County-Level Variation in Triple Guideline-Directed Medical Therapy in Heart Failure With Reduced Ejection Fraction

Background Current guidelines recommend simultaneous initiation of multidrug guideline-directed medical therapy classes for heart failure with reduced ejection fraction. Objectives The purpose of this study was to evaluate county-level variation in use of triple guideline-directed medical therapy, defined as simultaneous prescription fills for beta-blockers, renin-angiotensin system inhibitors or angiotensin receptor neprilysin inhibitors, and mineralocorticoid receptor antagonists, in heart failure with reduced ejection fraction. Methods We conducted a cohort study using Medicare Fee-for-Service claims data (parts A, B, and D between 2013 and 2019). Features of counties including area-level indicators of poverty, employment, and educational attainment and aggregated patient-level sociodemographic and medical history variables were compared by quintiles of triple therapy use. A multilevel logistic regression model was constructed to estimate the contextual effect of clustering by counties, which was expressed as a median OR. Results 304,857 patients from 2,600 counties (83% of all U.S. counties) were included. The median for triple therapy use was 14.3% (IQR: 10.3%-18.8%) across included counties with a wide variation (range: 0%-54.5%). Compared to counties in the highest use quintile, counties in lowest triple therapy use quintile had worse area-level indicators of socioeconomic status (% unemployment 6.8% vs 6.2%). Counties in lowest quintile had higher proportion of Black patients (13.3% vs 5.7% in highest quintile) and patients with low-income subsidy (29.3% vs 25.8% in highest quintile). The median OR was 1.30 (95% CI: 1.28-1.33). Conclusions We observed variation in triple therapy use across counties in the United States with suboptimal local use patterns correlating with indicators of socioeconomic disadvantage.

I n patients with heart failure with reduced ejection fraction (HFrEF), consensus guidelines recommend rapid sequence initiation of available guidelinedirected medical therapy (GDMT), including evidence-based beta-blockers, reninangiotensin system inhibitors (RASi) or angiotensin receptor neprilysin inhibitors, mineralocorticoid receptor antagonists (MRAs), and sodium-glucose co-transporter 2 inhibitors (SGLT2). 1 Prior research has documented substantial variation in use of GDMT among patients with HFrEF, 2,3 yet there is a paucity of information regarding the use of multidrug GDMT regimen in nationally representative cohorts of patients with HFrEF treated in routine clinical care.
Geographic variation in outcomes of HF including reduced and preserved ejection fraction subtypes, including HF-related mortality, is well described; 4,5 yet, little is known regarding geographic variation in treatment utilization in the United States.One study based on the Get With The Guidelines-Heart Failure (GWTG-HF) registry noted no differences in quality of care measures, which included GDMT use, across census regions in the United States. 6However, grouping of geographic regions into just 4 broad census regions may have masked important intraregion differences.A more recent investigation conducted in patients from the Veteran's Affairs revealed important geographic variation in GDMT use; however, owing to the peculiarities of the Veteran's Affairs health system which includes w98% males and provides a generally superior access to health care services, 7 it is unclear if those findings are generalizable to the U.S. population.Medicare enrollees represent a critically important population to investigate for HFrEF since a large majority of these patients are insured with this program. 8The burden of worsening HF in Medicare enrolled patients with HFrEF in routine clinical care is substantial with 1year cumulative incidence estimated to be 42.3%. 9It is plausible to hypothesize that with optimization of GDMT in this population, a portion of this high burden may be addressable.In this study, we aimed to address this knowledge gap and describe utiliza-   1A).Among the individual GDMT classes, beta-blockers were used most frequently and MRA least frequently (Figure 1B).Counties in the lowest quintile of triple therapy appeared to be concentrated in the South (Figure 2).Among patient-level sociodemographic factors, proportion of Black patients and patients with lowincome subsidy was higher in lower use quintiles.
Among medical history variables, counties in the lowest use quintile tended to have a higher burden of comorbid conditions including diabetes, renal dysfunction, and chronic obstructive pulmonary disease.1.9%-2.6%) of the residual variation in triple therapy use that persisted after adjustment for included demographics and medical history variables was explained by systematic differences between counties.The MOR was 1.30 (95% CI: 1.28-1.33),suggesting that when comparing 2 identical patients from randomly selected counties, the odds of triple therapy use were 30% higher if the patient moved to higher use vs lower use county.fraction subtypes, in the United States is welldescribed. 4,5A state-level analysis found that Southern states of Alabama, Mississippi, Oklahoma, and Arkansas consistently had among the highest age-adjusted mortality rates between 1999 and 2017. 4r findings suggest that some of these same geographic regions also lag in implementation of optimal GDMT, which represents an important opportunity to address the variation in mortality observed in prior studies.As clinical evidence and practice guidelines strongly support multidrug, comprehensive medical therapy initiated simultaneously or in rapid sequence, it is critical to consider and address existing disparities.Importantly, our analysis demonstrates that the most vulnerable and high-risk patients, including those in communities with the greatest socioeconomic disadvantages and those with multimorbid cardiometabolic conditions, continue to be those least well served in terms of implementation of disease-modifying HF therapy. 17,18ile a complex set of issues related to access, health literacy, and affordability may play a role in explaining our observations, it is important to explicitly consider the known disparities when periods and similarly exists in other countries. 22other study identified that among Veterans with HFrEF, racial and ethnic minority patients were not less likely to receive GDMT; however, they were less likely to be treated with appropriate dose uptitration to target. 3 important consideration when evaluating realworld use patterns of medications is cost implications for patients.Modeling studies have demonstrated multidrug regimen to be cost-effective at $150,000 per quality adjusted life year threshold in HFrEF. 23,24However, the cost burden on patients remains high even after Medicare coverage, especially for treatments that are not available as generics including sacubitril/valsartan and SGLT2 inhibitors with an average out of pocket costs estimated to be about $1,300 annually for triple therapy and $2,200 for quadruple therapy when these 2 medications are used as parts of the multidrug regimen. 25While the low use of triple therapy we observed in our study could be partly explained by high cost sharing for sacubitril/valsartan, we note that use of other RASi options including angiotensin converting enzyme inhibitors and angiotensin receptor blockers which are available as generics has relatively low out-ofpocket burden with an estimated annual cost of $159 to the patients with Medicare coverage. 25Therefore, our findings are likely not fully explained by financial deterrence.Some degree of clinical inertia, lack of knowledge or urgency among providers, and reservations due to increased side effect burden may play a role in explaining low triple therapy use observed in our study.Furthermore, in 2025 under the provisions of inflation reduction act, out-of-pocket cost will be capped at $2,000 annually for Medicare beneficiaries, which should help avoiding cost-related nonuse of medication to some extent. 26me strengths and limitations of the current study deserve discussion.Representativeness, including a national sample of patients with HFrEF treated in ambulatory care settings and hospitals is a major strength of the current study; to our knowledge, this is the largest study assessing county-level variation in the uptake of GDMT in patients with HF.Reliable capture of medication use through prescription dispensing records is another strength.
However, misclassification of HF phenotype and comorbid conditions is possible in the context of this of GDMT classes at county level in a populationbased cohort of Medicare enrollees across the United States using data from 2013 to 2019.Since SGLT2 inhibitors were only recently approved for the indication of HFrEF, we restricted this evaluation to use of triple therapy consisting of beta-blockers, RASi (including angiotensin receptor blockers, angiotensin-converting enzyme inhibitors) or sacubitril/valsartan, and MRA.METHODS DATA SOURCE AND STUDY COHORT.Medicare feefor-service claims data (2013-2019) from part A (inpatient coverage), part B (outpatient coverage), and part D (prescription medications) were used to conduct this study.A cohort of patients including incident and prevalent HF cases was assembled based on diagnosis codes of HF in inpatient or outpatient claims.Cohort entry date was defined as date of a medical encounter with a recorded HF diagnosis after a 6-month baseline period of continuous enrollment in Medicare parts A, B, and D. We required patients to have procedure codes for echocardiography or cardiac catheterization, in the 30 days prior to (and including) the cohort entry date, to improve the specificity of HF diagnosis and ensure close contact with the health care system. 9Given the lack of EF results availability, we then applied a validated claims-based probabilistic phenotyping model using 35 predictors identified in the baseline period to classify patients into HFrEF vs heart failure with preserved ejection fraction (HFpEF) (overall accuracy 82%; positive predicted value for HFrEF 73% in internal and external validation).
OF TRIPLE THERAPY.In the multilevel logistic regression model, we observed that model fit substantially improved after including random effects for counties compared to a fixed effects-only model (P value for likelihood ratio test <0.0001).Based on the calculated variance partition coefficient, we noted that 2.3% (95% CI:

FIGURE 1
FIGURE 1 Guideline-Directed Medical Therapy Use in Heart Failure With Reduced Ejection Fraction, Medicare Data 2013 to 2019 developing and evaluating novel interventions to bridge implementation gaps in care of HF patients. 19,20In counties with suboptimal use, community-based interventions, such as those which have proven effective in other disease states, 21 hold promise and require prospective evaluation.Our findings documenting low use of GDMT in HFrEF add to an existing body of literature detailing pervasive implementation gaps in the United States.In an U.S.-based ambulatory HF registry CHAMP (Change the Management of Patients with Heart Failure), Greene et al 2 previously reported that fewer than 1 in 4 patients were treated simultaneously with beta-blockers, RASi, and MRAs.Recent multinational data found these implementation gaps have not been closed in more contemporary time

FIGURE 2
FIGURE 2 Triple Therapy Use by Counties in Contiguous United States, Medicare Data 2013 to 2019 ADDRESS FOR CORRESPONDENCE: Dr Rishi J. Desai, Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030-R, Boston, Massachusetts 02120, USA.E-mail: rdesai@bwh.harvard.edu.PERSPECTIVES COMPETENCY IN PATIENT CARE: In patients with HFrEF, consensus guidelines recommend rapid sequence initiation of available GDMT.TRANSLATIONAL OUTLOOK: In a nationally representative sample of Medicare enrollees including patients from 2,600 counties (83% of the total counties in the U.S.), we observed that fewer than 1 in 7 patients with HFrEF were on 3 GDMT classes concurrently.We also documented appreciable geographic variation across counties and identified several features of counties with low observed use including worse indicators of area-level SES, higher density of Black patients, and greater medical complexity.Effective implementation of GDMT will require addressing structural and environmental barriers to access.

Table 2
provides OR and 95% CIs for the fixed effect variables which generally indicated lower odds of treatment with triple therapy among patients with various comorbid conditions.The effect sizes were especially large for history of conditions that indicate higher potential for adverse event risks with triple therapy including hyperkalemia (0.62 [95% CI: 0.60ciable geographic variation across counties and identified several features of counties with low observed use including worse indicators of area-level SES, higher density of Black patients, and greater medical complexity.Geographic variation in mortality related to heart failure, including reduced and preserved ejection

TABLE 1
County-Level Characteristics by Triple Therapy Use Quintiles cohort included patients with HFrEF with an average age of 76 years.Among older patients, treatment decisions are often influenced by considerations related to adverse events of medications.In our multivariable models, we observed history of hyperkalemia, renal dysfunction, and hypotension to be associated with lower odds of triple therapy use.Therefore, in agnostics, Sanofi, and Tricog Health; has speaker engagements with AstraZeneca, Novartis, and Roche Diagnostics; and participates on clinical trial committees for studies sponsored by Galmed, Novartis, Bayer AG, Occlutech, and Impulse Dynamics.All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.

TABLE 2
Patient-Level Characteristics and Their Association With Receipt of Values are n (%) or mean AE SD unless otherwise indicated.a Estimates from the multilevel logistic regression model with triple therapy use as the outcome variable including random effects for counties and fixed effects for variables in this table.