Association of Rankings With Cardiovascular Outcomes at Top-Ranked Hospitals vs Nonranked Hospitals in the United States

Key Points

Question  Do US News & World Report top-ranked hospitals for cardiovascular care have better outcomes than nonranked hospitals?

Findings  In this cross-sectional study, US News & World Report top-ranked hospitals for cardiovascular care had lower 30-day mortality rates for acute myocardial infarction, heart failure, and coronary artery bypass grafting and higher patient satisfaction ratings compared with nonranked hospitals. However, 30-day readmission rates were either similar (for acute myocardial infarction and coronary artery bypass grafting) or higher (for heart failure) at top-ranked hospitals compared with nonranked hospitals.

Meaning  The observed discrepancy between readmissions and other performance measures raises concern that readmissions may not be an adequate metric of hospital care quality.

Importance  The US News & World Report (USNWR) identifies the “Best Hospitals” for “Cardiology and Heart Surgery.” These rankings may have significant influence on patients and hospitals.

Objective  To determine whether USNWR top-ranked hospitals perform better than nonranked hospitals on mortality rates and readmission measures as well as patient satisfaction.

Design, Setting, and Participants  This national retrospective study evaluated outcomes at 3552 US hospitals from 2014 to 2017.

Exposures  US News & World Report 2018 to 2019 Cardiology and Heart Surgery rankings (top-ranked vs nonranked hospitals).

Main Outcomes and Measures  Hospital-level 30-day risk-standardized mortality and readmission rates for Medicare fee-for-service beneficiaries age 65 years or older hospitalized for 3 cardiovascular conditions: acute myocardial infarction (AMI), heart failure (HF), and coronary artery bypass grafting (CABG) as well as Hospital Consumer Assessment of Healthcare Providers and Systems patient satisfaction star ratings obtained from publicly available Centers for Medicaid and Medicare Services data.

Results  Thirty-day mortality rates at top-ranked hospitals (n = 50), compared with nonranked hospitals (n = 3502), were lower for AMI (11.9% vs 13.2%, P < .001), HF (9.5% vs 11.9%; P < .001), and CABG (2.3%vs 3.3%; P < .001). Thirty-day readmission rates at the top-ranked hospitals (n = 50) when compared with nonranked hospitals (n = 2841) were similar for AMI (16.7% vs 16.5%; P = .64) and CABG (14.1% vs 13.7%; P = .15) but higher for HF (21.0% vs 19.2%; P < .001), Finally, patient satisfaction was higher at top-ranked hospitals (n = 50) compared with nonranked hospitals (n = 3412) (3.9 vs 3.3; P < .001).

Conclusions and Relevance  We found that USNWR top-ranked hospitals for cardiovascular care had lower 30-day mortality rates for AMI, HF, and CABG and higher patient satisfaction ratings compared with nonranked hospitals. However, 30-day readmission rates were either similar (for AMI and CABG) or higher (for HF) at top-ranked compared with nonranked hospitals. This discrepancy between readmissions and other performance measures raises concern that readmissions may not be an adequate metric of hospital care quality.

Hospital rankings that evaluate and publicly compare hospital performance for cardiovascular care have become increasingly common as part of a movement to enhance transparency in health care. The US News & World Report (USNWR) is one of the most influential rankings and identifies the “Best Hospitals” for “Cardiology and Heart Surgery” among other categories.¹ Hospital rankings should guide patients to high-quality health care, which can be measured by outcomes that matter most to patients, such as mortality rates and readmissions. However, the USNWR hospital-ranking methods also emphasize metrics, such as hospital reputation, that may not reflect care quality. While prior research has shown that top-ranked hospitals may perform better than nonranked hospitals, these studies were for a specific subset of patients, such as those undergoing percutaneous coronary intervention,¹ or were performed nearly a decade ago.^2-4 These findings may no longer hold true, as hospitals have increasingly been required to participate in national quality improvement initiatives, such as pay-for-performance, public reporting, and readmission reduction programs, and because the USNWR methods have evolved in recent years.^5,6

Understanding whether USNWR top-ranked hospitals perform better than nonranked hospitals on mortality rates and readmission measures is important given the significant influence that these rankings may have on patients and hospitals. This study examined whether USNWR top-ranked hospitals compared with nonranked hospitals have better outcomes as assessed by mortality rates, readmission rates, and patient satisfaction for 3 cardiovascular conditions. This study also evaluated, among hospitals that were ranked, whether a higher ranking was associated with better outcomes.

We obtained hospital-level risk-standardized 30-day mortality rates and readmission rates for acute myocardial infarction (AMI), heart failure (HF), and coronary artery bypass grafting (CABG) from the Centers for Medicare and Medicaid Services (CMS) Hospital Compare Data set.⁷ Risk standardization of mortality and readmission rates is performed by CMS to account for variation in patient case mix indices among hospitals.⁸ We also obtained Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) star ratings, which are a composite measure of patient satisfaction for each hospital (assigned 1-5 stars).⁷

Hospitals were stratified into 2 groups: top-ranked hospitals (defined as those in the top 50 of the 2018-2019 USNWR “Best Hospitals” for “Cardiology and Heart Surgery”) and nonranked hospitals (defined as those not identified as top-50 hospitals). We then compared risk-standardized 30-day mortality and readmission rates between top-ranked and nonranked hospitals for the 3 cardiovascular conditions as well as HCAPHS star ratings at these hospitals using a linear regression model.

Additionally, we divided the top 50 ranked hospitals into quintiles by hospital rank (1-10, 11-20, 21-30, 31-40, and 41-50). We then compared the aforementioned outcome measures by hospital quintile using a linear regression model to evaluate if, among top-ranked hospitals, a higher rank was associated with better outcomes.

Thirty-day mortality rates at top-ranked hospitals (n = 50) compared with nonranked hospitals (n = 3502) were lower for AMI (11.9% vs 13.2%; P < .001), HF (9.5% vs 11.9%; P < .001), and CABG (2.3% vs 3.3%; P < .001) (Figure 1A). In addition, 30-day readmission rates (Figure 1) at top-ranked hospitals (n = 50) compared with nonranked hospitals (n = 2841) were similar for AMI (16.7% vs 16.5%; P = .64) and CABG (14.1% vs 13.7%; P = .15), but higher for HF (21.0% vs 19.2%; P < .001). Finally, patient satisfaction was higher at top-ranked hospitals (n = 50) compared with nonranked hospitals (n = 3412) (3.9 vs 3.3; P < .001) (Figure 2).

Among the top 50 ranked hospitals, 30-day mortality rates at the top-ranked quintile (1-10) vs the lowest-ranked quintile (41-50) were lower for 2 of the 3 conditions (AMI, 10.5% vs 12.5%; P < .001; HF, 8.2% vs 10.7%; P = .001; CABG, 2.0% vs 2.4%; P = .23) (Table). Thirty-day readmission rates at top-ranked vs lowest-ranked quintile hospitals did not significantly differ for any of the 3 cardiovascular conditions (AMI, 17.4% vs 16.2%; P = .11; HF, 21.0% vs 20.5%; P = .73; CABG, 13.8% vs 13.4%; P = .49). Patient satisfaction, as measured by the HCAPHS star rating, was higher at the top-ranked quintile vs the lowest-ranked quintile hospitals (4.2 vs 3.8; P = .01).

Our findings demonstrate that 30-day risk-standardized mortality rates for CABG, AMI, and HF at top-ranked USNWR cardiology hospitals are lower than at nonranked hospitals. In contrast, 30-day risk-standardized readmission rates for AMI and CABG are similar at top-ranked compared with nonranked hospitals, while HF readmissions are significantly higher. Finally, we also found that patient satisfaction scores were higher at top-ranked USNWR cardiology hospitals compared with nonranked hospitals.

Although the USNWR methods have changed in recent years, top-ranked cardiology hospitals have lower mortality rates and higher patient satisfaction scores compared with nonranked hospitals. This likely reflects the fact that 37.5% of USNWR rankings are based on raw mortality data that are obtained from the CMS, although the approach to risk adjustment is different than that of the CMS’ risk standardization.⁹ However, readmission rates are similar for AMI and CABG at top-ranked vs nonranked hospitals and are higher for HF at top-ranked hospitals. This disconnect between mortality rates and readmission outcomes at top-ranked hospitals compared with nonranked hospitals highlights the ongoing uncertainty as to whether readmissions rates are an adequate surrogate for quality of care,¹⁰ particularly for cardiovascular conditions, such as HF.^6,11

In recent years, financial incentives for hospitals to reduce readmissions following the enactment of the Hospital Readmissions Reduction Program have been 10-fold to 15-fold greater than incentives to improve mortality rates and have resulted in significant declines in cardiovascular readmissions.¹² Our finding that top-ranked hospitals have lower mortality rates than nonranked hospitals but have generally similar readmission rates might reflect these incentives. It is possible that top-ranked and nonranked hospitals have focused substantial resources on reducing readmissions rather than mortality rates given the financial push of the Hospital Readmission Reduction Program, which resulted in generally similar readmission rates but disparate mortality rates between these hospital groups.

Our study has limitations. We chose to use USNWR hospital rankings, which use CMS data to determine the rankings. Other hospital ranking systems use all-payer data, which capture non-Medicare patients and result in different hospital rankings than USNWR.¹³

We found that USNWR top-ranked hospitals for cardiovascular care had lower 30-day mortality rates for AMI, HF, and CABG and higher patient satisfaction ratings compared with nonranked hospitals. However, 30-day readmission rates were either similar (for AMI and CABG) or higher (for HF) at top-ranked compared with nonranked hospitals. This discrepancy between readmissions and other performance measures raises concern that readmissions may not be an adequate metric of hospital care quality.

Accepted for Publication: October 8, 2018.

Corresponding Author: Deepak L. Bhatt, MD, MPH, Brigham and Women's Hospital Heart & Vascular Center, Harvard Medical School, 75 Francis St, Boston, MA 02115 (dlbhattmd@post.harvard.edu).

Correction: This article was corrected online January 2, 2019, for errors in the corresponding author’s email address, study dates in the Abstract, and formatting in the Table.

Published Online: November 28, 2018. doi:10.1001/jamacardio.2018.3951

Author Contributions: Dr Wang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Wang, Wadhera, Bhatt.

Acquisition, analysis, or interpretation of data: Wang, Bhatt.

Drafting of the manuscript: Wang, Wadhera.

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

Statistical analysis: Wang, Wadhera.

Study supervision: Bhatt.

Conflict of Interest Disclosures: Dr Wadhera reported prior personal fees from Regeneron outside the submitted work. Dr Bhatt reported grants from Amarin, AstraZeneca, Bristol-Myers Squibb, Roche, Eisai, Ethicon, Medtronic, Sanofi Aventis, Pfizer, Forest Laboratories/AstraZeneca, Ischemix, Amgen, Lilly, Chiesi, Ironwood, Abbott, Regeneron, PhaseBio, Idorsia, Synaptic, and The Medicines Company; unfunded research support from FlowCo, PLx Pharma, Novo Nordisk, Takeda, and Merck; serving on the advisory boards of Medscape Cardiology, Regado Biosciences, Cardax; serving on the board of directors of Boston VA Research Institute; being the deputy editor of Clinical Cardiology; being a site coinvestigator for Biotronik, Boston Scientific, Svelte, and St Jude Medical (now Abbott); receiving honoraria from the American College of Cardiology; serving on clinical trial committees funded by Bayer and Boehringer Ingelheim; personal fees from Duke Clinical Research Institute, Mayo Clinic, Population Health Research Institute, Belvoir Publications, Slack Publications, WebMD, Elsevier, Society of Cardiovascular Patient Care, HMP Global, Harvard Clinical Research Institute (now Baim Institute for Clinical Research), Journal of the American College of Cardiology, Cleveland Clinic, Mount Sinai School of Medicine, and TobeSoft; nonfinancial support from American College of Cardiology, American Heart Association, and the Society of Cardiovascular Patient Care. No other disclosures were reported.