Lower Readmissions Linked to Higher Risk of Death

An alarming new study suggests that the emphasis on reducing hospital 30-day readmissions has inadvertently led to increased risk of death for Medicare patients hospitalized with heart failure.

As many as 10,000 heart failure patients could die prematurely each year because of misguided efforts that keep them out of the hospital to avoid the financial penalties attached to higher readmissions, according to a study published this week in JAMA Cardiology.

Study co-author Gregg C. Fonarow, MD, a researcher and professor of cardiovascular medicine at UCLA, spoke with HealthLeaders Media. The following is an edited transcript.

HLM: The implications of this study are staggering. What are we to make of them?

Fonarow: This study represents the sum of all fears. Concerns were raised about flaws in the readmission metric, flaws in the ways the penalties were being constructed and the potential and concern for unintended consequences. Now we’re seeing that.

That’s not to say that trying to reduce preventable readmissions is not important. But, it needs to be coupled with strong efforts to ensure patient safety and to reduce preventable deaths.

HLM: How many lives are we talking about?

Fonarow: If we were to extrapolate this to all Medicare beneficiaries hospitalized with heart failure, we are talking about maybe 10,000 patients a year with heart failure losing their lives as a consequence of this program. Even one patient being harmed isn’t worth any degree of readmission reduction but to have potentially each year 10,000 or more patients having been potentially impacted with increased mortality is just an absolutely devastating level of potential harm.

HLM: What’s wrong with the design of the policy?

Fonarow: The way Medicare constructed their readmission reduction program was solely focused on 30-day readmissions with strong financial penalties up to 3% of every Medicare dollar. But, when you look at what was Medicare doing to ensure patient safety and trying to incentivize around lower mortality rates, the maximum penalty a hospital could face, even with a 100% mortality rate, would be 0.2% of their total Medicare revenues. That’s sending a message 15X from a financial standpoint that it’s more important to reduce readmissions than to be focused on patient safety or mortality.

HLM: Why the focus on heart failure patients?

Fonarow: The concern was greatest for heart failure patients. This study adds very striking, statistically significant, clinically relevant evidence that risk-adjusted mortality rates for Medicare beneficiaries hospitalized with heart failure have gone up. They’ve gone up in the first 30 days and that continues on out to one year. It looks as if the hospital readmission reduction program has been associated with a serious and devastating unintended consequence of increased mortality for heart failure patients.

HLM: How does this unfold at the hospital level?

Fonarow: Part of it is the financial resources being taken away by virtue of the penalty. Those of lower socioeconomic status treated at safety net hospitals are more likely to be re-hospitalized in a way that wasn’t adequately captured in the 30-day readmission metric. Those hospitals got oversized penalties. The most vulnerable patients are treated at the most vulnerable hospital, and desperately needed resources were taken from those hospitals, for which they no longer had available for staff or programs on patient safety and key therapies that can improve outcomes.

But there are also other ways. This financial penalty and public reporting led to hospitals putting pressure on clinicians to reduce 30-day readmissions and keep patients out of the hospital. So, the clinician trying to respond to the pressure and incentive has a patient who is at home and not doing well but they’d been hospitalized 21 days before and you’re in this dilemma. The clinical situation may necessitate hospitalization. You want to do the right thing for the patient, but maybe if I can buy a little more time and leave them at home, take that little extra risk it will work out OK and if they get hospitalized beyond 30 days that is OK.

There are patients being shunted from the ER to an outpatient observation unit rather than being hospitalized because of Medicare policies. Now, that patient can no longer for qualify for home health or skilled nursing facility or other resources because they were not hospitalized.

So, there are a number of ways by which these incentives inadvertently could have led to this harm that’s been observed.

HLM: Do you believe readmissions should be a quality metric?

Fonarow: Readmission reductions in isolation is not patient-centered. It needs to be in conjunction with meaningful patient-centered metrics including their health status and patient survival and coupled with strong measures to ensure there is not gaming and not unintended consequences. In light of this data, for heart failure patients, readmissions needs to completely and immediately cease and we must find ways to mitigate the damage that has already been done before ever reconsidering that as a valid metric for heart failure patients.

HLM: What can be done to restore 30-day readmissions as an accurate metric for value-based, care coordination?

Fonarow: There are a number of measures where you can look at the processes that have been associated with outcomes, and also directly measure patients’ health status as well as, most critically, patient survival. They can be integrated into a multi-dimensional component of assessment and still incentivize that kind of quality of care and moving beyond the hospital walls and investing in heart failure disease management programs that have been shown to improve all components of care and outcomes.

HLM: What should be done with the study?

Fonarow: I would like to see immediate action taken by Medicare to convene a multi-stakeholder panel to discuss steps forward to try and modify the program, suspend it with regards to heart failure patients, and develop proactive steps to mitigate it and better understand lessons learned and how these kinds of unintended consequences can be avoided in the future.

HLM: How did this metric morph into its evil twin?

Fonarow: You can say it did reduce readmissions. It didn’t backfire that way, but it backfired in a far more disastrous way.

It’s why it’s so important, just as we not unleash a new therapy on patients without having tested it, it’s critical to recognize that policy decisions can have disastrous side effects. Major policies should be pilot tested. There needs to be proactive close monitoring for any unintended consequences.

This was unleashed with no monitoring. Here we are talking about data that took a few years to assemble, but as early as 2013-14 there was a clear increase in mortality that we are just finding out about. Now that this data has come to light, it’s critical that we act.