Long-term Survival After Heart Failure Discharge Tied to Hospital Performance

March 21, 2018

ORLANDO — A new analysis of patients with heart failure (HF) discharged from hospitals participating in a large quality-improvement program bolsters the case for preferring mortality over rehospitalization at 30 days as the better performance metric, say investigators.

Survival at 1 to 5 years was better for such patients cared for at Get With The Guidelines (GWTG) Heart Failure centers that had lower 30-day post–HF-discharge risk-standardized mortality rates (RSMR), they found.

Centers in the lowest quartile for 30-day RSMR tended to be those most adherent to many evidence-based HF quality measures, including assurance of follow-up within a week of discharge and prescription of device therapies.

An analysis limited to patients who survived the first 30 postdischarge days produced similar findings.

"The survival advantage at centers with low 30-day risk-standardized mortality rates continues to accrue beyond the first 30 days and persists long-term," said Ambarish Pandey, MD, University of Texas Southwestern Medical Center, Dallas.

"Taken together, these findings suggest that the metric of 30-day RSMR may be useful to incentivize quality care and improve long-term outcomes in patients hospitalized with heart failure," said Pandey, presenting the study here at the American College of Cardiology (ACC) 2018 Annual Scientific Sessions.

The findings underscore that "hospital performance does matter," Gregg C Fonarow, MD, University of California Los Angeles told theheart.org | Medscape Cardiology. He is senior author on the GWTG Heart Failure registry analysis, published March 12 in JAMA Cardiology to coincide with its presentation at the ACC meeting.

"We saw meaningful differences in the types of evidence-based therapies provided by the different quartiles of hospitals grouped by their performance on the 30-day risk-standardized mortality metric," he said.

That patients discharged from the better-performing centers also had a survival benefit at 1, 3, and 5 years in the analysis shows that 30-day RSMR "is an actionable metric that really makes a large and substantial difference in meaningful life for patients hospitalized with heart failure," Fonarow said.

The current findings make a good case that adjusted 30-day mortality after a heart-failure hospitalization "does a pretty good job of predicting both short-term and long-term outcomes," according to Karen Joynt Maddox, MD, Washington University School of Medicine, St Louis, Missouri, who isn't connected to the study.

"That short-term mortality correlates with knowing that the hospitals are doing more of the right things — guideline-based therapy — in the inpatient setting, and that it carries over into long-term effects, I think is very reassuring," she said in an interview.

Mortality vs Rehospitalization

The findings complement assertions in a recent report from the same GWTG registry, from many of the same authors, that the 30-day RSMR and 30-day risk-standardized readmission rate (RSRR) metrics used in the Hospital Readmissions Reduction Program may be somewhat at odds with each other.

The Centers for Medicare & Medicaid Services (CMS) uses the metrics, especially RSRR, to guide penalty adjustments to diagnosis-related group reimbursements as a way to incentivize  hospital performance, to the chagrin of many HF specialists.

"The current financial incentive program used by CMS is heavily skewed in favor of 30-day RSRR-based hospital performance, with up to 15-fold greater penalties for excess readmission rates (3%) than excess mortality rates (0.2%) in financial year 2016," according to the current published report.

"As a result, there is an increasing drive to invest hospital resources in programs focused on reducing readmissions. While such programs may have led to a decline in readmission rates, there has been a significant and steady increase in risk adjusted mortality rates in HF over the same period."

The current analysis argues in favor of RSMR as the predominant target, as opposed to the 30-day readmission metric, "where you find an inverse relationship, if anything, to mortality, you don't find any of the traditional process of care measures judging quality to be associated with it, and no relationship or even an inverse relationship with 1- and 3-year mortality," Fonarow said.

"A decade's worth of gradual decline in 30-day and 1-year mortality" was "completely reversed" after CMS started using the metric to determine financial penalties, Fonarow said. That "does make a pretty compelling case for why it needs to reform their penalty program and put greater weight on meaningful metrics, such as 30-day risk-standardized mortality."

Although Joynt Maddox questions whether efforts to rein in 30-day HF readmission rates necessarily raise postdischarge mortality in patients with HF, "I'm not so sure readmission is a great stand-alone hospital quality measure."

The current analysis "tells you more about hospital quality than most of the readmissions measures do," she said. "Readmission fundamentally happens outside of the hospital, and when you look at the predictive power of risk-adjusted models, the mortality ones tend to be much better. We understand why people die better than we understand why they're readmitted."

Short-term Mortality Predicts Long-term Outcomes

The current analysis looked at 106,304 patients with HF, aged 65 years or older, at 317 GWTG hospitals from 2005 to 2013; about 54% were women. The centers' 30-day RSMR ranged from 8.6% in the first quartile to 10.7% in the fourth quartile.

Unadjusted mortality at 1, 3, and 5 years after the HF discharge went up in graded fashion with increasing RSMR quartile for the respective centers, in an analysis covering all patients and in another limited to patients who survived at least 30 days after discharge.

Table 1. Unadjusted Mortality After HF Hospitalization by 30-Day RSMR Quartile, Overall and Among 30-Day Survivors

Endpoints 30-Day Survivors (%) Overall (%)
Q1 Q4 Q1 Q4
1-y mortality 30.2 32.9 35.3 41.1
3-y mortality 57.8 60.9 60.9 65.7
5-y mortality 73.7 76.8 75.6 79.6
Q1 = quartile of centers with the lowest 30-day RSMR; Q4 = quartile of centers with the highest 30-day RSMR.


In adjusted analysis, a similar pattern of greater mortality hazard ratios (HRs) with increasing hospital 30-day RSMR quartile emerged at 1, 3, and 5 years.

Table 2. Adjusteda  HRs (95% CIs) for Mortality by 30-Day RSMR Quartile Among Survivors 30 Days After HF Discharge

Endpoints 1-Year Mortality 3-Year Mortality 5-Year Mortality
Q2 vs Q1 1.05 (1.00 - 1.11) 1.04 (1.00 - 1.09) 1.04 (1.00 - 1.09)
Q3 vs Q1 1.06 (1.02 - 1.11) 1.05 (1.01 - 1.09) 1.05 (1.01 - 1.08)
Q4 vs Q1 1.17 (1.12 - 1.23) 1.14 (1.10 - 1.18) 1.14 (1.10 - 1.18)
Q1 = quartile of centers with the lowest 30-day RSMR; Q4 = quartile of centers with the highest 30-day RSMR.

aAdjusted for demographics, socioeconomic status, medical history, baseline left ventricular ejection fraction, vital signs, laboratory test results, medications, and number of all-cause and HF hospitalizations within 6 months before index hospitalization.


The results for patients at Q4 hospitals vs Q1 hospitals were similar in the adjusted analysis across the entire cohort, including those who died within 30 days after discharge, with greater long-term mortality HRs.

  • 1-year mortality: HR, 1.31 (95% CI, 1.26 - 1.36);

  • 3-year mortality: HR, 1.23 (95% CI, 1.19 - 1.27); and

  • 5-year mortality: HR, 1.22 (95% CI, 1.18 - 1.26).

In a secondary finding of the study, the associations between 30-day RSMR quartile and 3- and 5-year mortality were strongest for patients who had been hospitalized with HF with reduced ejection fraction (HFrEF) and attenuated for HF with preserved ejection fraction.

Fonarow said that would certainly be expected given that for HFrEF, "we have far more in the way of evidence-based therapies that can impact survival directly."

The GWTH-Heart Failure program is sponsored by the American Heart Association and has previously received funding from Medtronic, GlaxoSmithKline, Ortho-McNeil Pharmaceutical, and the American Heart Association Pharmaceutical Roundtable. Pandey had no disclosures. Fonarow discloses receiving research funding from the National Institutes of Health and consulting for Amgen, Bayer, Janssen, Novartis, Medtronic, and St Jude Medical. Joynt Maddox has recently reported she has no relevant financial relationships.

American College of Cardiology (ACC) 2018 Annual Scientific Session. Session 412-08. Presented March 12, 2018.

JAMA Cardiol. Published online March 22, 2018. Abstract

Follow Steve Stiles on Twitter: @SteveStiles2. For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: