed to assess utilization of SAC/VAL and conceptually how it may affect provider performance under MSSP and BPCI from the viewpoint of providers.
In 2021, the ACC released its expert consensus decision pathway that recommends an ACEI, ARB, or ARNI as frontline therapy, with ARNI preferred, for management of HFrEF. Approximately 50% of patients with HFrEF included in the BPCI analysis and 34% of the patients in the MSSP analysis received neither SAC/VAL nor an ACEI/ARB, reflecting significant room for improvement in optimizing medical therapies for patients with HF. Despite proven clinical efficacy, guideline recommendations, and quality improvement initiatives, uptake of SAC/VAL has been relatively low in the United States. In this study, use among BPCI and MSSP participants was low (less than 6%) among beneficiaries. Barriers to uptake of SAC/VAL include the medication cost45 and, in inpatient settings, the required washout period for switching from an ACEI/ARB to SAC/VAL that can increase length of stay.46 The value-based incentives under APMs mitigate both of these barriers as provider performance is measured by relatively longer-term patient outcomes (up to 90 days post discharge under BPCI and annually under MSSP) as opposed to a single clinical encounter under the traditional FFS model.
We believe that findings from this study contribute substantial evidence and data for informing the dialogue around the value of SAC/VAL and other innovative medicines, particularly in the context of value-based payment models such as the BPCI and MSSP examined in this study. Given that APMs have seen mixed success in the cardiovascular space,36 this study’s ability to tie health economic outcomes specific to APMs to the use of SAC/VAL, a guideline-directed medication therapy for treatment of HF, underscores the need for high-quality care in the context of program performance. This study is the first to provide real-world evidence that SAC/VAL for management of HF could provide benefit in reducing expensive medical events in the context of APMs such as BPCI and MSSP and could, in turn, improve health system and APM performance.
Several study limitations should be noted. First, the study population included Medicare FFS beneficiaries, thus the study findings may not be generalizable to the commercially insured or other insured patient populations. Clinical information such as disease stage and HF severity (eg, New York Heart Association functional class), laboratory results (eg, ejection fraction values, blood pressure values), and beneficiaries’ preferences for health care services are important factors that may influence the study outcomes of interest. Those data elements are not available in administrative claims data and were not accounted for in the analyses. In the absence of clinical information, this study utilized common methods for claims-based research to identify patients with HFrEF.43,47-49 Although administrative claims are an excellent data source for capturing total health care expenditures, the ascertainment of clinical or disease-specific outcomes such as HF hospitalizations may be subject to misclassification bias due to inconsistent or inaccurate recording of ICD-10 diagnosis codes. The analysis also excluded beneficiaries in BPCI who died prior to the end of the episode, which eliminated a subset of beneficiaries with at-risk health status (ie, survival bias). However, this decision mirrored the design of the BPCI Model 2 and MSSP to allow for the examination of patient outcomes from the perspective of providers participating in these programs. In addition, patients who were nonadherent or had a treatment gap were also excluded from the data analysis. Thus, the study findings are not generalizable to patients with poor treatment adherence.
Optimizing utilization of SAC/VAL for beneficiaries with HFrEF could reduce medical expenditures and improve provider performance in Medicare APMs.
The study team acknowledges Emily Gillen for her support in drafting the manuscript and Kevin Dietz for his analytic support.
Author Affiliations: Novartis Pharmaceuticals Corporation (XS, CA, JC), East Hanover, NJ; Avalere Health (TTS, GS, MA, MF, AP), Washington, DC.
Source of Funding: Novartis Pharmaceuticals Corporation.
Author Disclosures: Dr Shen, Dr Abbas, and Mr Cristino are employed by Novartis Pharmaceuticals Corporation and own Novartis stock. Novartis manufactures Entresto (sacubitril/valsartan), which is an FDA-approved therapy for management of chronic heart failure by reducing cardiovascular death and hospitalization. Mr Schwartz is an employee of Avalere Health, which received consulting fees from Novartis. Mr Sullivan, Mr Adelsberg, Mr Francis, and Ms Petrilla were previously employees of Avalere Health when this study was completed. Ms Petrilla is currently an employee of Inovalon, which receives consulting fees from life sciences companies, including Novartis.
Authorship Information: Concept and design (XS, TTS, GS, MA, MF, AP, CA, JC); acquisition of data (MA); analysis and interpretation of data (XS, TTS, GS, MA, MF, AP, CA, JC); drafting of the manuscript (XS, TTS, GS, AP, JC); critical revision of the manuscript for important intellectual content (XS, TTS, AP, CA, JC); statistical analysis (MA, MF); obtaining funding (XS); administrative, technical, or logistic support (TTS, GS, MF, AP); and supervision (TTS, AP, JC).
Send Correspondence to: Xian Shen, PhD, Novartis Pharmaceuticals Corporation, One Health Plaza, East Hanover, NJ 07936-1080. Email: email@example.com.
1. Virani SS, Alonso A, Benjamin EJ, et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics—2020 update: a report from the American Heart Association. Circulation. 2020;141(9):e139-e596. doi:10.1161/CIR.0000000000000757
2. Roger VL. Epidemiology of heart failure. Circ Res. 2013;113(6):646-659. doi:10.1161/CIRCRESAHA.113.300268
3. Mozaffarian D, Benjamin EJ, Go AS, et al; American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Executive summary: heart disease and stroke statistics—2016 update: a report from the American Heart Association. Circulation. 2016;133(4):447-454. doi:10.1161/CIR.0000000000000366
4. Benjamin EJ, Blaha MJ, Chiuve SE, et al; American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics—2017 update: a report from the American Heart Association. Circulation. 2017;135(10):e146-e603. doi:10.1161/CIR.
5. Steinberg BA, Zhao X, Heidenreich PA, et al; Get With the Guidelines Scientific Advisory Committee and Investigators. Trends in patients hospitalized with heart failure and preserved left ventricular ejection fraction: prevalence, therapies, and outcomes. Circulation. 2012;126(1):65-75. doi:10.1161/CIRCULATIONAHA.111.080770
6. Owan TE, Hodge DO, Herges RM, Jacobsen SJ, Roger VL, Redfield MM. Trends in prevalence and outcome of heart failure with preserved ejection fraction. N Engl J Med. 2006;355(3):251-259. doi:10.1056/NEJMoa052256
7. Gerber Y, Weston SA, Redfield MM, et al. A contemporary appraisal of the heart failure epidemic in Olmsted County, Minnesota, 2000 to 2010. JAMA Intern Med. 2015;175(6):996-1004. doi:10.1001/jamainternmed.2015.0924
8. Ward C, Ewald E, Koenig K, Schluterman N. Prevalence and health care expenditures among Medicare beneficiaries aged 65 years and over with heart conditions. CMS. December 2017. Accessed April 13, 2020. https://www.cms.gov/Research-Statistics-Data-and-Systems/Research/MCBS/Downloads/HeartConditions_DataBrief_2017.pdf
9. Braunschweig F, Cowie MR, Auricchio A. What are the costs of heart failure? Europace. 2011;13(suppl 2):ii13-ii17. doi:10.1093/europace/eur081
10. Jackson SL, Tong X, King RJ, Loustalot F, Hong Y, Ritchey MD. National burden of heart failure events in the United States, 2006 to 2014. Circ Heart Fail. 2018;11(12):e004873. doi:10.1161/CIRCHEARTFAILURE.117.004873
11. Shetty S, Malik AH, Ali A, Yang YC, Briasoulis A, Alvarez P. Characteristics, trends, outcomes, and costs of stimulant-related acute heart failure hospitalizations in the United States. Int J Cardiol. 2021;331:158-163. doi:10.1016/j.ijcard.2021.01.060
12. Kilgore M, Patel HK, Kielhorn A, Maya JF, Sharma P. Economic burden of hospitalizations of Medicare beneficiaries with heart failure. Risk Manag Healthc Policy. 2017;10:63-70. doi:10.2147/RMHP.S130341
13. Lesyuk W, Kriza C, Kolominsky-Rabas P. Cost-of-illness studies in heart failure: a systematic review 2004-2016. BMC Cardiovasc Disord. 2018;18(1):74. doi:10.1186/s12872-018-0815-3
14. Maddox TM, Januzzi JL Jr, Allen LA, et al. 2021 update to the 2017 ACC Expert Consensus Decision Pathway for Optimization of Heart Failure Treatment: answers to 10 pivotal issues about heart failure with reduced ejection fraction: a report of the American College of Cardiology Solution Set Oversight Committee. J Am Coll Cardiol. 2021;77(6):772-810. doi:10.1016/j.jacc.2020.11.022
15. Jessup M. Neprilysin inhibition—a novel therapy for heart failure. N Engl J Med. 2014;371(11):1062-1064. doi:10.1056/NEJMe1409898
16. McMurray JJ, Packer M, Desai AS, et al; PARADIGM-HF Investigators and Committees. Angiotensin–neprilysin inhibition versus enalapril in heart failure. N Engl J Med. 2014;371(11):993-1004. doi:10.1056/NEJMoa1409077
17. Shafrin J, Aliyev ER, Brauer M, Park S, Shen X. Alternative payment models and innovation: a case study of US health system adoption of sacubitril/valsartan to treat acute decompensated heart failure. J Med Econ. 2020;23(12):1450-1460. doi:10.1080/13696998.2020.1825454
18. Velazquez EJ, Morrow DA, DeVore AD, et al. Angiotensin–neprilysin inhibition in acute decompensated heart failure. N Engl J Med. 2019;380(6):539-548. doi:10.1056/NEJMoa1812851
19. Albert NM, Swindle JP, Buysman EK, Chang C. Lower hospitalization and healthcare costs with sacubitril/valsartan versus angiotensin-converting enzyme inhibitor or angiotensin-receptor blocker in a retrospective analysis of patients with heart failure. J Am Heart Assoc. 2019;8(9):e011089. doi:10.1161/JAHA.118.011089
20. Lewis EF, Claggett BL, McMurray JJV, et al. Health-related quality of life outcomes in PARADIGM-HF. Circ Heart Fail. 2017;10(8):e003430. doi:10.1161/CIRCHEARTFAILURE.116.003430
21. Balmforth C, Simpson J, Shen L, et al. Outcomes and effect of treatment according to etiology in HFrEF: an analysis of PARADIGM-HF. JACC Heart Fail. 2019;7(6):457-465. doi:10.1016/j.jchf.2019.02.015
22. Chandra A, Lewis EF, Claggett BL, et al. Effects of sacubitril/valsartan on physical and social activity limitations in patients with heart failure: a secondary analysis of the PARADIGM-HF trial. JAMA Cardiol. 2018;3(6):498-505. doi:10.1001/jamacardio.2018.0398
23. Docherty KF, McMurray JJV. PIONEER-HF: a new frontier in the role of neprilysin inhibition in the management of heart failure with reduced ejection fraction. Cardiovasc Res. 2019;115(13):e136-e139. doi:10.1093/cvr/cvz223
24. Costanzo MR. Similar yet different: examining the effects of sacubitril/valsartan by race in the PIONEER-HF trial. JACC Heart Fail. 2020;8(10):867-869. doi:10.1016/
25. Ambrosy AP, Braunwald E, Morrow DA, et al; PIONEER-HF Investigators. Angiotensin receptor-neprilysin inhibition based on history of heart failure and use of renin-angiotensin system antagonists. J Am Coll Cardiol. 2020;76(9):1034-1048. doi:10.1016/j.jacc.2020.06.073
26. Čerlinskaitė K, Hollinger A, Mebazaa A, Cinotti R. Finding the balance between costs and quality in heart failure: a global challenge. Eur J Heart Fail. 2018;20(8):1175-1178. doi:10.1002/ejhf.1195
27. Feeley TW, Mohta NS. New marketplace survey: transitioning payment models: fee-for-service to value-based care. NEJM Catal Innov Care Deliv. 2018;4(6).
28. Burwell SM. Setting value-based payment goals—HHS efforts to improve US health care. N Engl J Med. 2015;372(10):897-899. doi:10.1056/NEJMp1500445
29. Brown B, Crapo J. The key to transitioning from fee-for-service to value-based reimbursement. Health Catalyst. 2017. Accessed April 13, 2020. https://bit.ly/41vBj2q
30. Nunlist MM, Uiterwyk S, Nicoletti B. Gambling on the transition from fee-for-service to value-based care. Fam Pract Manag. 2014;21(5):6-7.
31. Schroeder SA, Frist W; National Commission on Physician Payment Reform. Phasing out fee-for-service
payment. N Engl J Med. 2013;368(21):2029-2032. doi:10.1056/NEJMsb1302322
32. Innovation models. CMS. Accessed April 13, 2020.
33. Hirsch JA, Leslie-Mazwi TM, Barr RM, et al. The Bundled Payments for Care Improvement initiative. J Neurointerv Surg. 2016;8(5):547-548. doi:10.1136/neurintsurg-2015-011746
34. CMS, HHS. Medicare program; Medicare shared savings program: accountable care organizations. final rule. Fed Regist. 2011;76(212):67802-67990.
35. CMS, HHS. Medicare program; Medicare shared savings program: accountable care organizations. final rule. Fed Regist. 2015;80(110):32691-32845.
36. Joynt Maddox K, Bleser WK, Crook HL, et al; American Heart Association Value-Based Models Learning Collaborative. Advancing value-based models for heart failure: a call to action from the value in healthcare initiative’s value-based models learning collaborative. Circ Cardiovasc Qual Outcomes. 2020;13(5):e006483. doi:10.1161/CIRCOUTCOMES.120.006483
37. Accountable care organization participants. CMS. 2018. Accessed April 13, 2020. https://bit.ly/3Ie18v0
38. Bundled Payments for Care Improvement (BPCI) initiative: general information. CMS. 2018. Accessed April 13, 2020. https://bit.ly/3J1UQAr
39. Risk adjustment. CMS. Updated May 27, 2022. Accessed June 20, 2022. https://go.cms.gov/3IDjRQT
40. Caliendo M, Kopeinig S. Some practical guidance for the implementation of propensity score matching. J Econ Surv. 2008;22(1):31-72. doi:10.1111/j.1467-6419.2007.00527.x
41. Zhang Z, Kim HJ, Lonjon G, Zhu Y; AME Big-Data Clinical Trial Collaborative Group. Balance diagnostics after propensity score matching. Ann Transl Med. 2019;7(1):16. doi:10.21037/atm.2018.12.10
42. APM measurement: progress of alternative payment models. Health Care Payment Learning and Action Network. 2019. Accessed April 13, 2020. https://hcp-lan.org/workproducts/apm-methodology-2019.pdf
43. Tan NY, Sangaralingham LR, Sangaralingham SJ, Yao X, Shah ND, Dunlay SM. Comparative effectiveness of sacubitril-valsartan versus ACE/ARB therapy in heart failure with reduced ejection fraction. JACC Heart Fail. 2020;8(1):43-54. doi:10.1016/j.jchf.2019.08.003
44. Gaziano TA, Fonarow GC, Velazquez EJ, Morrow DA, Braunwald E, Solomon SD. Cost-effectiveness of sacubitril-valsartan in hospitalized patients who have heart failure with reduced ejection fraction. JAMA Cardiol. 2020;5(11):1236-1244. doi:10.1001/jamacardio.2020.2822
45. Shah KS, Ziaeian B, Mody FV, Nsair A, Fonarow GC. Perceived barriers to sacubitril/valsartan use in patients with heart failure. J Card Fail. 2018;24(8):S99. doi:10.1016/j.cardfail.2018.07.378
46. Sauer AJ, Cole R, Jensen BC, et al. Practical guidance on the use of sacubitril/valsartan for heart failure. Heart Fail Rev. 2019;24(2):167-176. doi:10.1007/s10741-018-9757-1
47. Desai RJ, Patorno E, Vaduganathan M, et al. Effectiveness of angiotensin-neprilysin inhibitor treatment versus renin-angiotensin system blockade in older adults with heart failure in clinical care. Heart. 2021;107(17):1407-1416. doi:10.1136/heartjnl-2021-319405
48. Griffin J, Cheng R. Real-world utilisation of angiotensin-neprilysin inhibitors in older adults with heart failure. Heart. 2021;107(17):1364-1366. doi:10.1136/heartjnl-2021-319545
49. Desai AS, Claggett BL, Packer M, et al; PARADIGM-HF Investigators. Influence of sacubitril/valsartan (LCZ696) on 30-day readmission after heart failure hospitalization. J Am Coll Cardiol. 2016;68(3):241-248. doi:10.1016/