Quality of Care and Outcomes Among Patients Hospitalized for Heart Failure in Rural vs Urban US Hospitals: The Get With The Guidelines–Heart Failure Registry

Jacob B Pierce; Uchechukwu Ikeaba; Anthony E Peters; Adam D DeVore; Karen Chiswell; Larry A Allen; Nancy M Albert; Clyde W Yancy; Gregg C Fonarow; Stephen J Greene

doi:10.1001/jamacardio.2023.0241

. 2023 Feb 20;8(4):376–385. doi: 10.1001/jamacardio.2023.0241

Quality of Care and Outcomes Among Patients Hospitalized for Heart Failure in Rural vs Urban US Hospitals

The Get With The Guidelines–Heart Failure Registry

Jacob B Pierce ¹, Uchechukwu Ikeaba ², Anthony E Peters ^2,³, Adam D DeVore ^2,³, Karen Chiswell ², Larry A Allen ⁴, Nancy M Albert ⁵, Clyde W Yancy ^6,⁷, Gregg C Fonarow ^8,⁹, Stephen J Greene ^2,^3,^✉

¹Department of Medicine, Duke University School of Medicine, Durham, North Carolina

²Duke Clinical Research Institute, Durham, North Carolina

³Division of Cardiology, Duke University School of Medicine, Durham, North Carolina

⁴Division of Cardiology, University of Colorado Anschutz Medical Campus, Aurora

⁵Nursing Institute and Kaufman Center for Heart Failure, Cleveland Clinic, Cleveland, Ohio

⁶Division of Cardiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois

⁷Deputy Editor, JAMA Cardiology

⁸Ahmanson-UCLA Cardiomyopathy Center, University of California, Los Angeles

⁹Associate Editor for Health Care Quality and Guidelines, JAMA Cardiology

Accepted for Publication: January 30, 2023.

Published Online: February 20, 2023. doi:10.1001/jamacardio.2023.0241

^✉

Corresponding Author: Stephen J. Greene, MD, Duke Clinical Research Institute, 300 W Morgan St, Durham, NC 27701 (stephen.greene@duke.edu).

Author Contributions: Drs Chiswell and Greene had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Pierce, Peters, Allen, Yancy, Fonarow, Greene.

Acquisition, analysis, or interpretation of data: Pierce, Ikeaba, DeVore, Chiswell, Albert, Fonarow, Greene.

Drafting of the manuscript: Pierce, Greene.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Ikeaba, DeVore, Chiswell.

Obtained funding: Pierce, Fonarow.

Administrative, technical, or material support: Fonarow, Greene.

Study supervision: Allen, Yancy, Fonarow, Greene.

Conflict of Interest Disclosures: Dr Pierce has received grants from the American Heart Association during the conduct of the study. Dr Peters has received grants from the National Heart, Lung, and Blood Institute during the conduct of the study as well as personal fees from Cytokinetics outside the submitted work. Dr DeVore has received grants from the American Heart Association during the conduct of the study; grants from Biofourmis, Bodyport, Cytokinetics, American Regent, National Heart, Lung, and Blood Institute, Novartis, and Story Health; personal fees from Abiomed, AstraZeneca, Cardionomic, InnaMed, LivaNova, Natera, Novartis, Novo Nordisk, Procyrion, Story Health, Vifor, and Zoll; and nonfinancial support from Abbott outside the submitted work. Dr Allen has received grants from Patient-Centered Outcomes Research Institute, American Heart Association, and National Institutes of Health as well as personal fees from ACI Clinical, Boston Scientific, Cytokinetics, Novartis, UpToDate, and Quidel during the conduct of the study. Dr Albert reported grants from Novartis and AstraZeneca as well as personal fees from Boehringer Ingelheim/Eli Lilly, Boston Scientific, Cytokinetics, Novartis, and Merck outside the submitted work. Dr Fonarow has received personal fees from Abbott, Amgen, AstraZeneca, Bayer, Cytokinetics, Eli Lilly, Janssen, Medtronic, Merck, Novartis, and Pfizer outside the submitted work. Dr Greene has received research support from the Duke University Department of Medicine Chair’s Research Award, American Heart Association, National Heart, Lung, and Blood Institute, Amgen, AstraZeneca, Bristol Myers Squibb, Cytokinetics, Merck, Novartis, Pfizer, and Sanofi as well as personal fees from Amgen, AstraZeneca, Bayer, Bristol Myers Squibb, Boehringer Ingelheim/Eli Lilly, Corteria Therapeutics, Cytokinetics, CSL Vifor, Merck, PharmaIN, Tricog Health, Sanofi, Urovant Pharmaceuticals, Roche Diagnostics, and scPharmaceuticals outside the submitted work. No other disclosures were reported.

Funding/Support: This study was funded by an American Heart Association Early Career Investigator Get With The Guidelines Database Research Seed Grant to Dr Pierce. The Get With The Guidelines–Heart Failure program is provided by the American Heart Association and sponsored, in part, by Novartis, Boehringer Ingelheim and Eli Lilly Diabetes Alliance, Novo Nordisk, Sanofi, AstraZeneca, and Bayer.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: Dr Yancy is Deputy Editor and Dr Fonarow is Associate Editor for Health Care Quality and Guidelines of JAMA Cardiology, but they were not involved in any of the decisions regarding review of the manuscript or its acceptance.

Meeting Presentation: This paper was presented at the ACC.23 Annual Scientific Session and Expo; February 20, 2023; New Orleans, Louisiana.

Data Sharing Statement: See Supplement 2.

^✉

Corresponding author.

PMCID: PMC9941973 PMID: 36806447

Key Points

Question

Are there differences in the quality of care and outcomes of patients hospitalized for heart failure at rural vs urban hospitals in the US?

Findings

In this cohort study including 774 419 patients in the Get With The Guidelines–Heart Failure registry, patients at rural hospitals were significantly less likely to be prescribed cardiac resynchronization therapy, angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, and angiotensin receptor–neprilysin inhibitor therapy at discharge, but other quality metrics were similar to urban hospitals. After adjustment for confounders, there were no significant differences between rural and urban hospitals regarding in-hospital or 30-day outcomes.

Meaning

In this study, disparities existed in the quality of heart failure care provided at rural vs urban hospitals, but in-hospital and 30-day postdischarge outcomes were similar.

This cohort study assesses quality of care and clinical outcomes for US patients hospitalized for heart failure at rural vs urban hospitals in the Get With The Guidelines–Heart Failure registry.

Abstract

Importance

Prior studies have suggested patients with heart failure (HF) from rural areas have worse clinical outcomes. Contemporary differences between rural and urban hospitals in quality of care and clinical outcomes for patients hospitalized for HF remain poorly understood.

Objective

To assess quality of care and clinical outcomes for US patients hospitalized for HF at rural vs urban hospitals.

Design, Setting, and Participants

This retrospective cohort study analyzed 774 419 patients hospitalized for HF across 569 sites in the Get With The Guidelines–Heart Failure (GWTG-HF) registry between January 1, 2014, and September 30, 2021. Postdischarge outcomes were assessed in a subset of 161 996 patients linked to Medicare claims. Data were analyzed from August 2022 to January 2023.

Main Outcomes and Measures

GWTG-HF quality measures, in-hospital mortality, length of stay, and 30-day mortality and readmission outcomes.

Results

This study included 19 832 patients (2.6%) and 754 587 patients (97.4%) hospitalized at 49 rural hospitals (8.6%) and 520 urban hospitals (91.4%), respectively. Of 774 419 included patients, 366 161 (47.3%) were female, and the median (IQR) age was 73 (62-83) years. Compared with patients at urban hospitals, patients at rural hospitals were older (median [IQR] age, 74 [64-84] years vs 73 [61-83] years; standardized difference, 10.63) and more likely to be non-Hispanic White (14 572 [73.5%] vs 498 950 [66.1%]; standardized difference, 34.47). In adjusted models, patients at rural hospitals were less likely to be prescribed cardiac resynchronization therapy (adjusted risk difference [aRD], −13.5%; adjusted odds ratio [aOR], 0.44; 95% CI, 0.22-0.92), angiotensin-converting enzyme inhibitor or angiotensin receptor blocker (aRD, −3.7%; aOR, 0.71; 95% CI, 0.53-0.96), and an angiotensin receptor–neprilysin inhibitor (aRD, −5.0%; aOR, 0.68; 95% CI, 0.47-0.98) at discharge. In-hospital mortality was similar between rural and urban hospitals (460 of 19 832 [2.3%] vs 20 529 of 754 587 [2.7%]; aOR, 0.86; 95% CI, 0.70-1.07). Patients at rural hospitals were less likely to have a length of stay of 4 or more days (aOR, 0.75; 95% CI, 0.67-0.85). Among Medicare beneficiaries, there were no significant differences between rural and urban hospitals in 30-day HF readmission (adjusted hazard ratio [aHR], 1.03; 95% CI, 0.90-1.19), all-cause readmission (aHR, 0.97; 95% CI, 0.91-1.04), and all-cause mortality (aHR, 1.05; 95% CI, 0.91-1.21).

Conclusions and Relevance

In this large contemporary cohort of US patients hospitalized for HF, care at rural hospitals was independently associated with lower use of some guideline-recommended therapies at discharge and shorter length of stay. In-hospital mortality and 30-day postdischarge outcomes were similar at rural and urban hospitals.

Introduction

Despite major advances in therapy, rates of both heart failure (HF) mortality and hospitalization are rising in the US, with evidence of persistent disparities in HF outcomes between patients living in rural vs urban communities.^1,2,3 Those living in rural areas may face unique challenges in achieving optimal cardiovascular health, potentially related to socioeconomic disadvantage, geographic proximity to specialty care, and quality of cardiovascular care.^4,5,6 Similarly, rural hospitals may have more difficulty recruiting physicians, lower patient volumes, and higher rates of uninsured patients or uncompensated care, which may limit achievement of quality metrics and threaten overall financial viability.^{7,8,9,10,11,12} These challenges faced by rural hospitals were recently highlighted in the American Heart Association Presidential Advisory and Call to Action on Rural Health.¹³ However, contemporary differences between rural and urban hospitals in the quality of care and clinical outcomes for patients hospitalized for HF remain poorly understood.

In this context, the Get With The Guidelines–Heart Failure (GWTG-HF) registry affords a unique opportunity to comprehensively characterize differences in patient-level and hospital-level characteristics, quality of care, and clinical outcomes for patients hospitalized with HF in rural vs urban hospitals in the US. The objective of this study was to characterize the existence, nature, and magnitude of any disparities in HF care and patient outcomes at rural vs urban hospitals, laying a foundation for potential subsequent public health interventions and policies aimed at ensuring equitable care for patients hospitalized at both types of US hospitals.

Methods

Data Sources

This retrospective cohort study evaluated patients hospitalized for decompensated HF at participating GWTG-HF registry sites. The GWTG-HF registry is an ongoing, nationwide quality improvement registry sponsored by the American Heart Association.^14,15 In brief, trained personnel at participating sites prospectively collect data for patients hospitalized with decompensated HF, including medical history, hospital characteristics, vital and laboratory values, echocardiographic data, medications, and in-hospital outcomes using standardized case report forms. Institutional review boards approved study protocols at each participating site. As data are used primarily for quality improvement, participant informed consent was waived under the Common Rule. To assess the availability of interventional cardiac catheterization and heart transplant services at individual hospitals, data for participating GWTG-HF sites were linked to the 2018 American Hospital Association Annual Survey. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Study Population

We identified patients 18 years or older admitted for HF to participating GWTG-HF hospitals between January 1, 2014, through September 30, 2021. Race and ethnicity were identified by self-report. Sites missing hospital characteristics were excluded. We excluded patients who were transferred to another acute care facility, discharged to hospice or palliative care, or left against medical advice. Patients with missing data for age, sex, or patient disposition were excluded. After applying exclusion criteria, this cohort was used to assess in-hospital outcomes and quality of care (Figure 1A). Baseline characteristics of patients excluded from the study cohort due to transfer to another facility (stratified by transfer out from a rural vs urban hospital) are shown in eTable 1 in Supplement 1.

Figure 1. — Selection of primary GWTG-HF cohort (A) and secondary Medicare-linked cohort (B). In analyses stratified by ejection fraction (EF), patients missing quantitative EF data were excluded. CMS indicates US Centers for Medicare & Medicaid Services; HF, heart failure.

To assess postdischarge outcomes, we subsequently linked GWTG-HF records for patients 65 years and older to available fee-for-service Medicare claims data using a validated approach.¹⁶ The same exclusion criteria were applied to the Medicare-linked cohort (Figure 1B). For analyses stratified by ejection fraction (EF), those missing quantitative EF were excluded.

Definition of Rural vs Urban Hospital

Similar to previous GWTG-HF analyses,^17,18 rural hospitals were classified according to the American Hospital Association definition, which defined rural hospitals as those not located within a metropolitan area designated by the US Office of Management and Budget and the Census Bureau. Patients were grouped by rural or urban status of the hospital at which they were admitted. To investigate whether an alternative classification scheme would significantly change rural-urban designations in our registry sample, we assessed the degree of agreement between the American Hospital Association and Rural-Urban Commuting Area (RUCA) code definitions. As per prior studies,^19,20,21,22 RUCA codes 1 through 3 were considered urban and RUCA codes 4 through 10 were considered rural. Using RUCA codes, 16 of 569 GWTG-HF hospitals (2.8%) were reclassified as either rural or urban, suggesting a high degree of agreement between the American Hospital Association and RUCA definitions (eTables 3 and 4 in Supplement 1).

Study Outcomes

Primary study outcomes were in-hospital mortality and length of stay of 4 or more days. We also assessed rates of adherence to 15 established GWTG-HF quality metrics for the index hospitalization. These measures included measurement of left ventricular function, scheduling postdischarge follow-up, scheduling postdischarge follow-up within 7 days or less, smoking cessation counseling, placement or prescription of cardiac resynchronization therapy (CRT) at discharge, deep vein thrombosis prophylaxis during hospitalization, influenza vaccination during influenza season, pneumococcal vaccination, outpatient cardiac rehabilitation program referral, outpatient HF disease management program referral, and prescription of guideline-directed medical therapy (GDMT) at discharge among eligible patients with HF with reduced EF (angiotensin-converting enzyme inhibitor [ACEI] or angiotensin receptor blocker [ARB], angiotensin receptor–neprilysin inhibitor [ARNI], β-blockers, mineralocorticoid receptor antagonist [MRA], and combination hydralazine and nitrate). Analyses of quality measures were restricted to patients eligible for each measure by treatment guidelines. Among patients in the Medicare-linked cohort, secondary outcomes included HF readmission, all-cause readmission, and all-cause mortality within 30 days of hospital discharge.

Statistical Analysis

We assessed for differences in patient and hospital characteristics between rural vs urban hospitals. Groups were compared using absolute standardized differences, whereby percentages were calculated by dividing the difference between the rural and urban means by the within-group standard deviation. An absolute standardized difference of more than 10% indicates nonnegligible imbalance between groups.

Multivariable logistic regression models were used to assess for differences between rural and urban hospitals in the primary outcomes of in-hospital mortality and length of stay (4 or more days) as well as prescription of GDMT at discharge and adherence to GWTG-HF quality metrics. Generalized estimating equations (GEE) with independence working correlation structure imposed on the models were used to adjust for in-hospital clustering. Comparisons between rural and urban hospitals were also quantified as risk differences, estimated using GEE from models with a linear link function. Among the subset of Medicare beneficiaries 65 years and older, we investigated differences in 30-day postdischarge outcomes by estimating Cox proportional hazard models. For in-hospital mortality, length of stay, and postdischarge outcomes, analyses were further stratified by EF category of 40% or less, 41% to 49%, and 50% or more. Interaction testing was performed to assess for heterogeneity by EF group in the association between rural vs urban hospital and outcome. Multivariable regression models were adjusted for 20 prespecified covariates, including demographic characteristics, medical history, and hospital characteristics (eMethods and eTable 2 in Supplement 1). Models investigating 30-day postdischarge outcomes were adjusted first for patient-level characteristics alone (ie, partially adjusted) and second for patient-level and hospital-level characteristics (ie, fully adjusted).

Additional details on the statistical methods are described in the eMethods in Supplement 1. For all analyses, P values were 2-sided, and a P value less than .05 was considered statistically significant. All statistical analyses were performed at the Duke Clinical Research Institute using SAS version 9.4 (SAS Institute).

Results

Patient Characteristics

After applying selection criteria, there were 774 419 patients included from 569 hospitals in the GWTG-HF registry (Figure 1). Of 774 419 included patients, 366 161 (47.3%) were female, and the median (IQR) age was 73 (62-83) years (Table 1). Compared with patients at urban hospitals, patients at rural hospitals were older (median [IQR] age, 74 [64-84] years vs 73 [61-83] years; standardized difference, 10.63) and more likely to be non-Hispanic White (14 572 [73.5%] vs 498 950 [66.1%]; standardized difference, 34.47) with Medicare insurance (11 027 [58.4%] vs 368 589 [51.1%]; standardized difference, 16.88). Patients at rural hospitals also tended to have higher median (IQR) EF (48% [30%-60%] vs 45% [28%-58%]) and systolic blood pressure (142 [123-165] mm Hg vs 139 [120-160] mm Hg) and lower N-terminal pro–brain natriuretic peptide (4614 [2170-9930] pg/mL vs 4854 [2126-11 117] pg/mL) compared with those at urban hospitals. Rates of comorbidities and other laboratory values on admission were generally similar between those admitted at rural and urban hospitals.

Table 1. Baseline Characteristics of Patients Hospitalized With Heart Failure at Rural and Urban Get With The Guidelines–Heart Failure Sites.

Characteristic	No. (%)			Absolute standardized difference, %^a
Characteristic	Overall (N = 774 419)	Rural (n = 19 832)	Urban (n = 754 587)	Absolute standardized difference, %^a
Demographic
Age, median (IQR), y	73 (62-83)	74 (64-84)	73 (61-83)	10.63
Sex
Female	366 161 (47.3)	9763 (49.2)	356 398 (47.2)	4.00
Male	408 258 (52.7)	10 069 (50.8)	398 189 (52.8)	4.00
Race and ethnicity^b
Asian	14 462 (1.9)	162 (0.8)	14 300 (1.9)	34.47
Black	166 850 (21.5)	4355 (22.0)	162 495 (21.5)
Hispanic	57 417 (7.4)	208 (1.0)	57 209 (7.6)
White	513 522 (66.3)	14 572 (73.5)	498 950 (66.1)
Other race	22 095 (2.9)	535 (2.7)	21 560 (2.9)
Insurance status
Private, HMO, or other	201 685 (27.3)	3983 (21.1)	197 702 (27.4)	16.88
Medicaid	129 583 (17.5)	3301 (17.5)	126 282 (17.5)
Medicare	379 616 (51.3)	11 027 (58.4)	368 589 (51.1)
Uninsured	28 679 (3.9)	574 (3.0)	28 105 (3.9)
Medical history
Ejection fraction
Median (IQR), %	45 (28-58)	48 (30-60)	45 (28-58)	10.40
HFrEF (≤40%)	329 996 (42.6)	7692 (38.8)	322 304 (42.7)	6.59
HFmrEF (41%-49%)	73 713 (9.5)	2019 (10.2)	71 694 (9.5)
HFpEF (≥50%)	342 673 (44.2)	9007 (45.4)	333 666 (44.2)
Missing	28 037 (3.6)	1114 (5.6)	26 923 (3.6)
Diabetes	365 955 (48.1)	9253 (47.2)	356 702 (48.1)	1.88
Hypertension	659 428 (86.7)	16 430 (83.8)	642 998 (86.7)	8.37
Hyperlipidemia	453 189 (59.6)	11 016 (56.2)	442 173 (59.6)	7.05
Coronary artery disease	369 216 (48.5)	9494 (48.4)	359 722 (48.5)	0.23
Prior myocardial infarction	159 615 (21.0)	4277 (21.8)	155 338 (21.0)	2.08
Stroke or transient ischemic attack	130 170 (17.1)	3127 (15.9)	127 043 (17.1)	3.21
Peripheral vascular disease	93 438 (12.3)	2251 (11.5)	91 187 (12.3)	2.54
Atrial fibrillation or flutter	317 665 (41.7)	8110 (41.4)	309 555 (41.8)	0.82
COPD or asthma	277 167 (36.4)	7937 (40.5)	269 230 (36.3)	8.54
CRT-D or CRT-P	52 137 (6.9)	1060 (5.4)	51 077 (6.9)	6.19
Chronic kidney disease	206 374 (27.1)	5461 (27.8)	200 913 (27.1)	1.66
Current smoking	131 642 (17.2)	3672 (18.5)	127 970 (17.2)	3.58
Laboratory values and vitals at admission, median (IQR)
Systolic blood pressure, mm Hg	139 (120-160)	142 (123-165)	139 (120-160)	11.86
Heart rate, beats per min	84 (72-98)	85 (72-100)	84 (72-98)	6.96
BMI^c	30.0 (25.1-36.7)	30.5 (25.6-37.4)	30.0 (25.1-36.7)	5.71
Potassium, mEq/L	4.2 (3.8-4.6)	4.2 (3.8-4.6)	4.2 (3.8-4.6)	2.54
Serum sodium, mEq/L	139 (136-141)	138 (135-140)	139 (136-141)	6.88
Serum creatinine, mg/dL	1.3 (1.0-1.9)	1.3 (0.9-1.8)	1.3 (1.0-1.9)	1.01
eGFR, mL/min/1.73 m²^d	50.9 (32.2-72.7)	52.1 (33.4-75.1)	50.9 (32.2-72.7)	5.53
NT-proBNP, pg/mL	4850 (2127-11 100)	4614 (2170-9930)	4854 (2126-11 117)	11.32
Laboratory values and vitals at discharge, median (IQR)
Systolic blood pressure, mm Hg	123 (110-139)	124 (111-140)	123 (110-139)	6.90
Heart rate, beats per min	76 (67-87)	76 (68-88)	76 (67-87)	4.75
BMI^c	29.3 (24.5-35.9)	29.9 (25.1-36.8)	29.3 (24.5-35.9)	6.92
Potassium, mEq/L	4.0 (3.7-4.3)	4.0 (3.7-4.3)	4.0 (3.7-4.3)	3.83
Serum sodium, mEq/L	138 (136-140)	138 (135-140)	138 (136-140)	13.80
Serum creatinine, mg/dL	1.3 (1.0-1.9)	1.3 (1.0-1.8)	1.3 (1.0-1.9)	2.78
eGFR, mL/min/1.73 m²^d	51.2 (32.5-73.1)	52.1 (33.3-74.7)	51.2 (32.5-72.9)	4.26

Open in a new tab

Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; CRT-D, cardiac resynchronization therapy–implantable cardioverter-defibrillator; CRT-P, cardiac resynchronization therapy–pacemaker; eGFR, estimated glomerular filtration rate; HFpEF, heart failure with preserved ejection fraction; HFmrEF, heart failure with mildly reduced ejection fraction; HFrEF, heart failure with reduced ejection fraction; HMO, health maintenance organization; NT-proBNP, N-terminal pro–brain natriuretic peptide.

SI conversion factor: To convert creatinine to μmol/L, multiply by 88.4.

^{^a}

Absolute standardized difference greater than 10% indicates nonnegligible imbalance between groups.

^{^b}

Race and ethnicity were identified by self-report. The other race category includes American Indian or Alaska Native, Native Hawaiian or Pacific Islander, and unknown race.

^{^c}

Calculated as weight in kilograms divided by height in meters squared.

^{^d}

eGFR was calculated using the 2021 Chronic Kidney Disease–Epidemiology Collaboration equation.

In the subset of Medicare beneficiaries used in secondary end point analyses, 161 996 patients were included from 513 sites (Figure 1). Baseline comorbidities, vitals, and laboratory values were similar between patients at rural and urban hospitals except for systolic blood pressure, which tended to be higher for patients at rural hospitals (143 [124-165] mm Hg vs 140 [122-159] mm Hg) (eTable 5 in Supplement 1).

Hospital Characteristics

A total of 19 832 patients (2.6%) and 754 587 patients (97.4%) were hospitalized at 49 rural hospitals (8.6%) and 520 urban hospitals (91.4%), respectively. Rural hospitals had fewer beds than urban hospitals (median [IQR] beds, 107 [25-160] vs 288 [169-450]) (eTable 6 in Supplement 1). Rural hospitals were more frequently in the Midwest (20 [40.8%] vs 136 [26.2%]) and South (17 [34.7%] vs 169 [32.5%]) compared with urban hospitals. Additionally, rural hospitals were less likely to be teaching hospitals (13 [26.5%] vs 364 [70%]) and less likely to offer interventional cardiac catheterization (21 [43.8%] vs 429 [84.3%]) and heart transplant services (2 [4.2%] vs 131 [25.7%]) compared with urban hospitals. Similar differences were observed among rural and urban sites within the Medicare-linked cohort (eTable 7 in Supplement 1).

In-Hospital Outcomes and Quality Measures

In-Hospital Mortality and Length of Stay

Rates of in-hospital mortality were similar between patients admitted to rural and urban hospitals (460 of 19 832 [2.3%] vs 20 529 of 754 587 [2.7%]) (Table 2). There was no significant difference in the odds of in-hospital mortality between patients admitted to rural and urban in adjusted logistic regression models (adjusted odds ratio [aOR], 0.86; 95% CI, 0.70-1.07). There was a statistically significant interaction by EF subgroup, with a numerically higher risk of in-hospital mortality at rural hospitals for patients with EF of 41% to 49%.

Table 2. In-Hospital Outcomes, Quality Measures, and Guideline-Directed Medical Therapy (GDMT) Among Patients Hospitalized With Heart Failure (HF) at Rural vs Urban Get With The Guidelines–Heart Failure (GWTG-HF) Sites.

Outcome	No. (%)		Unadjusted		P value^b	Adjusted		P value^b
Outcome	Rural (n = 19 832)	Urban (n = 754 587)	Risk difference, % (95% CI)	OR (95% CI)^a	P value^b	Risk difference, % (95% CI)	OR (95% CI)^a^,^c	P value^b
In-hospital mortality
Overall	460 (2.3)	20 529 (2.7)	NA	0.85 (0.71-1.02)^d	.08	NA	0.86 (0.70-1.07)^d	.19
HFrEF (≤40%)	176 (2.3)	9608 (3.0)	NA	0.76 (0.60-0.97)	.03	NA	0.78 (0.59-1.03)	.08
HFmrEF (41%-49%)	53 (2.6)	1597 (2.2)	NA	1.18 (0.89-1.57)	.24	NA	1.22 (0.87-1.71)	.25
HFpEF (≥50%)	164 (1.8)	7499 (2.2)	NA	0.81 (0.63-1.04)	.09	NA	0.85 (0.65-1.11)	.23
Length of stay ≥4 d (all patients)
Overall	6892 (43.9)	329 622 (55.4)	NA	0.63 (0.56-0.71)^d	<.001	NA	0.75 (0.67-0.85)^d	<.001
HFrEF (≤40%)	2790 (44.0)	150 743 (57.0)	NA	0.59 (0.53-0.66)	<.001	NA	0.71 (0.63-0.79)	<.001
HFmrEF (41%-49%)	705 (43.3)	30 077 (53.4)	NA	0.67 (0.56-0.79)	<.001	NA	0.79 (0.68-0.93)	.004
HFpEF (≥50%)	3084 (45.0)	138 785 (54.8)	NA	0.67 (0.59-0.78)	<.001	NA	0.80 (0.70-0.92)	.002
Achievement of GWTG-HF quality measures among eligible patients
Measure LV function	17 945 (97.7)	702 136 (99.0)	−1.3 (−2.6 to 0.02)	0.44 (0.24-0.81)	.008	−1.1 (−2.4 to 0.2)	0.51 (0.29-0.91)	.02
Postdischarge appointment for patients with HF	12 073 (81.3)	458 243 (81.4)	−0.1 (−7.4 to 7.2)	0.99 (0.62-1.60)	.98	2.8 (−4.6 to 10.2)	1.21 (0.75-1.97)	.44
Follow-up visit within ≤7 d	9992 (57.2)	396 376 (59.4)	−2.2 (−9.6 to 5.2)	0.91 (0.67-1.24)	.56	−0.40 (−8.3 to 7.5)	0.98 (0.71-1.36)	.92
Smoking cessation	3126 (88.0)	111 597 (90.1)	−2.1 (−6.7 to 2.6)	0.81 (0.52-1.26)	.35	−2.1 (−6.8 to 2.7)	0.81 (0.51-1.28)	.37
CRT-D or CRT-P placed or prescribed at discharge	34 (18.3)	2480 (42.3)	−24.0 (−35.1 to −12.9)	0.31 (0.15-0.62)	.001	−13.5 (−24.1 to −2.9)	0.44 (0.22-0.92)	.03
DVT prophylaxis during hospitalization (all patients)	5025 (97.2)	250 648 (92.8)	4.4 (1.8 to 7.1)	2.70 (1.48-4.92)	.001	3.9 (−0.4 to 8.1)	2.60 (1.16-5.84)	.02
Influenza vaccination during influenza season	5999 (89.7)	230 830 (86.2)	3.5 (−1.4 to 8.4)	1.40 (0.85-2.30)	.19	1.5 (−3.5 to 6.5)	1.19 (0.70-2.00)	.52
Pneumococcal vaccination	10 926 (70.2)	429 356 (73.1)	−2.9 (−14.6 to 8.8)	0.87 (0.49-1.52)	.62	−3.7 (−15.5 to 8.1)	0.82 (0.45-1.51)	.53
Outpatient cardiac rehabilitation program referral	763 (4.3)	44 720 (6.8)	−2.5 (−5.5 to 0.6)	0.62 (0.31-1.25)	.18	−3.2 (−6.6 to 0.2)	0.55 (0.27-1.14)	.11
Outpatient HF disease management program referral	5244 (32.5)	298 532 (45.3)	−12.8 (−25.7 to 0.1)	0.58 (0.32-1.04)	.07	−2.8 (−15.6 to 10.0)	0.88 (0.49-1.60)	.69
GDMT prescribed at discharge among patients with HFrEF
ACEI/ARB	4866 (80.6)	196 713 (87.5)	−6.9 (−11.6 to −2.2)	0.60 (0.44-0.81)	.001	−3.7 (−7.8 to 0.4)	0.71 (0.53-0.96)	.03
ARNI^e	516 (13.3)	25 770 (17.8)	−4.6 (−9.3 to 0.1)	0.71 (0.47-1.05)	.09	−5.0 (−9.1 to −0.8)	0.68 (0.47-0.98)	.04
β-Blocker	7200 (94.4)	288 471 (96.0)	−1.6 (−3.1 to −0.1)	0.70 (0.52-0.94)	.02	−0.6 (−1.9 to 0.7)	0.82 (0.63-1.08)	.16
MRA	1921 (43.1)	90 544 (50.4)	−7.3 (−13.6 to −1.0)	0.75 (0.58-0.96)	.02	−1.9 (−7.9 to 4.0)	0.92 (0.70-1.20)	.53
Hydralazine nitrate	549 (30.9)	20 689 (28.7)	2.2 (−3.4 to 7.8)	1.11 (0.85-1.45)	.44	9.2 (2.4 to 15.9)	1.62 (1.14-2.30)	.007

Open in a new tab

Abbreviations: ACEI/ARB, angiotensin-converting enzyme inhibitors/angiotensin receptor blockers; ARNI, angiotensin receptor blocker–neprilysin inhibitor; CRT-D, cardiac resynchronization therapy–implantable cardioverter-defibrillator; CRT-P, cardiac resynchronization therapy–pacemaker; DVT, deep vein thrombosis; HFpEF, HF with preserved ejection fraction; HFmrEF, HF with mildly reduced ejection fraction; HFrEF, HF with reduced ejection fraction; LV, left ventricle; MRA, mineralocorticoid receptor antagonist; NA, not applicable; OR, odds ratio.

^{^a}

Patients hospitalized at urban hospitals were used as reference group for all regression models.

^{^b}

P values correspond with ORs.

^{^c}

Models were adjusted for demographic characteristics (age, sex, race and ethnicity, and insurance status), LV ejection fraction, medical history (hypertension, coronary artery disease, prior myocardial infarction, peripheral vascular disease, smoking, chronic obstructive pulmonary disease or asthma, atrial fibrillation or flutter, chronic kidney disease, cardiac resynchronization therapy, diabetes, hyperlipidemia, and stroke or transient ischemic attack), and hospital characteristics (geographic region, number of beds, and teaching status).

^{^d}

P for interaction < .01. Interaction P value tests if the association between rural/urban hospital and outcomes varied by ejection fraction phenotype.

^{^e}

Data for ARNI were reported starting in 2015.

Patients admitted to rural hospitals were significantly less likely to have a length of stay of 4 or more days (6892 [43.9%] vs 329 622 [55.4%]; aOR, 0.75; 95% CI, 0.67-0.85). The association between rural status and length of stay varied by EF, with lowest odds of length of stay of 4 or more days for rural vs urban hospitals among patients with EF of 40% or less (Table 2).

GWTG-HF Quality Measures

Patients admitted to rural hospitals were less likely to have measured left ventricular function (adjusted risk difference [aRD], −1.1%; 95% CI, −2.4 to 0.2; aOR, 0.51; 95% CI, 0.29-0.91) or CRT placed or prescribed at discharge (aRD, −13.5%; 95% CI, −24.1 to −2.9; aOR, 0.44; 95% CI, 0.22-0.92) but were more likely to receive deep vein thrombosis prophylaxis while hospitalized (aRD, 3.9%; 95% CI, −0.4 to 8.1; aOR, 2.60; 95% CI, 1.16-5.84) (Table 2; Figure 2). There were no statistically significant differences between rural and urban hospitals in adherence to other quality measures.

Figure 2. — Achievement of GWTG-HF quality measures (A) and GDMT prescription (B) among rural vs urban hospitals. Data for angiotensin receptor blocker–neprilysin inhibitor (ARNI) were reported starting in 2015. Influenza vaccination only applied to discharges during influenza season. ACEI/ARB indicates angiotensin-converting enzyme inhibitors/angiotensin receptor blockers; CRT, cardiac resynchronization therapy; DVT, deep vein thrombosis; LV, left ventricle; MRA, mineralocorticoid receptor antagonist.

^aAdjusted odds ratio P < .05.

GDMT Prescribed at Discharge

In unadjusted models, eligible patients from rural hospitals were less likely to be discharged taking ACEI/ARB (4866 [80.6%] vs 196 713 [87.5%]; OR, 0.60; 95% CI, 0.44-0.81), β-blocker (7200 [94.4%] vs 288 471 [96.0%]; OR, 0.70; 95% CI, 0.52-0.94), and MRA (1921 [43.1%] vs 90 544 [50.4%]; OR, 0.74; 95% CI, 0.58-0.96), but there were no significant differences in prescription of ARNI (516 [13.3%] vs 25 770 [17.8%]) or hydralazine/nitrate (549 [30.9%] vs 20 689 [28.7%]) (Table 2; Figure 2). However, after adjusting for patient and hospital characteristics, rural patients were significantly less likely to receive ARNI (aRD, −5.0%; 95% CI, −9.1 to −0.8; aOR, 0.68; 95% CI, 0.47-0.98) and more likely to receive hydralazine/nitrate (aRD, 9.2%; 95% CI, 2.4-15.9; aOR, 1.62; 95% CI, 1.14-2.30), whereas there were no significant differences in β-blocker (aOR, 0.82; 95% CI, 0.63-1.08) and MRA (aOR, 0.92; 95% CI, 0.70-1.20) prescription at discharge. Rural hospital admission remained significantly associated with lower odds of ACEI/ARB prescription in adjusted models (aRD, −3.7%; 95% CI, −7.8 to 0.4; aOR, 0.71; 95% CI, 0.53-0.96).

Postdischarge Outcomes

Among the subset of Medicare beneficiaries (n = 161 996), incidence rates for all three 30-day outcomes were high among patients discharged from rural and urban hospitals (Figure 3; eTable 8 in Supplement 1). In adjusted models, there were no significant differences between patients discharged from rural and urban hospitals for HF readmission (8.2% [95% CI, 7.4-9.1] vs 7.9% [95% CI, 7.7-8.0]; aHR, 1.03; 95% CI, 0.90-1.19), all-cause readmission (21.2% [95% CI, 20.0-22.5] vs 21.8% [95% CI, 21.6-22.0]; aHR, 0.97; 95% CI, 0.91-1.04), and all-cause mortality (6.3% [95% CI, 5.6-7.0] vs 5.6% [95% CI, 5.5-5.7]; aHR, 1.05; 95% CI, 0.91-1.21). There was a significant interaction between EF subgroup and rural status for all-cause mortality but not for HF or all-cause readmission (Figure 3). Results were consistent between unadjusted, partially adjusted, and fully adjusted models (eTable 8 in Supplement 1).

Figure 3. — Patients hospitalized at urban hospitals were used as the reference group for all regression models. Models were adjusted for demographic characteristics (age, sex, race and ethnicity, and insurance status), left ventricular ejection fraction, medical history (hypertension, coronary artery disease, prior myocardial infarction, peripheral vascular disease, smoking, chronic obstructive pulmonary disease/asthma, atrial fibrillation/flutter, chronic kidney disease, cardiac resynchronization therapy, diabetes, hyperlipidemia, and stroke or transient ischemic attack), and hospital characteristics (geographic region, number of beds, and teaching status). HF indicates heart failure; HFpEF, HF with preserved ejection fraction; HFmrEF, HF with mildly reduced ejection fraction; HFrEF, HF with reduced ejection fraction.

Discussion

In this retrospective cohort study of US patients hospitalized for HF, we found significant differences in multiple quality measures among patients discharged from rural vs urban hospitals. Although adjusted analyses supported most quality-of-care metrics as similar between rural and urban hospitals, eligible patients discharged from rural hospitals were less likely to be prescribed CRT and key elements of GDMT, including ACEI/ARB and ARNI therapy, even after adjustment for other hospital characteristics, such as geographic region, bed size, and teaching status. Despite these differences in HF care quality, rates of 30-day readmission and mortality were similarly high between the 2 groups, and patients at rural hospitals were more likely to have a shorter length of stay.

To our knowledge, this is the first study to examine patient-level and hospital-level differences in HF quality of care and outcomes in a contemporary nationwide cohort of patients hospitalized with HF at rural and urban hospitals. Compared with prior analyses that may rely on diagnosis codes alone, the GWTG-HF registry captures detailed clinical information, such as vital signs, laboratory values, medications, and device therapy. Likewise, GWTG-HF includes data on quantitative EF, which allowed assessment of differential associations between rural and urban hospital status and outcomes by EF phenotype. In addition, in contrast to prior studies more singularly focused on clinical outcomes, outcome data in the current analyses are complemented by comprehensive assessment of achievement of evidence-based quality measures.

Similar to our findings, a recent analysis of the National Inpatient Sample²³ found no significant difference between rural and urban hospitals for in-hospital mortality and demonstrated shorter length of stay for patients admitted for HF at rural hospitals. In-hospital mortality in both our analysis and the National Inpatient Sample²³ were 2% to 3% for both rural and urban hospitals, supporting generalizability of the GWTG-HF registry with other nationwide data across rural and urban hospitals.

However, the current results contrast with a prior study of Medicare claims that observed higher adjusted 30-day postdischarge all-cause mortality risk for Medicare beneficiaries admitted for HF to rural hospitals.²² In that study, relative risk of mortality following discharge from rural hospitals was similarly elevated in both unadjusted (HR, 1.13; 95% CI, 1.11-1.15) and adjusted (aHR, 1.15; 95% CI, 1.13-1.16) regression models.²² In the present study, while the unadjusted model showed a similar point estimate for 30-day mortality (HR, 1.12; 95% CI, 0.97-1.30), this was attenuated with multivariable adjustment (aHR, 1.05; 95% CI, 0.91-1.21). Thus, it is plausible that more rigorous risk adjustment in the current analysis with inclusion of clinical variables not otherwise available in a claims-based analysis could explain these differences between studies. Alternatively, differences between the 2 studies may be due to select hospitals being included in the GWTG-HF registry compared with the nationwide Medicare cohort in the prior study.

Our study is among the first nationwide studies to demonstrate reduced use of GDMT among rural patients with HF. One prior study in the Veterans Affairs system demonstrated that veterans in rural communities were less likely to receive target doses of β-blocker or ACEI/ARB therapy.²⁴ Similarly, we found lower use of ACEI/ARB or ARNI therapy among patients discharged from rural hospitals despite these patients having similar rates of chronic kidney disease and higher systolic blood pressure on admission than those discharged from urban hospitals. Although there was higher utilization of hydralazine-nitrates among rural hospitals, guidelines recommend hydralazine and isosorbide dinitrate for Black patients who remain symptomatic only after initiation of core GDMTs or who cannot tolerate first-line therapies, like ACEI/ARB or ARNI. In contrast, ACEI/ARB and ARNI medications are foundational for all eligible patients with HF with reduced EF and definitively proven to reduce all-cause mortality. Underuse of ACEI/ARB, ARNI, and CRT in the present study potentially contribute to higher longer-term HF mortality rates in rural communities.^1,3 These data support the need for evidence-based health policy and implementation strategies to reduce disparities in quality of care between rural and urban hospitals. Randomized trials of health policy and implementation strategies specifically targeted at rural hospitals should be considered to prevent potential unintended consequences from empirical adoption of untested interventions.

It is notable that despite differences in use of disease-modifying therapies, the current study found patients hospitalized at rural and urban hospitals to have similar risk-adjusted 30-day mortality and readmission. It is possible that the magnitude of differences in evidence-based therapy were not large enough to result in significant differences in outcomes or that the 30-day follow-up period was too brief to detect differences. It is also possible that other potential drivers of adverse HF outcomes in rural communities (eg, access to care, social determinants of health) may have lesser impact on outcomes soon after hospital discharge compared with effects on longer-term morbidity and mortality.

As a further consideration of underuse of GDMT at time of hospital discharge, prior studies have demonstrated minimal initiation and titration of medications for HF during longitudinal outpatient follow-up, including after hospitalization.^25,26 Larger gaps in quality of care at time of discharge from rural hospitals may support persistent longer-term underuse of proven therapies, with associated negative consequences on long-term morbidity and mortality. In addition, disproportionate financial pressure on rural hospitals to decrease length of stay may lead to less GDMT optimization prior to discharge. While shorter length of stay has been associated with increased rates of HF readmission,²⁷ shorter length of stay did not translate to excess 30-day readmission among rural hospitals in this study.

Limitations

This study has limitations. First, there is no consensus definition of rural vs urban hospitals. Although this analysis used a standard definition from the American Hospital Association, different definitions of rural and urban may lead to alternative results.²⁸ Second, this observational study cannot definitively determine cause-effect relationships. Third, formal interaction testing revealed statistically significant interaction between EF subgroups and in-hospital and postdischarge outcomes. However, these results should be interpreted with caution and may not necessarily reflect clinical meaningful differences by EF group. Fourth, this study examined specific elements of hospital-based HF care and did not investigate potential differences in outpatient HF diagnosis and treatment. Fifth, transfers were excluded from this analysis, which may bias length of stay outcomes given higher rates of transfer among rural hospitals.²⁹ Sixth, participation in the GWTG-HF registry is voluntary and may not be representative of rural or urban hospitals. For example, rural hospitals in GWTG-HF tended to be larger than most rural US hospitals,³⁰ and comorbidity burden among patients at rural GWTG-HF hospitals was similar to those at urban hospitals, which is different from some prior analyses of rural-urban disparities.^31,32 Although prior studies have supported generalizability to US practice, quality of care and outcomes in hospitals participating in GWTG-HF may not reflect care delivered at all rural or urban hospitals.³³ Additionally, given the potential for significant diversity among rural communities depending on geography, community demographic characteristics, and resources available to hospitals and patients, effective quality improvement initiatives may vary between rural hospitals. For example, critical access hospitals in remote rural areas likely require different support mechanisms compared with other rural hospitals.

Conclusions

In this study of US patients admitted to rural vs urban hospitals for HF, although most quality metrics were similar, patients at rural hospitals were less likely to receive multiple elements of guideline-directed HF care, such as CRT, ACEI/ARB, and ARNI therapies. Despite these differences in HF care, there were no significant differences between rural and urban hospitals regarding in-hospital mortality or 30-day postdischarge outcomes. Focused efforts to ensure consistent quality of HF care across rural and urban US hospitals may narrow rural-urban disparities in longer-term HF outcomes.

Supplement 1.

eMethods.

eTable 1. Baseline Characteristics of Patients Transferred to Another Acute Care Facility From Rural or Urban GWTG-HF Hospitals

eTable 2. Missing Variables of Baseline Characteristics for the Primary GWTG-HF and Secondary Medicare-Linked Cohorts

eTable 3. Comparison of American Hospital Association and Rural-Urban Commuting Area Code Classification Schemes for Defining Rural vs Urban GWTG-HF hospital Sites

eTable 4. Comparison of American Hospital Association and Rural-Urban Commuting Area Code Classification Schemes for Defining Rural vs Urban GWTG-HF Patients

eTable 5. Baseline Characteristics of Patients Hospitalized With Heart Failure at Rural and Urban GWTG-HF Sites Among the Medicare-Linked Cohort

eTable 6. Hospital Characteristics of Rural and Urban GWTG-HF Sites in Full Cohort

eTable 7. Hospital Characteristics for GWTG-HF Sites in the Medicare-Linked Cohort

eTable 8. Postdischarge Readmission and Mortality Outcomes at 30 Days Among Patients Admitted to Rural vs Urban Hospitals

Click here for additional data file.^{(267.6KB, pdf)}

Supplement 2.

Data Sharing Statement

Click here for additional data file.^{(13.3KB, pdf)}

References

1.Pierce JB, Shah NS, Petito LC, et al. Trends in heart failure-related cardiovascular mortality in rural versus urban United States counties, 2011-2018: a cross-sectional study. PLoS One. 2021;16(3):e0246813. doi: 10.1371/journal.pone.0246813 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Chang PP, Wruck LM, Shahar E, et al. Trends in hospitalizations and survival of acute decompensated heart failure in four US communities (2005-2014): ARIC Study Community Surveillance. Circulation. 2018;138(1):12-24. doi: 10.1161/CIRCULATIONAHA.117.027551 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Khan MS, Kumar P, Sreenivasan J, et al. Rural-urban differences in mortality from ischemic heart disease, heart failure, and stroke in the United States. Circ Cardiovasc Qual Outcomes. 2021;14(4):e007341. doi: 10.1161/CIRCOUTCOMES.120.007341 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.O’Hanlon CE, Kranz AM, DeYoreo M, Mahmud A, Damberg CL, Timbie J. Access, quality, and financial performance of rural hospitals following health system affiliation. Health Aff (Millwood). 2019;38(12):2095-2104. doi: 10.1377/hlthaff.2019.00918 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.McCarthy S, Moore D, Smedley WA, et al. Impact of rural hospital closures on health-care access. J Surg Res. 2021;258:170-178. doi: 10.1016/j.jss.2020.08.055 [DOI] [PubMed] [Google Scholar]
6.Joynt KE, Harris Y, Orav EJ, Jha AK. Quality of care and patient outcomes in critical access rural hospitals. JAMA. 2011;306(1):45-52. doi: 10.1001/jama.2011.902 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Dranove D, Garthwaite C, Ody C. Uncompensated care decreased at hospitals in Medicaid expansion states but not at hospitals in nonexpansion states. Health Aff (Millwood). 2016;35(8):1471-1479. doi: 10.1377/hlthaff.2015.1344 [DOI] [PubMed] [Google Scholar]
8.Germack HD, Kandrack R, Martsolf GR. When rural hospitals close, the physician workforce goes. Health Aff (Millwood). 2019;38(12):2086-2094. doi: 10.1377/hlthaff.2019.00916 [DOI] [PubMed] [Google Scholar]
9.Gong G, Phillips SG, Hudson C, Curti D, Philips BU. Higher US rural mortality rates linked to socioeconomic status, physician shortages, and lack of health insurance. Health Aff (Millwood). 2019;38(12):2003-2010. doi: 10.1377/hlthaff.2019.00722 [DOI] [PubMed] [Google Scholar]
10.Lindrooth RC, Perraillon MC, Hardy RY, Tung GJ. Understanding the relationship between Medicaid expansions and hospital closures. Health Aff (Millwood). 2018;37(1):111-120. doi: 10.1377/hlthaff.2017.0976 [DOI] [PubMed] [Google Scholar]
11.Bai G, Yehia F, Chen W, Anderson GF. Varying trends in the financial viability of US rural hospitals, 2011-17. Health Aff (Millwood). 2020;39(6):942-948. doi: 10.1377/hlthaff.2019.01545 [DOI] [PubMed] [Google Scholar]
12.Johnston KJ, Wen H, Joynt Maddox KE. Lack of access to specialists associated with mortality and preventable hospitalizations of rural Medicare beneficiaries. Health Aff (Millwood). 2019;38(12):1993-2002. doi: 10.1377/hlthaff.2019.00838 [DOI] [PubMed] [Google Scholar]
13.Harrington RA, Califf RM, Balamurugan A, et al. Call to action: rural health: a Presidential Advisory from the American Heart Association and American Stroke Association. Circulation. 2020;141(10):e615-e644. doi: 10.1161/CIR.0000000000000753 [DOI] [PubMed] [Google Scholar]
14.Hong Y, LaBresh KA. Overview of the American Heart Association “Get With The Guidelines” programs: coronary heart disease, stroke, and heart failure. Crit Pathw Cardiol. 2006;5(4):179-186. doi: 10.1097/01.hpc.0000243588.00012.79 [DOI] [PubMed] [Google Scholar]
15.Smaha LA; American Heart Association . The American Heart Association Get With The Guidelines Program. Am Heart J. 2004;148(5)(suppl):S46-S48. doi: 10.1016/j.ahj.2004.09.015 [DOI] [PubMed] [Google Scholar]
16.Hammill BG, Hernandez AF, Peterson ED, Fonarow GC, Schulman KA, Curtis LH. Linking inpatient clinical registry data to Medicare claims data using indirect identifiers. Am Heart J. 2009;157(6):995-1000. doi: 10.1016/j.ahj.2009.04.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Ambardekar AV, Fonarow GC, Dai D, et al. ; Get With The Guidelines Steering Committee and Hospitals . Quality of care and in-hospital outcomes in patients with coronary heart disease in rural and urban hospitals (from Get With The Guidelines-Coronary Artery Disease Program). Am J Cardiol. 2010;105(2):139-143. doi: 10.1016/j.amjcard.2009.09.003 [DOI] [PubMed] [Google Scholar]
18.Bahiru E, Ziaeian B, Moucheraud C, et al. Association of dual eligibility for Medicare and Medicaid with heart failure quality and outcomes among Get With The Guidelines–Heart Failure hospitals. JAMA Cardiol. 2021;6(7):791-800. doi: 10.1001/jamacardio.2021.0611 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Wilcock AD, Zachrison KS, Schwamm LH, Uscher-Pines L, Zubizarreta JR, Mehrotra A. Trends among rural and urban Medicare beneficiaries in care delivery and outcomes for acute stroke and transient ischemic attacks, 2008-2017. JAMA Neurol. 2020;77(7):863-871. doi: 10.1001/jamaneurol.2020.0770 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Loccoh E, Joynt Maddox KE, Xu J, et al. Rural-urban disparities in all-cause mortality among low-income Medicare beneficiaries, 2004-17. Health Aff (Millwood). 2021;40(2):289-296. doi: 10.1377/hlthaff.2020.00420 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Redmond N, Richman J, Gamboa CM, et al. Perceived stress is associated with incident coronary heart disease and all-cause mortality in low- but not high-income participants in the Reasons for Geographic And Racial Differences in Stroke study. J Am Heart Assoc. 2013;2(6):e000447. doi: 10.1161/JAHA.113.000447 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Loccoh EC, Joynt Maddox KE, Wang Y, Kazi DS, Yeh RW, Wadhera RK. Rural-urban disparities in outcomes of myocardial infarction, heart failure, and stroke in the United States. J Am Coll Cardiol. 2022;79(3):267-279. doi: 10.1016/j.jacc.2021.10.045 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Minhas AMK, Sheikh AB, Ijaz SH, et al. Rural-urban disparities in heart failure and acute myocardial infarction hospitalizations. Am J Cardiol. 2022;175:164-169. doi: 10.1016/j.amjcard.2022.04.014 [DOI] [PubMed] [Google Scholar]
24.Witting C, Zheng J, Tisdale RL, et al. Treatment differences in medical therapy for heart failure with reduced ejection fraction between sociodemographic groups. JACC Heart Fail. 2022;S2213-1779(22)00573-X. doi: 10.1016/j.jchf.2022.08.023 [DOI] [PMC free article] [PubMed]
25.Greene SJ, Fonarow GC, DeVore AD, et al. Titration of medical therapy for heart failure with reduced ejection fraction. J Am Coll Cardiol. 2019;73(19):2365-2383. doi: 10.1016/j.jacc.2019.02.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Rao VN, Murray E, Butler J, et al. In-hospital initiation of sodium-glucose cotransporter-2 inhibitors for heart failure with reduced ejection fraction. J Am Coll Cardiol. 2021;78(20):2004-2012. doi: 10.1016/j.jacc.2021.08.064 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Khan H, Greene SJ, Fonarow GC, et al. ; EVEREST Trial Investigators . Length of hospital stay and 30-day readmission following heart failure hospitalization: insights from the EVEREST trial. Eur J Heart Fail. 2015;17(10):1022-1031. doi: 10.1002/ejhf.282 [DOI] [PubMed] [Google Scholar]
28.US Department of Agriculture . Rural classifications. Accessed December 16, 2022. https://www.ers.usda.gov/topics/rural-economy-population/rural-classifications/
29.Bennett KJ, Probst JC, Bullard JC, Crouch E. The importance of rural hospitals: transfers and 30-day readmissions among rural residents and patients presenting at rural hospitals. Popul Health Manag. 2019;22(2):120-126. doi: 10.1089/pop.2018.0050 [DOI] [PubMed] [Google Scholar]
30.American Hospital Association . Fast Facts: U.S. rural hospitals infographic. Accessed January 26, 2023. https://www.aha.org/infographics/2021-05-24-fast-facts-us-rural-hospitals-infographic
31.Hillerson D, Li S, Misumida N, et al. Characteristics, process metrics, and outcomes among patients with ST-elevation myocardial infarction in rural vs urban areas in the US: a report from the US National Cardiovascular Data Registry. JAMA Cardiol. 2022;7(10):1016-1024. doi: 10.1001/jamacardio.2022.2774 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Turecamo SE, Xu M, Dixon D, et al. Association of rurality with risk of heart failure. JAMA Cardiol. Published online January 25, 2023. doi: 10.1001/jamacardio.2022.5211 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Curtis LH, Greiner MA, Hammill BG, et al. Representativeness of a national heart failure quality-of-care registry: comparison of OPTIMIZE-HF and non-OPTIMIZE-HF Medicare patients. Circ Cardiovasc Qual Outcomes. 2009;2(4):377-384. doi: 10.1161/CIRCOUTCOMES.108.822692 [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