Readmission-Reduction Program Linked to Higher Heart-Failure Mortality

Patrice Wendling

November 21, 2017

ANAHEIM, CA — Controversial new research suggests the Centers for Medicare & Medicaid Services (CMS) Hospital Readmission Reduction Program (HRRP) may have slowed the revolving door for Medicare patients hospitalized with heart failure (HF), but the lower rate of readmissions may have come at a price of higher 30-day and 1-year mortality[1].

Dr Gregg C Fonarow

"It was really the worst of our fears realized," particularly as mortality among these patients was previously going down, study author Dr Gregg Fonarow (University of California Los Angeles) told | Medscape Cardiology.

"You could say, 'what's a 1% or 2% increase in 30-day mortality? That seems small,' but actually when applied to just the Medicare beneficiaries being hospitalized, it's a very large number in absolute terms of patients that may have been harmed by this untested policy that was launched without close scrutiny and monitoring."

At the national level, this could be as many as 10,000 extra lives being lost per year in just the first 30 days, Fonarow suggests. There's also the possibility of a spillover effect to non-Medicare patients and that the observed mortality increase is continuing today, past the 2-year lag in the data.

Not everyone is convinced, however. Social media erupted in a flurry of sharp responses when the data were presented at the American Heart Association (AHA) 2017 Scientific Sessions and simultaneously published in JAMA Cardiology, with first author Dr Ankur Gupta (Brigham and Women's Hospital Heart & Vascular Center and Harvard Medical School, Boston, MA).

At the center of the argument is a recent Yale University study that looked at within-hospital trends in HF readmissions and mortality from more than 4000 hospitals, in which reductions in 30-day readmissions were weakly but significantly correlated with lower 30-day mortality[2].

"This is news that has made it to hospital administrators," American College of Cardiology (ACC) president Dr Mary Norine Walsh (St Vincent Medical Group, Indianapolis, IN) said in an interview.

"This has been a very loud study in that regard. I've already been emailed by my own institution: 'Have you seen these data?' My answer is everybody has to know their own mortality statistics to make sure that this hasn't happened in their hospital and to analyze why—what strategies did we put in place, what have we changed, and if we have seen an increase in mortality, how are we going to change?" she said.

The present study examined data from 115,245 fee-for-service Medicare patients at 416 US hospitals in the AHA Get with the Guidelines (GWTG)-Heart Failure registry during three phases: before HRRP implementation (January 1, 2006 to March 31, 2010); during HRRP implementation (April 1, 2010 to September 30, 2012); and during the HRRP penalty phase requiring the CMS to dock payments to hospitals with excess readmissions (October 1, 2012 to study end on December 30, 2014).

The 30-day risk-adjusted readmission rate fell from 20.0% before HRRP to 18.4% in the penalty phase (hazard ratio 0.91, 95% CI 0.87–0.95), while 30-day risk-adjusted mortality increased over the same period from 7.2% to 8.6% (HR 1.18, 95% CI 1.10–1.27).

Similarly, 1-year risk-adjusted readmissions declined over the same period from 57.2% to 56.3% (HR 0.9, 95% CI 0.89–0.96; P<0.001), while 1-year risk-adjusted mortality rose from 31.3% before HRRP to 36.3% in the penalty phase (HR 1.10, 95% CI 1.06–1.14; all comparisons P<0.001).

"In their data, the decline in readmission just about mirrored the increase in mortality," Dr Harlan Krumholz (Yale University School of Medicine, New Haven, CT), who coauthored the Yale paper, told | Medscape Cardiology.

"So how does that work? Is it that every additional person not admitted after the penalty died? Is that really plausible? It just doesn't pass the sniff test," Krumholz said.

He noted that the GWTG registry is not a representative sample, the investigators didn't compare the same hospitals over time, and that the readmission declines in the ecological study aren't consistent with several other studies that found such declines largely occurred between 2010 and 2012 and not after 2013.

"If the problem is that some doctors are putting their hospitals' interest before their patients—I don't believe it, but that must be the premise—isn't that the issue to attack and identify where and how that is happening and fix it, rather than focus on the measure?" Krumholz added.

Fonarow and his colleagues suggest that pressure to reduce readmissions and costly penalties has created an incentive to delay necessary hospital readmissions beyond the 30-day window; funnel patients into early hospice care; and to use observation stays even when admission is clinically indicated, a practice that leaves patients ineligible for home healthcare, nursing, and other Medicare benefits that follow an inpatient HF hospitalization.

"It's something that's certainly amenable to a policy change," Fonarow said. "If you're going to create a financial penalty about inpatient and outpatient status then you need to remove the barrier to patients getting the care they need after that observation stay."

Also ripe for change is the 15-fold gap between the maximum penalties of up to 3% for readmissions vs only 0.2% for mortality.

"That needs to be reversed. The Medicare policies need to value Medicare beneficiaries' lives. Patients do want to be at home but not at the price of dying," he added.

Fonarow acknowledged that they cannot fully assign causation to the HRRP but said recent national data also show similar mortality trends[3].

Heart Failure Society of America (HFSA) president Dr Christopher O'Connor (Inova Heart & Vascular Institute, Falls Church, VA) said the competing risk of early mortality could be an issue as well as changing demographics and patient populations over the study period.

"But the 1-year mortality is a more durable end point, and I think showing that that end point was also increased is one that makes me nervous about the policy," he said in an interview. "The authors have done a really good job of trying to adjust for the covariates. If there's a covariate that explains this, we need to hear it loud and clear."

Notably there was no difference in the findings between teaching and nonteaching hospitals, despite concerns that financial penalties are disproportionately levied against academic medical centers and safety-net hospitals that care for sicker patients.

Interventional cardiologist Dr Robert W Yeh (Beth Israel Deaconess Medical Center, Boston, MA) said he's not entirely convinced hospital readmissions and the rising mortality rate are related and suggested that crude mortality rates appeared to be going up for HF even before implementation of the HRRP.

On the other hand, he noted that years after policies for public reporting of PCI-related mortality were initiated, there is now robust evidence of unintended consequences.

"You can tell from the reaction, the public reaction, especially among the heart-failure community, is, 'We knew it, we knew this was the case, and we know this based on how we treat our patients.' To me that is almost the more interesting and compelling part of what this research means," Yeh said. "The research has provoked a lot of emotion and when it provokes emotion, you know there's something behind that. So that's what makes me concerned, actually."

Several experts noted that the focus on excess readmissions has prompted positive changes such as better communication and coordination of care, as well as putting in place early-access clinics, transition-of-care clinics, patient navigators, and postdischarge coordinators. Still, O'Connor observed that the course of heart failure isn't just 30 days but runs months longer.

"The bundled-care initiatives are looking at 90-day event rates, but it's very clear that it gets harder and harder," he said. "Maybe this is the direction we should be pushing hospitals, to be more integrated healthcare systems that integrate outpatient and inpatient care, but most of the hospitals aren't suited to do that in this country."

Asked whether a 90-day metric would be better for heart-failure patients, Dr Mariell Jessup (LeDucq Foundation, Boston, MA) told | Medscape Cardiology, "I don't think we know enough about the trajectory of disease. People can come back in the hospital because they don't take their medicines or because they don't have an environment where they're going to do well, or it may be their heart failure is just terrible."

She added, "I think it's a really legitimate question, and I think it's important to explore it further and understand what we've learned over the past few years because people have learned a lot, but I do think that it requires an issue of saying was this the right metric or is it time to change the metric."

The ACC's Walsh said she's not sure what the correct metric is, but that the CMS and patients need to be involved in the discussion regarding the different data sets and any potential policy changes.

"Readmission is often stated to be a patient-centered metric, and it isn't always," she said. "Most patients don't want to be in the hospital, of course, but I think sometimes going back in the hospital is something that a patient would actually embrace if they feel poorly enough. When we do level penalties on institutions for readmissions, I don't know where the patient voice is."

Despite the potential uphill battle of changing any federal policy, Fonarow said he remains optimistic and that the HFSA, AHA, and ACC hope to get a dialogue moving with the CMS in the next 3 to 6 months.

"These data are alarming enough that healthcare policy makers, who I really believe have the best intentions in mind but got in this mind-set that this was going to be a win-win, are going to see that there have been unintended consequences and will take action," he said.

This study was funded in part by a grant from the National Institutes of Health (NIH) and by a Young Investigator seed grant award to Gupta from the AHA GWTG-HF program. The GWTG-HF is sponsored in part by Amgen Cardiovascular and has been supported in the past by Medtronic, GlaxoSmithKline, Ortho-McNeil, and the AHA Pharmaceutical Roundtable. Dr Fonarow reported receiving research support from the NIH; consulting with Abbott, Amgen, Novartis, and Medtronic; and serving as a member of the AHA. Disclosures for coauthors are listed in the papers. O'Connor discloses consulting or serving on an advisory board for ResMed, Merck, Bristol-Myer Squibb, Actelion, ARCA, and Bayer. Walsh has no relevant financial relationships. Jessup is an advisor for Broadview Ventures.

Follow Patrice Wendling on Twitter: @pwendl. For more from | 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.