Abstract
Background
Black and Hispanic patients with heart failure (HF) have a higher risk of adverse clinical outcomes. Currently, it is unclear whether there are disparities in referral to outpatient HF management programs based on race and ethnicity.
Methods and Results
We used the American Heart Association GWTG‐HF (Get With The Guidelines‐Heart Failure) registry to examine 402 225 patients hospitalized for acute HF from January 1, 2010 to December 31, 2021. Logistic regression was used to examine the association of race and ethnicity with the likelihood of referral to outpatient HF management programs, adjusted for demographics, hospital characteristics, distressed community index score, comorbidities, and indicators of HF severity. Of the 402 225 patients hospitalized for acute HF during the study period (mean age 72 years, 47% female, 44% with ejection fraction <40%), 220 354 (55%) patients were referred to an outpatient HF management program at hospital discharge. In fully adjusted models, patients who self‐identified as Hispanic (odds ratio [OR], 0.87 [95% CI, 0.84–0.90]), Asian (OR, 0.74 [95% CI, 0.70–0.78]), and other (American Indian, Alaska Native, Hawaiian Native, or Pacific Islander, OR, 0.85 [95% CI, 0.82–0.89]) had a lower likelihood of referral to outpatient HF management programs than White patients. There were no differences in referral likelihood between Black and White patients.
Conclusions
In the GWTG‐HF registry, patients from minoritized racial and ethnic groups, aside from Black patients, were less likely than White patients to be referred to outpatient HF management programs after HF hospitalization. Addressing these differences in referral practices may improve HF outcomes in minoritized communities.
Keywords: health disparities, heart failure, heart failure management, race and ethnicity, referral
Subject Categories: Heart Failure, Disparities, Race and Ethnicity
Nonstandard Abbreviations and Acronyms
- DCI
distressed community index
- GWTG‐HF
Get With The Guidelines‐Heart Failure
Clinical Perspective.
What Is New?
In the GWTG‐HF (Get With The Guidelines‐Heart Failure) registry, patients from minoritized racial and ethnic groups, aside from Black patients, were less likely than White patients to be referred to outpatient heart failure management programs at hospital discharge.
What Are the Clinical Implications?
Variation in referral to outpatient heart failure management programs across racial and ethnic groups may perpetuate existing disparities in heart failure outcomes.
Further investigations are needed to identify individual‐ and system‐level factors that contribute to these differences in referral practices.
In patients with heart failure (HF), timely referral for HF specialty care is recommended to improve patient outcomes. 1 , 2 HF management programs involve specialized, multidisciplinary care models that include optimization of guideline‐directed medical and device therapies, patient education, nutrition counseling, behavioral modifications, management of comorbidities, and assessment of candidacy for advanced HF therapies. 3 Prior studies have shown that patients with HF who participated in HF management programs had fewer rehospitalizations, lower mortality risk, and better quality of life compared with patients with HF who received conventional care. 3 , 4 , 5
In the United States, racial and ethnic disparities in HF outcomes persist despite rapid pharmacologic and therapeutic advancements in HF care. Black and Hispanic adults have the highest risk for incident HF, as well as HF‐related hospitalizations and mortality. 6 , 7 , 8 Asian adults have better HF outcomes compared with other racial and ethnic groups, 6 , 9 although data in this population are limited and may not be generalizable to all Asian subgroups due to vast heterogeneity in this population. Data on HF among American Indian and Alaska Native individuals are also underreported, but prior studies have shown greater burden of traditional cardiovascular risk factors that may increase HF risk in this population. 10
Although lower socioeconomic status and barriers to health care access contribute to the persistence of disparities in HF outcomes, the role of clinical practice patterns across race and ethnicity has yet to be fully explored. Prior studies confirm that Black and Hispanic patients are more likely to be cared for at poorer‐quality hospitals, likely as a result of structural racism, which continues to contribute to the widespread disparities in health outcomes. 11 However, it is currently unknown whether there are variations in patterns of referral to outpatient HF management programs across racial and ethnic groups. Identification of potential variations in clinical practice patterns is crucial to informing future interventions to reduce racial and ethnic disparities in HF care. Using the American Heart Association GWTG‐HF (Get With The Guidelines‐Heart Failure) registry, we examined referral patterns to outpatient HF management programs for patients hospitalized for acute HF between January 1, 2010 and December 31, 2021 at 1056 participating hospital sites across the United States. We hypothesized that patients from minoritized racial and ethnic groups would be less likely than White patients to be referred to outpatient HF management programs after hospitalization for acute HF, and that neighborhood‐level socioeconomic status may influence these patterns.
METHODS
Data Availability
The data used for this study cannot be made publicly available by the authors. Researchers interested in accessing data from the American Heart Association GWTG‐HF registry for research or validation purposes can submit proposals at www.heart.org/qualityresearch.
Data Source and Study Population
This retrospective cohort study included 1 493 407 patients hospitalized for acute HF between January 1, 2010 and December 31, 2021 at 1056 hospital sites participating in the American Heart Association GWTG‐HF registry, which is an ongoing quality improvement database of patients hospitalized with acute HF at participating sites throughout the United States. 12 We excluded patients with missing data for the primary exposure of interest (race or ethnicity), important covariates (sex, residential zip code or distressed community index [DCI] score), and the primary outcome of interest (referral to an outpatient HF management program). We also excluded patients with a history of heart or kidney transplant and ventricular‐assist devices because these patients are already established with an outpatient HF management program. Trained personnel at each participating site prospectively collected data on demographics, health insurance, hospital characteristics, medical history, clinical presentation, laboratory results, echocardiogram results, and clinical outcomes using standardized case report forms. Protocols were approved by institutional review boards at each participating site, including the institutional review board at Emory University. Each participating site received either human research approval to enroll cases without individual patient consent under the common rule, or a waiver of authorization and exception from subsequent review by their institutional review board. Advarra, the institutional review board for the American Heart Association, determined that this study is exempt from institutional review board oversight.
Outcome of Interest
The primary outcome of interest for this study was referral to an outpatient HF management program, which was collected by trained personnel at each participating site using standardized case reports forms. The outcome was determined to be present if referral occurred during the index hospitalization or any subsequent hospitalizations over the study period. If a patient had multiple referrals from different hospitalizations, they were counted to have the outcome only once.
Primary Exposure of Interest
The primary exposure of interest for this study was race and ethnicity. Race was self‐reported on case report forms as White, Black, Asian, or “Other.” Based on the demographic distribution of this patient population, we expect the “Other” racial and ethnic group to be composed of patients who self‐identified as American Indian, Alaska Native, Hawaiian Native, or Pacific Islander. Hispanic ethnicity was self‐reported as yes or no.
Distressed Community Index Score
Each patient's residential zip code as documented on the case report forms was linked to a DCI score based on the Economic Innovation Group DCI database. 13 This database captures 7 metrics of community‐level socioeconomic indicators including percentage of individuals 25 years or older without a high school diploma, housing vacancy rate, percentage of unemployment among adults ages 25 to 54 years, poverty rate, median household income, percent change in employment (number of jobs), and percent change in business establishments. DCI scores range between 0 to 100, with higher scores corresponding to communities with greater distress. We adjusted for DCI to minimize confounding from socioeconomic factors.
Patient‐ and Hospital‐Level Covariates
The rationale for our control covariates are as follows: We adjusted for demographic factors (age and sex) because prior studies have suggested that age and sex are important predictors for referral to cardiac services. For instance, it has been shown that women and older individuals are less likely to be referred to cardiac rehabilitation. 14 We adjusted for DCI because we wanted to minimize confounding from socioeconomic factors. We adjusted for health insurance because insurance status can determine whether a referral can be placed or not. We adjusted for length of hospital stay because longer length of stay may give providers more time and hence, opportunity to place a referral before discharge. We adjusted for hospital characteristics (bed size, location, and teaching site) because larger, academic hospitals in urban settings may have more resources to establish HF management programs than smaller, community hospitals in rural areas. Lastly, we adjusted for comorbidities and indicators of HF severity because patients who are sicker may have more obvious reasons to be referred. Missing data were low (<5%) for all covariates, except insurance (21%), prior HF hospital admissions (62%), heart rate (8%), mean arterial pressure (9%), serum sodium (15%), and estimated glomerular filtration rate (6%). For the insurance covariate, patients with missing data were grouped with the “no insurance or not documented” subgroup. Given our primary outcome was determined to be present if referral occurred during the index hospitalization or any subsequent hospitalizations over the study period, we wanted to account for the increased number of opportunities an individual would have with more HF hospitalizations and hence, derived a HF hospital admissions covariate that included the number of prior HF hospitalizations and the number of subsequent HF hospitalizations during the study period (not including the index hospitalization). For patients with only a single index hospitalization, they were grouped with the “unknown” subgroup if they had missing data for prior HF hospitalizations.
Statistical Analysis
The distribution of data was presented as mean±SD for normally distributed continuous variables, median (interquartile range) for skewed continuous variables, and proportions for categorical variables. Baseline characteristics were compared across racial and ethnic groups as well as DCI quartiles using Kruskal–Wallis test for continuous variables and Pearson's chi‐square test for categorical variables. We also compared baseline characteristics between patients excluded for missing data versus patients included in the final analytic. Multivariable logistic regression models were used to examine the association of race and ethnicity with the likelihood of referral to outpatient HF management program. Three sequential models were estimated: model 1 adjusted for sociodemographic factors (age, sex, DCI score, insurance type), hospital characteristics (bed size, hospital location, and teaching site), and length of stay; model 2 additionally adjusted for patient comorbidities (coronary artery disease [CAD], diabetes, hypertension, hyperlipidemia, peripheral vascular disease, cerebrovascular accident/transient ischemic accident, atrial fibrillation/atrial flutter, obstructive lung disease, chronic kidney disease, and depression); and model 3 additionally adjusted for indicators of HF severity (number of HF hospitalizations and ejection fraction, heart rate, mean arterial pressure, serum sodium, and estimated glomerular filtration rate on admission during the index hospitalization). Given a significant proportion (21%) of the study population was missing insurance data, sensitivity analyses were performed using multiple imputation to generate expected values for the missing insurance variable. All covariates in model 3 of the primary analysis were included in the imputation model. A total of 5 imputed data sets were generated and then analyzed using multivariable logistic regression. These results were combined for comparison with the fully adjusted model (model 3) of the primary analysis. Multivariable logistic regression models with race or ethnicity‐DCI interaction terms were also performed to examine the association of race and ethnicity with the likelihood of referral to outpatient HF management program within different DCI quartiles. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).
RESULTS
Study Population
From January 1, 2010 to December 31, 2021, 1 493 407 adult patients were hospitalized for acute HF at 1056 participating hospitals in the GWTG‐HF registry. Of these, 58 425 patients were excluded for missing data on race and ethnicity or sex, 3781 patients were excluded for history of heart or kidney transplant and ventricular‐assist devices, 733 509 patients were excluded for missing data on referral to outpatient HF management program, and 295,467 patients were excluded for missing residential zip code or DCI scores (Figure 1). This resulted in a final analytic cohort of 402 225 adult patients hospitalized for acute HF. A comparison of baseline characteristics between patients excluded for missing data versus those included in the final analytic cohort demonstrated similar baseline characteristics across all 3 groups, except there was a larger proportion of patients with no insurance/not documented insurance and unknown number of HF hospital admissions in the excluded cohorts (Table S1).
Figure 1. Consolidated Standards of Reporting Trials diagram for selection of Get With The Guidelines‐Heart Failure final analysis cohort.
HF indicates heart failure.
Baseline Characteristics
Among the 402 225 patients included in the analytic cohort, the mean age was 72 (SD±15) years and 47% were women (Table 1). White patients were older compared with other racial and ethnic groups. Black and Hispanic patients were more likely to live in communities with higher DCI scores, indicating greater socioeconomic distress relative to other racial and ethnic groups. The proportion of uninsured patients was comparable across all groups, except patients who self‐identified as “other” had a higher proportion of patients with no insurance or insurance not documented. Hospital characteristics, including bed size, location (urban versus rural), and teaching site, were similar across all groups. Most patients were hospitalized at large, urban, and teaching hospitals. Comorbidities including CAD, hyperlipidemia, peripheral vascular disease, atrial fibrillation/atrial flutter, and depression were higher among White patients compared with other racial and ethnic groups. Black patients had the lowest prevalence of CAD but had the highest prevalence of hypertension and HF with reduced ejection fraction. The proportion of patients with diabetes was highest among Hispanic and Asian patients. Asian patients were also more likely to have a history of chronic kidney disease. The number of HF hospitalizations and heart rate and serum sodium on index admission were similar across all groups.
Table 1.
Baseline Characteristics of Study Cohort, Stratified by Race and Ethnicity
| Baseline characteristics | Total (n=402 225) | White (n=272 165) | Black (n=80 081) | Hispanic (n=27 748) | Asian (n=7392) | Other (n=14 839) | P value |
|---|---|---|---|---|---|---|---|
| Age, y | 72±15 | 75±13 | 63±15 | 67±15 | 72±15 | 69±16 | <0.0001 |
| Female sex | 190305 (47%) | 129110 (47%) | 38441 (48%) | 12712 (46%) | 3377 (46%) | 6665 (45%) | <0.0001 |
| Distressed community index score | 51±29 | 45±28 | 68±26 | 61±25 | 42±26 | 49±28 | <0.0001 |
| Length of stay, d | 6±13 | 6±11 | 6±14 | 6±22 | 6±11 | 6±8 | <0.0001 |
| Insurance type | |||||||
| Medicaid | 47427 (12%) | 19273 (7%) | 17967 (22%) | 6666 (24%) | 1424 (19%) | 2097 (14%) | <0.0001 |
| Medicare | 168931 (42%) | 129417 (48%) | 24707 (31%) | 8494 (31%) | 2583 (35%) | 3730 (25%) | |
| Other | 89278 (22%) | 64919 (24%) | 15031 (19%) | 5341 (19%) | 1505 (20%) | 2482 (17%) | |
| No Insurance or not documented | 96589 (24%) | 58556 (22%) | 22376 (28%) | 7247 (26%) | 1880 (25%) | 6530 (44%) | |
| Bed size (number of beds) | 446±214 | 425±209 | 517±220 | 478±211 | 416±188 | 415±210 | <0.0001 |
| Hospital location | |||||||
| Urban | 380775 (95%) | 255508 (94%) | 76927 (96%) | 27363 (99%) | 6885 (93%) | 14092 (95%) | <0.0001 |
| Rural | 21340 (5%) | 16583 (6%) | 3148 (4%) | 359 (1%) | 507 (7%) | 743 (5%) | |
| Teaching site | 318171 (79%) | 210710 (77%) | 68300 (85%) | 22552 (81%) | 5714 (77%) | 10895 (73%) | <0.0001 |
| Coronary artery disease | 181816 (45%) | 134694 (50%) | 26958 (34%) | 11072 (40%) | 3203 (43%) | 5889 (40%) | <0.0001 |
| Diabetes | 182122 (45%) | 116649 (43%) | 39349 (49%) | 15734 (57%) | 3836 (52%) | 6554 (44%) | <0.0001 |
| Hyperlipidemia | 223625 (56%) | 160024 (59%) | 38013 (48%) | 14577 (53%) | 4135 (56%) | 6876 (46%) | <0.0001 |
| Hypertension | 332861 (83%) | 223688 (82%) | 69437 (87%) | 22791 (82%) | 6079 (82%) | 10866 (73%) | <0.0001 |
| Peripheral vascular disease | 44990 (11%) | 34129 (13%) | 6651 (8%) | 2527 (9%) | 431 (6%) | 1252 (8%) | <0.0001 |
| Cerebrovascular accident/transient ischemic accident | 62693 (16%) | 43699 (16%) | 12535 (16%) | 3674 (13%) | 984 (13%) | 1801 (12%) | <0.0001 |
| Atrial fibrillation/flutter | 155545 (40%) | 123569 (46%) | 18178 (23%) | 6943 (26%) | 2320 (32%) | 4535 (34%) | <0.0001 |
| Chronic obstructive pulmonary disease/asthma | 133376 (33%) | 94518 (35%) | 26594 (33%) | 6826 (25%) | 1492 (20%) | 3946 (27%) | <0.0001 |
| Chronic kidney disease | 92326 (23%) | 57187 (21%) | 22238 (28%) | 7138 (26%) | 2331 (32%) | 3432 (23%) | <0.0001 |
| Depression | 61293 (15%) | 47642 (18%) | 8176 (10%) | 3409 (12%) | 454 (6%) | 1612 (11%) | <0.0001 |
| Heart failure hospital admissions | |||||||
| 0 | 66468 (17%) | 46868 (17%) | 12589 (16%) | 4320 (16%) | 1049 (14%) | 1642 (11%) | <0.0001 |
| 1 | 35434 (9%) | 23135 (9%) | 7711 (10%) | 2869 (10%) | 655 (9%) | 1064 (7%) | |
| 2 | 13394 (3%) | 8107 (3%) | 3525 (4%) | 1157 (4%) | 226 (3%) | 379 (3%) | |
| >2 | 13882 (4%) | 7329 (3%) | 4548 (6%) | 1341 (5%) | 242 (3%) | 422 (3%) | |
| Unknown | 273047 (68%) | 186726 (69%) | 51708 (65%) | 18061 (65%) | 5220 (71%) | 11332 (76%) | |
| Ejection fraction | |||||||
| EF<40% | 175476 (43.6%) | 108760 (40.0%) | 43354 (54.1%) | 12993 (46.8%) | 3130 (42.3%) | 7239 (48.8%) | <0.0001 |
| EF 41%–49% | 37904 (9.4%) | 26848 (9.9%) | 6572 (8.2%) | 2511 (9.0%) | 636 (8.6%) | 1337 (9.0%) | |
| EF≥50% | 182890 (45.5%) | 132824 (48.8%) | 29067 (36.3%) | 11736 (42.3%) | 3478 (47.1%) | 5785 (39.0%) | |
| Heart rate on admission, bpm | 87±20 | 86±20 | 90±20 | 87±20 | 86±21 | 87±20 | <0.0001 |
| Mean arterial pressure on admission, mm Hg | 100±20 | 99±19 | 108±22 | 102±20 | 101±20 | 100±20 | <0.0001 |
| Serum sodium on admission, mEq/L | 138±7 | 137±7 | 139±8 | 137±7 | 136±7 | 137±8 | <0.0001 |
| Estimated glomerular filtration rate on admission, mL/min per 1.73 m2 | 44 (25–67) | 45 (27–68) | 41 (21–64) | 45 (22–71) | 39 (19–63) | 44 (24–70) | <0.0001 |
Values are mean±SD, median (interquartile range), or N (%).EF indicates ejection fraction.
Association of Race and Ethnicity With Likelihood of Referral to Outpatient HF Management Program
Of the 402 225 patients hospitalized for acute HF during the study period, 220 354 (55%) patients were referred to an outpatient HF management program at hospital discharge. Of the 220 354 patients referred to an outpatient HF management program, 146 056 (66%) patients self‐identified as White, 47 671 (22%) as Black, 15 538 (7%) as Hispanic, 3696 (2%) as Asian, and 7,393 (3%) as "other". Within each racial and ethnic group, 146,056 (54%) of White patients, 47,671 (60%) of Black patients, 15,538 (56%) of Hispanic patients, 3,696 (50%) of Asian patients, and 7,393 (50%) of “other” patients were referred to an outpatient HF management program (Table S2). In fully adjusted models controlling for demographics, DCI, insurance, hospital characteristics, patient comorbidities, and indicators of HF severity, Hispanic (OR, 0.87 [95% CI, 0.84–0.90] P<0.0001), Asian (OR, 0.74 [95% CI 0.70–0.78] P<0.0001), and “other” (OR, 0.85 [95% CI 0.82–0.89] P<0.0001) patients had a lower likelihood of referral to outpatient HF management program than White patients (Table 2). There were no differences in referral likelihood between Black and White patients after multivariable adjustments (Table 2). Similar results were found in sensitivity analyses with imputed insurance variables (Table S3).
Table 2.
Likelihood of Referral to Outpatient Heart Failure Management Program at Hospital Discharge According to Patient Race and Ethnicity
| Race or ethnicity | Referral to outpatient HF management program | |||
|---|---|---|---|---|
| Unadjusted OR (95% CI) | Model 1 OR (95% CI) | Model 2 OR (95% CI) | Model 3 OR (95% CI) | |
| White | Reference | Reference | Reference | Reference |
| Black | 1.27 (1.25–1.29)* | 1.00 (0.98–1.02) | 1.03 (1.01–1.05)* | 1.02 (0.99–1.04) |
| Hispanic | 1.10 (1.07–1.13)* | 0.91 (0.89–0.94)* | 0.91 (0.88–0.93)* | 0.87 (0.84–0.90)* |
| Asian | 0.86 (0.82–0.90)* | 0.74 (0.70–0.78)* | 0.74 (0.70–0.78)* | 0.74 (0.70–0.78)* |
| Other† | 0.86 (0.83–0.89)* | 0.83 (0.80–0.86)* | 0.78 (0.75–0.81)* | 0.85 (0.82–0.89)* |
Model 1: Adjusted for age, sex, distressed community index score, length of stay, insurance, bed size, hospital location, and teaching site. Model 2: Model 1 + adjusted for coronary artery disease, diabetes, hypertension, hyperlipidemia, peripheral vascular disease, cerebrovascular accident/transient ischemic accident, atrial fibrillation/flutter, chronic obstructive pulmonary disease/asthma, chronic kidney disease, and depression. Model 3: Model 2 + adjusted for number of prior and subsequent HF hospitalizations, ejection fraction, and heart rate, mean arterial pressure, serum sodium, and estimated glomerular filtration rate on index admission.HF indicates heart failure; and OR, odds ratio.
P value <0.05.
Other indicates patients who self‐identified as American Indian, Alaska Native, Hawaiian Native, or Pacific Islander.
Association of Race and Ethnicity With Likelihood of Referral to Outpatient HF Management Program Within Distressed Community Index Quartiles
When examining the likelihood of referral to outpatient HF management, there was a significant interaction between race or ethnicity and DCI score (P value for interaction <0.0001). In fully adjusted models, patients who self‐identified as Hispanic and "other" had lower likelihood of referral to outpatient HF management programs than White patients in the lower 3 DCI quartiles but had comparable referral likelihood in communities with the greatest socioeconomic distress (Table 3 and Figure 2). Asian patients had lower likelihood of referral to outpatient HF management programs across all DCI quartiles (Table 3 and Figure 2). Black patients had higher likelihood of referral to outpatient HF management programs than White patients within DCI quartiles 1 and 3 but had no difference in referral likelihood within DCI quartiles 2 and 4 (Table 3 and Figure 2).
Table 3.
Likelihood of Referral to Outpatient Heart Failure Management Program at Hospital Discharge According to Patient Race or Ethnicity and DCI
| Referral to outpatient HF management program | ||||
|---|---|---|---|---|
| DCI quartile 1 (least distressed) | DCI quartile 2 | DCI quartile 3 | DCI quartile 4 (most distressed) | |
| White | Reference | Reference | Reference | Reference |
| Black | 1.10 (1.04–1.16)* | 0.98 (0.93–1.02) | 1.05 (1.01–1.09)* | 1.00 (0.97–1.04) |
| Hispanic | 0.89 (0.82–0.97)* | 0.73 (0.68–0.78)* | 0.87 (0.82–0.92)* | 0.96 (0.91–1.02) |
| Asian | 0.87 (0.79–0.95)* | 0.73 (0.65–0.82)* | 0.60 (0.53–0.67)* | 0.76 (0.62–0.92)* |
| Other† | 0.75 (0.68–0.82)* | 0.79 (0.72–0.87)* | 0.89 (0.82–0.97)* | 1.05 (0.95–1.17) |
Adjusted for race or ethnicity‐DCI interaction, age, sex, length of stay, insurance, bed size, hospital location, teaching site, coronary artery disease, diabetes, hypertension, hyperlipidemia, peripheral vascular disease, cerebrovascular accident/transient ischemic accident, atrial fibrillation/flutter, chronic obstructive pulmonary disease/asthma, chronic kidney disease, depression, number of prior and subsequent HF hospitalizations, ejection fraction, and heart rate, mean arterial pressure, serum sodium, and estimated glomerular filtration rate on index admission. DCI indicates distressed community index; and HF, heart failure.
P‐value <0.05.
Other indicates patients who self‐identified as American Indian, Alaska Native, Hawaiian Native, or Pacific Islander.
Figure 2. Forest plot for likelihood of referral to outpatient heart failure management program at hospital discharge according to patient race or ethnicity within DCI quartiles.
*Reference group=White. Adjusted for race or ethnicity‐DCI interaction, age, sex, length of stay, insurance, bed size, hospital location, teaching site, coronary artery disease, diabetes, hypertension, hyperlipidemia, peripheral vascular disease, cerebrovascular accident/transient ischemic accident, atrial fibrillation/atrial flutter, chronic obstructive lung disease/asthma, chronic kidney disease, depression, number of prior and subsequent heart failure hospitalizations, ejection fraction, and heart rate, mean arterial pressure, serum sodium, and estimated glomerular filtration on index admission. DCI indicates distressed community index (higher DCI score=greater community distress); and OR, odds ratio.
Association of Other Patient, Hospital, and Clinical Characteristics With Likelihood of Referral to Outpatient HF Management Program
In fully adjusted models, older age, female sex, lack of insurance, higher DCI scores (greater community distress), cerebrovascular accident/transient ischemic accident, chronic obstructive pulmonary disease/asthma, chronic kidney disease, and depression were associated with lower likelihood of referral to outpatient HF management programs at hospital discharge. In contrast, rural, teaching, and larger hospitals, Medicaid and Medicare insurance, CAD, hyperlipidemia, multiple HF hospital admissions, and lower ejection fraction were associated with higher likelihood of referral to outpatient HF management programs at hospital discharge (Table S4).
DISCUSSION
In a large contemporary cohort of US adult patients hospitalized for acute HF, patients who self‐identified as Hispanic, Asian, or other (American Indian, Alaska Native, Hawaiian Native, or Pacific Islander) had a lower likelihood of being referred to outpatient HF management programs at hospital discharge when compared with patients who self‐identified as White. In contrast, there were no differences in referral likelihood between Black and White patients. In fact, in some neighborhoods, Black patients had a higher referral likelihood than White patients. Our findings reveal that there are differential access to outpatient HF management programs following HF hospitalization, which may perpetuate existing racial and ethnic disparities in HF outcomes.
Despite significant improvements in HF care over the past decades achieved with the introduction of novel medical and device therapies and more emphasis on multidisciplinary care models, the inequitable distribution of therapies and access to HF specialty care remains a major concern. 11 Widening disparities in HF outcomes among individuals from minoritized racial and ethnic groups and lower socioeconomic status have been observed. 15 , 16 A comprehensive examination of how clinical practice patterns differ based on race and ethnicity is needed to identify disparities in HF care to inform future interventions that will enable more equitable HF care for all Americans.
The benefits of outpatient HF management programs are well‐supported, with prior studies showing less HF rehospitalizations, lower mortality rates, and improved quality of life. 3 , 4 , 5 Despite these known benefits, outpatient HF management programs are underused, with low referral rates being a major barrier to access. 17 , 18 Although the referral rate in this study was higher than what has been mentioned in prior studies, 17 , 18 it is still unacceptably low when considering the significant reduction in morbidity and mortality among patients with HF managed with outpatient HF management programs as opposed to conventional care. 5 The novelty of our study is that patients from minoritized racial and ethnic groups appear to be disproportionately affected by these low referral rates. More research is needed to identify strategies to improve HF referral rates, especially among minoritized racial and ethnic groups.
Our findings support the growing evidence showing that individuals from minoritized racial and ethnic groups are less likely to receive cardiovascular specialty care when compared with White patients. For instance, Black, Hispanic, and Asian patients are 20%, 36%, and 50% less likely, respectively, than White patients to be referred to cardiac rehabilitation following hospitalization for CAD. 19 Black and Hispanic patients are also less likely to receive primary care from a cardiologist when admitted for acute HF. 20 , 21 Even among patients who are able to access HF specialty care, Black and Hispanic patients are less likely than White patients to receive advanced HF therapies, such as left ventricular assist devices and heart transplantation, after adjusting for HF severity, quality of life, and social determinants of health. 22 , 23 , 24 , 25 These findings suggest that structural racism or provider bias may play a role in perpetuating racial and ethnic disparities in HF care and outcomes.
In our analysis, we found that patients from minoritized racial and ethnic groups, except for Black patients, were less likely be referred to outpatient HF management programs following hospitalization for acute HF. Black patients either had no difference or higher likelihood of being referred to outpatient HF management programs when compared with White patients, which is discordant with the current body of literature showing that Black adults have worse HF outcomes. 26 A potential explanation is that hospitals participating in the GWTG‐HF registry may have increased awareness about HF disparities in this population and subsequently have better care processes compared with hospitals not participating in the GWTG‐HF registry. 27 Existing data on cardiovascular outcomes among Asian individuals generally show lower cardiovascular‐related mortality compared with White individuals, 9 , 28 which may suggest lower disease severity and hence explain why Asian patients had a lower likelihood in referral to outpatient HF management programs. However, data on HF outcomes among Asian Americans should be interpreted cautiously due to its limited availability, as well as the challenges that come with using broad groupings in research when there is significant heterogeneity within the population. Data on HF outcomes among Hispanic individuals generally show worse HF outcomes when compared with White individuals. 6 , 8 Recent data from the Centers for Disease Control and Prevention's Wide‐Ranging Online Data for Epidemiologic Research death certificate database revealed that Hispanic adults under age 65, particularly Hispanic men, experienced HF‐related deaths at accelerated rates between 1999 and 2018 when compared with White adults. 29 This may be due to variation in clinical practice patterns based on race and ethnicity, such as lower referral likelihood to outpatient HF management program in Hispanic patients versus White patients as seen in our analysis. Our study also revealed that patients who self‐identified as “other” (American Indian, Alaska Native, Native Hawaiian, or Pacific Islander) had lower likelihood of referral to outpatient HF management programs, revealing potential disparities in HF care among these populations. A recent large cohort study of American Indian and Alaska Native Medicare beneficiaries revealed a high burden of HF. 30 Likewise, prior studies have shown high prevalence of cardiometabolic risk factors among Native Hawaiian and Pacific Islander individuals, 31 which may increase their risk for HF. Currently, data on HF outcomes in these populations are lacking and are another important avenue for future disparity research.
When examining the association of race and ethnicity with referral likelihood within different DCI quartiles, we found persistent racial and ethnic disparities. Patients from minoritized racial and ethnic groups were more likely to live in neighborhoods with higher DCI scores; however, the referral patterns differed by race and ethnicity. Among Black patients, the only difference in referral likelihood existed among those patients living in DCI quartile 1 and 3, where there was a higher likelihood of being referred to outpatient HF management programs when compared with White patients. 26 One hypothesis is the increasing awareness that Black patients have worse clinical outcomes with HF with reduced ejection fraction, 32 , 33 perhaps prompting earlier referral to outpatient HF management programs. Asian patients had lower likelihood of referral to outpatient HF management programs compared with White patients across all community‐level socioeconomic strata. Asian patients who lived in communities with higher DCI scores had worse left ventricular function and kidney function compared with White patients within the same DCI quartiles (Table S5) and continued to have lower referral likelihood after adjustments for these markers of HF severity. This may be due to stereotypical perceptions that Asian patients generally have better health outcomes, causing their HF risks and needs to be overlooked. 34 Further investigation into how stereotypes and implicit biases have shaped the Asian American experience within the health care system is needed. Finally, patients who self‐identified as Hispanic or "other" ethnicities had lower likelihood of being referred to outpatient HF management programs across all DCI quartiles except for those living in the most distressed communities. This may be because lower socioeconomic status is associated with worse HF severity and more frequent readmissions, 35 perhaps allowing clinicians to more easily recognize the need for HF specialty care in these populations.
There are limitations to our study that are worth noting. One limitation is the potential for selection bias given that patients with missing zip‐code or DCI scores and missing referral outcomes were excluded. Since these excluded groups had greater proportions of patients who were uninsured or did not have insurance information documented (Table S1), our study may have excluded individuals at greatest risk of being underreferred and thus our findings may be an underestimation of the true referral rates. Another limitation is that 21% of the study population was missing insurance data, which is an important factor in determining whether a patient is referred for additional care. Although our multiple imputation analysis reassuringly showed similar findings to our primary analysis, there is still some potential for biased estimates. We were also unable to fully account for HF severity due to significant missingness in important biomarkers (eg, troponin and B‐type natriuretic peptide) and because of the large proportion of patients with unknown number of prior HF hospital admissions. This makes it difficult to identify whether referrals were placed at similar points in the disease trajectory across racial and ethnic groups. Another limitation of our study is that patients who self‐identified as American Indian, Alaska Native, Hawaiian Native, or Pacific Islander were grouped into a single category called “other,” which was a predetermined category on the case report forms. Although the aggregation of data on different racial and ethnic groups is not ideal because it can mask health disparities that often exist among subgroups, we felt it was important to include these populations in our study, especially given the scant data we have on HF outcomes among American Indian, Alaska Native, Hawaiian Native, and Pacific Islander adults. Additionally, our findings may not be generalizable to patients hospitalized for acute HF at hospitals not included in the GWTG‐HF registry. Participation in the GWTG‐HF registry is voluntary and hospitals that choose to participate may be more likely to participate in evidence‐based practices. It is also worth noting that there may be variability in the organization and services provided by the HF disease management programs patients were referred to. Lastly, it is important to note that referral rates do not necessarily translate to participation because of potential barriers to access, such as lack of transportation, conflicting work schedules, or medical mistrust. Further studies are needed to examine whether referral rates are reflective of participation rates in outpatient HF management programs across different racial and ethnic groups. Nevertheless, the strengths of this study are that it was a large, multicenter study, it included racial and ethnic populations who are traditionally underrepresented in clinical research, and it addressed a knowledge gap in HF referral patterns across racial and ethnic groups. With respect to population health, our study indicated that referral patterns were especially low among patients who self‐identified as Asian, which is a group that is historically overlooked in disparity research.
Conclusions
In conclusion, in the GWTG‐HF registry, patients from minoritized racial and ethnic groups, aside from Black patients, were less likely than White patients to be referred to outpatient HF management programs at hospital discharge. These differences in referral practices may perpetuate existing racial and ethnic disparities in HF outcomes. Further studies are needed to identify individual‐ and system‐level factors that contribute to these differences in referral practices.
Sources of Funding
Maggie Wang is supported by the Multidisciplinary Research Training to Reduce Inequalities in Cardiovascular Health (METRIC) National Institutes of Health/National Heart, Lung, and Blood Institute T32 grant (T32HL130025). Shivani A. Patel is supported in part by the Georgia Diabetes Translation Research Center (P30DK111024). Khadijah Breathett receives funding from the National Heart, Lung, and Blood Institute K01HL142848, R01HL159216, R01HL160734, the Health Resources and Services Administration of the US Department of Health and Human Services, and the Indiana Clinical and Translational Sciences Institute. Alanna A. Morris is supported by funding from the American Heart Association and Merck. The GWTG‐HF (Get With The Guidelines–Heart Failure) program is provided by the American Heart Association. GWTG‐HF is sponsored, in part, by Novartis, Boehringer Ingelheim, Novo Nordisk, AstraZeneca, Bayer, Tylenol, and Alnylam Pharmaceuticals.
Disclosures
Alanna A. Morris reports consulting fees or honoraria from Abbott, BI Lilly, Cytokinetics, Ionis, Merck, Novo Nordisk, and Regeneron. Alanna A. Morris reports research funding from Ionis, Merck, and Novo Nordisk. Neal W. Dickert reports consulting and research funding from Abiomed and research funding from Merck. Ersilia M. DeFilippis serves on a clinical trial committee for Abiomed. Gregg C. Fonarow reports consulting for Abbott, Amgen, AstraZeneca, Bayer, Boehinger Ingelheim, Cytokinetics, Eli Lilly, Johnson & Johnson, Medtronic, Merck, Novartis, and Pfizer. Ambarish Pandey has received research support from the National Institute of Health, American Heart Association, Applied Therapeutics, Roche, Ultromics, and ScPharmaceuticals; has received honoraria outside of the present study as an advisor/consultant for Tricog Health Inc, Lilly USA, Rivus, Cytokinetics, Roche Diagnostics, Axon therapies, Medtronic, Edward Lifesciences, Science37, Novo Nordisk, Bayer, Merck, Sarfez Pharmaceuticals, Emmi Solutions, Anumana, Semler Scientific, Ultromics, Merck, Encarda, Kieele Health, Acorai; and has received nonfinancial support from Pfizer and Merck. Dr. Pandey is also a consultant for Palomarin with stock options. The remaining authors have no disclosures to report.
Supporting information
Tables S1–S5
This article was sent to Sula Mazimba, MD, MPH, Associate Editor, for review by expert referees, editorial decision, and final disposition.
This work was presented in part at the American College of Cardiology Scientific Sessions, April 6–8, 2024, in Atlanta, GA.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.124.036900
For Sources of Funding and Disclosures, see page 10.
References
- 1. Heidenreich P, Bozkurt B, Aguilar D, Allen LA, Byun J, Colvin MM, Drazner MH, Dunlay SM, Evers LR, Fang JC, et al. AHA/ACC/HFSA guideline for the Management of Heart Failure: executive summary. A report of the American College of Cardiology/American Heart Association joint committee on clinical practice guidelines. Circulation. 2022;145:e876–e894. doi: 10.1161/CIR.0000000000001063 [DOI] [PubMed] [Google Scholar]
- 2. Morris AA, Khazanie P, Drazner MH, Albert NM, Breathett K, Cooper LB, Eisen HJ, O'Gara P, Russell SD. Guidance for timely and appropriate referral of patients with advanced heart failure: a scientific statement from the American Heart Association. Circulation. 2021;144:e238–e250. doi: 10.1161/CIR.0000000000001016 [DOI] [PubMed] [Google Scholar]
- 3. Fonarow GC, Stevenson LW, Walden JA, Livingston NA, Steimle AE, Hamilton MA, Moriguchi J, Tillisch JH, Woo MA. Impact of a comprehensive heart failure management program on hospital readmission and functional status of patients with advanced heart failure. J Am Coll Cardiol. 1997;30:725–732. doi: 10.1016/S0735-1097(97)00208-8 [DOI] [PubMed] [Google Scholar]
- 4. Rich MW, Beckham V, Wittenberg C, Leven CL, Freedland KE, Carney RM. A multidisciplinary intervention to prevent the readmission of elderly patients with congestive heart failure. N Engl J Med. 1995;333:1190–1195. doi: 10.1056/nejm199511023331806 [DOI] [PubMed] [Google Scholar]
- 5. McAlister FA, Stewart S, Ferrua S, McMurray JJJV. Multidisciplinary strategies for the management of heart failure patients at high risk for admission. J Am Coll Cardiol. 2004;44:810–819. doi: 10.1016/j.jacc.2004.05.055 [DOI] [PubMed] [Google Scholar]
- 6. Ziaeian B, Kominski GF, Ong MK, Mays VM, Brook RH, Fonarow GC. National differences in trends for heart failure hospitalizations by sex and race/ethnicity. Circ Cardiovasc Qual Outcomes. 2017;10:e003552. doi: 10.1161/CIRCOUTCOMES.116.003552 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Glynn P, Lloyd‐Jones DM, Feinstein MJ, Carnethon M, Khan SS. Disparities in cardiovascular mortality related to heart failure in the United States. J Am Coll Cardiol. 2019;73:2354–2355. doi: 10.1016/j.jacc.2019.02.042 [DOI] [PubMed] [Google Scholar]
- 8. Bahrami H, Kronmal R, Bluemke DA, Olson J, Shea S, Liu K, Burke GL, Lima JAC. Differences in the incidence of congestive heart failure by ethnicity: the multi‐ethnic study of atherosclerosis. Arch Intern Med. 2008;168:2138–2145. doi: 10.1001/archinte.168.19.2138 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Jose PO, Frank ATH, Kapphahn KI, Goldstein BA, Eggleston K, Hastings KG, Cullen MR, Palaniappan LP. Cardiovascular disease mortality in Asian Americans. J Am Coll Cardiol. 2014;64:2486–2494. doi: 10.1016/j.jacc.2014.08.048 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Breathett K, Sims M, Gross M, Jackson EA, Jones EJ, Navas‐Acien A, Taylor H, Thomas KL, Howard BV. Cardiovascular health in American Indians and Alaska natives: a scientific statement from the American Heart Association. Circulation. 2020;141:e948–e959. doi: 10.1161/cir.0000000000000773 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Morris A, Shah KS, Enciso JS, Hsich E, Ibrahim NE, Page R, Yancy C, Butler J, Hsich E, Pressler SB, et al. The impact of health care disparities on patients with heart failure. J Card Fail. 2022;28:1169–1184. doi: 10.1016/j.cardfail.2022.04.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Smaha LA. The American Heart Association get with the guidelines program. Am Heart J. 2004;148:S46–S48. doi: 10.1016/j.ahj.2004.09.015 [DOI] [PubMed] [Google Scholar]
- 13. Distressed Communities Index . Economic Innovation Group. Accessed January 10, 2023. https://eig.org/distressed‐communities/.
- 14. Sun EY, Jadotte YT, Halperin W. Disparities in cardiac rehabilitation participation in the United States: a Systematic review and meta‐analysis. J Cardiopulm Rehabil Prev. 2017;37:2–10. doi: 10.1097/hcr.0000000000000203 [DOI] [PubMed] [Google Scholar]
- 15. Jain V, Minhas AMK, Khan SU, Greene SJ, Pandey A, Van Spall HGC, Fonarow GC, Mentz RJ, Butler J, Khan MS. Trends in HF hospitalizations among young adults in the United States from 2004 to 2018. JACC Heart Fail. 2022;10:350–362. doi: 10.1016/j.jchf.2022.01.021 [DOI] [PubMed] [Google Scholar]
- 16. Jain V, Minhas AMK, Morris AA, Greene SJ, Pandey A, Khan SS, Fonarow GC, Mentz RJ, Butler J, Khan MS. Demographic and regional trends of heart failure–related mortality in young adults in the US, 1999–2019. JAMA Cardiol. 2022;7:900–904. doi: 10.1001/jamacardio.2022.2213 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Gharacholou SM, Hellkamp AS, Hernandez AF, Peterson ED, Bhatt DL, Yancy CW, Fonarow GC. Use and predictors of heart failure disease management referral in patients hospitalized with heart failure: insights from the get with the guidelines program. J Card Fail. 2011;17:431–439. doi: 10.1016/j.cardfail.2010.12.005 [DOI] [PubMed] [Google Scholar]
- 18. Gravely S, Ginsburg L, Stewart DE, Mak S, Grace SL. Referral and use of heart failure clinics: what factors are related to use? Can J Cardiol. 2012;28:483–489. doi: 10.1016/j.cjca.2011.11.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Li S, Fonarow GC, Mukamal K, Xu H, Matsouaka RA, Devore AD, Bhatt DL. Sex and racial disparities in cardiac rehabilitation referral at hospital discharge and gaps in long‐term mortality. J Am Heart Assoc. 2018;7:e008088. doi: 10.1161/JAHA.117.008088 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Eberly LA, Richterman A, Beckett AG, Wispelwey B, Marsh RH, Manchanda ECC, Chang CY, Glynn RJ, Brooks KC, Boxer R, et al. Identification of racial inequities in access to specialized inpatient heart failure Care at an Academic Medical Center. Circ Heart Fail. 2019;12:e006214. doi: 10.1161/CIRCHEARTFAILURE.119.006214 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Breathett K, Liu WG, Allen LA, Daugherty SL, Blair IV, Jones J, Grunwald GK, Moss M, Kiser TH, Burnham E, et al. African Americans are less likely to receive care by a cardiologist during an intensive care unit admission for heart failure. JACC: Heart Fail. 2018;6:413–420. doi: 10.1016/j.jchf.2018.02.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Cascino TM, Colvin MM, Lanfear DE, Richards B, Khalatbari S, Mann DL, Taddei‐Peters WC, Jeffries N, Watkins DC, Stewart GC, et al. Racial inequities in access to ventricular assist device and transplant persist after consideration for preferences for care: a report from the REVIVAL study. Circ Heart Fail. 2023;16:e009745. doi: 10.1161/CIRCHEARTFAILURE.122.009745 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Wang X, Luke AA, Vader JM, Maddox TM, Maddox KEJ. Disparities and impact of Medicaid expansion on left ventricular assist device implantation and outcomes. Circ Cardiovasc Qual Outcomes. 2020;13:e006284. doi: 10.1161/CIRCOUTCOMES.119.006284 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Breathett K, Yee E, Pool N, Hebdon M, Crist JD, Knapp S, Larsen A, Solola S, Luy L, Herrera‐Theut K, et al. Does race influence decision making for advanced heart failure therapies? J Am Heart Assoc. 2019;8:e013592. doi: 10.1161/jaha.119.013592 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Dixon DD, Knapp SM, Ilonze O, Lewsey SC, Mazimba S, Mohammed S, Van Spall HGC, Breathett K. Racial and ethnic disparities in ambulatory heart failure ventricular assist device implantation and survival. JACC Heart Fail. 2023;11:1397–1407. doi: 10.1016/j.jchf.2023.05.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Nayak A, Hicks AJ, Morris AA. Understanding the complexity of heart failure risk and treatment in black patients. Circ Heart Fail. 2020;13:e007264. doi: 10.1161/CIRCHEARTFAILURE.120.007264 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Heidenreich PA, Hernandez AF, Yancy CW, Liang L, Peterson ED, Fonarow GC. Get with the guidelines program participation, process of care, and outcome for Medicare patients hospitalized with heart failure. Circ Cardiovasc Qual Outcomes. 2012;5:37–43. doi: 10.1161/CIRCOUTCOMES.110.959122 [DOI] [PubMed] [Google Scholar]
- 28. Shah NS, Xi K, Kapphahn KI, Srinivasan M, Au T, Sathye V, Vishal V, Zhang H, Palaniappan LP. Cardiovascular and cerebrovascular disease mortality in Asian American subgroups. Circ Cardiovasc Qual Outcomes. 2022;15:e008651. doi: 10.1161/CIRCOUTCOMES.121.008651 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Khan SU, Lone AN, Yedlapati SH, Dani SS, Khan MZ, Watson KE, Parwani P, Rodriguez F, Cainzos‐Achirica M, Michos ED. Cardiovascular disease mortality among Hispanic versus non‐Hispanic white adults in the United States, 1999 to 2018. J Am Heart Assoc. 2022;11:e022857. doi: 10.1161/JAHA.121.022857 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Eberly LA, Shultz K, Merino M, Brueckner MY, Benally E, Tennison A, Biggs S, Hardie L, Tian Y, Nathan AS, et al. Cardiovascular disease burden and outcomes among American Indian and Alaska Native Medicare beneficiaries. JAMA Netw Open. 2023;6:e2334923. doi: 10.1001/jamanetworkopen.2023.34923 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Mau MK, Sinclair K, Saito EP, Baumhofer KN, Kaholokula JK. Cardiometabolic health disparities in native Hawaiians and other Pacific Islanders. Epidemiol Rev. 2009;31:113–129. doi: 10.1093/ajerev/mxp004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Dries DL, Exner DV, Gersh BJ, Cooper HA, Carson PE, Domanski MJ. Racial differences in the outcome of left ventricular dysfunction. N Engl J Med. 1999;340:609–616. doi: 10.1056/nejm199902253400804 [DOI] [PubMed] [Google Scholar]
- 33. Chandra A, Skali H, Claggett B, Solomon SD, Rossi JS, Russell SD, Matsushita K, Kitzman DW, Konety SH, Mosley TH, et al. Race‐ and gender‐based differences in cardiac structure and function and risk of heart failure. J Am Coll Cardiol. 2022;79:355–368. doi: 10.1016/j.jacc.2021.11.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Parks M, Nallamothu B, Ho PM. Not so simple, nor model minority. Circ Cardiovasc Qual Outcomes. 2022;15:e008918. doi: 10.1161/CIRCOUTCOMES.122.008918 [DOI] [PubMed] [Google Scholar]
- 35. Mathews L, Ding N, Mok Y, Shin JI, Crews DC, Rosamond WD, Newton AK, Chang PP, Ndumele CE, Coresh J, et al. Impact of socioeconomic status on mortality and readmission in patients with heart failure with reduced ejection fraction: the ARIC study. J Am Heart Assoc. 2022;11:e024057. doi: 10.1161/JAHA.121.024057 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Tables S1–S5
Data Availability Statement
The data used for this study cannot be made publicly available by the authors. Researchers interested in accessing data from the American Heart Association GWTG‐HF registry for research or validation purposes can submit proposals at www.heart.org/qualityresearch.