Thinking Seriously About Readmissions—and Care Transitions | Healthcare Informatics Magazine | Health IT | Information Technology Skip to content Skip to navigation

Thinking Seriously About Readmissions—and Care Transitions

September 4, 2012
by Mark Hagland
| Reprints
A new report by CSC’s Jane Metzger offers insights on the challenges facing hospital leaders in averting readmissions going forward

In August, the Waltham, Mass.-based Global Institute for Emerging Healthcare Practices, a division of the Falls Church, Va.-based CSC, released a white paper entitled “Preventing Hospital Readmissions: The First Test Case for Continuity of Care.” That white paper, authored by Jane Metzger, principal research in the Global Institute, looks at a variety of factors that might help hospitals to reduce avoidable readmissions, in the context of the healthcare reform-driven mandate to do so, under the Affordable Care Act (ACA).

As Metzger notes in the introduction to the white paper, “Hospital readmissions were one of the earliest targets of both quality measurement and performance-based incentives for Medicare and other insurance programs,” partly because a great deal of information is easily available through claims data that can help payers (both public and private) to target issues in hospitals that might be behind readmission levels. It’s not surprising, then, of course, that avoidable readmissions work became a mandatory element of the ACA for all hospitals receiving Medicare payment.

What Metzger finds in her review of the literature is that averting the readmission of high-risk patients—primarily those defined as having heart failure, pneumonia, or an acute myocardial infarction (AMI)—“is very challenging, because so many community and patient factors contribute to the problem, many of them outside of the direct control of the hospital. However,” she adds in the white paper, “research, combined with practices in hospitals with a track record of reducing readmission, shows that comprehensive discharge planning and post-discharge care and support during the transition period reduces readmissions in high-risk patients.”

CMS, Metzger notes, has targeted four key outcomes measures, in the context of the mandatory readmissions reduction program under Medicare: 30-day risk-standardized readmission rates for AMI, heart failure, and pneumonia, and the 30-day risk-standardized readmission rate for all conditions.

In figure 1 below are the “Common process breakdowns associated with potentially preventable readmissions”:

Figure 1

Common Process Breakdowns Associated with Potentially Preventable Readmissions

Care Gaps During Stay

  • Patient safety (especially medication- and infection-related)
  • Medication reconciliation not completed or inaccurate at admission or discharge

Patient Factors

  • Lack of understanding of post-discharge plan of care
  • Lack of understanding of what to watch for (warning signs), how to respond
  • Non-compliance with any or all elements of post-discharge self-management and care
  • Lack of Timely Post-Discharge Care
  • No appointments available or no relationship with a primary care provider
  • Logistics, such as no transportation
  • Primary care physician unaware of hospitalization


  • Delayed, lacking or inadequate communication with next provider or direct care
  • Lacking or inadequate communication with home care provider (including family)

Such issues are going to become increasingly prominent in the near future, industry observers agree. Indeed, in releasing details of the program on Aug. 1, the Centers for Medicare & Medicaid Services (CMS) revealed that more than 2,200 hospitals, or two-thirds of U.S. hospital organizations, will see payment of reductions of up to 1 percent within the next year.

Jane Metzger spoke recently with HCI Editor-in-Chief Mark Hagland about her findings, and about their implication for healthcare and healthcare IT leaders. Below are excerpts from that interview.

The mandate for significantly reducing avoidable readmissions has really become an immediate challenge for hospitals, hasn’t it?

Yes, and hospital leaders are going to have to take the lead on it, because it’s here and now for them. Now, over time, the role of hospitals in transitional care will decline somewhat. But I think they’ve got a big problem today, because of the kinds of collaboration required, the continuity of information, and really having to do it all is a huge challenge.

Jane Metzger

What are the biggest challenges right now for hospital leaders in this area?

Well, I think the approach to readmissions in many hospitals has been somewhat narrowly focused. For example, a lot of programs you read about will be only for congestive heart failure (CHF) patients. And it is true that what programs look like will be different for different conditions. So I think that one big challenge for hospitals is that they’re going to have to think more about this process of transitional care and support, across larger numbers of patients. In other words, it’s no longer the CHF program. But I think that what they’ve been doing for CHF patients, except for the condition-specific elements, is a model for what they’re going to have to do for many patients. And though they’re going to have to do the same kinds of things for high-risk patients, what I’m arguing in the care plan is that I think there’s going to have to be a transitional care plan for all patients. And it may be simple to begin with.

But this whole problem of, we’ve got patients in for shorter stays—this has been happening for decades now—and we’ve got increasingly older and more fragile patients, with a bigger disease burden, so the discharge has become a riskier proposition. So instead of having a couple of nurses who work with the medical teams on a CHF program, I think they’re going to be organizing the transitional care as its own process. Because let’s say you’re doing a phone call to CHF patients on the day after discharge. Well, there are going to be many other patients who need phone calls. And you’re going to need to do that efficiently and in an organized way.

So yes, the nurse who knows what to say to a CHF patient, will need models for a pneumonia patient, or a really fragile diabetic, or a COPD [chronic obstructive pulmonary disease] patient. So you’re going to want to efficiently use the hospital resources available. And you’re going to want to think for each patient about what is at risk, what’s involved, and who is receiving the patient, and when the provider will connect with that patient.

So it makes sense for the hospital to connect as early as possible, and create the handoff. It may be this external program that actually executes the handoff the day after discharge. Or it may be a home visit. And the nurse practitioner from the PACE program is going to be visiting the home on day two. So I’m arguing that over time, you’re going to be connecting patients more formally with whatever that external care management party is; for some patients, it might be the patient’s primary care physician. And you’re going to want to work out elements like the med list, so you’re operating with useful information. But I think it’s a useful construct to operate with these formalized care plans.

But if you think about it, Medicare is reinforcing these silos, because the three conditions in the measures that Medicare is using are around heart failure, pneumonia, and cardiac issues (AMI—acute myocardial infarction—etc.). Now, in the accountable care organization program, they’re looking at all measures. In any case, it will be about taking this and making it into a much broader, non-siloed program, that comes up with a good transition plan for each patient.

I’ve reported on the Hospital at Home program that’s now live at Presbyterian Health Services in Albuquerque, which is transitioning patients very successfully from home care into a hospital-care-in-the-home setting, and back. What are your thoughts on programs like that one? Such programs like that could fit into this schematic, right?

Absolutely. And there are programs I referenced, not by name, but by concept, in the paper, that talk about front-loading post-discharge visits, including, if the patient can be gotten there, to the primary care physician on day one. In another instance, one program sends a team of a nurse practitioner, a social worker, and a pharmacist, and they go to the home. But this idea of providing pretty intensive care and support in those first few days—the Hospital at Home folks at Presbyterian call it an admission—that supports my thesis that we’ve got these categories—acute care, emergency care, and primary care, but that we have to start delivering transitional care. And the Presbyterian example is one set of interventions made available to the patient at home. And I think what you observed in Albuquerque is a model for what I’m writing about in this paper.

In fact, people have been working on readmissions and doing research on all this, for decades. So one of the things I tried to do in this white paper was to organize what’s known today. And that’s what I’m trying to do in the sections “What Research Tells Us” and “What Works in Readmissions.” And there are a lot of qualifiers to reach what we know. But what I decided after reading all the studies and program descriptions, was that several things stood out for me. One thing is that it’s pretty clear that doing a good job with the discharge plan helps, but high-risk patients, at least, need post-discharge care and support. And under, “What Research Tells Us,” I looked at programs that have achieved exemplary results, and have looked at what results they achieved, and the two tables, on and starting on page 6, and what I realized was, whether there’s been a formal research study saying a particular element works for particular patients, or whether it means looking at exemplary programs, as the Commonwealth Fund, the California Health Care Foundation, or IHI [Institute for Healthcare Improvement], through the Star Program, have found, when it came to comprehensive discharge planning, is that every one of these “validated” programs included one or more interventions under each of the three categories I list in the figure on page 6 of the report titled “Promising Components and Interventions—Comprehensive Discharge Planning.”

The reality is that we know who the patients are who are at highest risk; but we don’t know yet what the “killer risk” is yet, because most patients have a bucketful of risks. And so there’s no magic bullet saying, if we do X for patients living at home alone, they’ll do fine, right? And also, there’s quite a bit of research on the elements in the hospital stay that can contribute to avoidable readmissions.

But the bottom line is that there’s no magic bullet or magic combination of things to do, so the hospital is left trying to figure out what magical things to do for each patient. So you have to organize programs that try to address the various risks, whether it’s the typical risks for a fragile COPD patient, the typical risks of someone going home to poor supports, the typical risks involved if someone doesn’t have a primary care provider with whom to arrange follow-up care. So it really is individualized, and it’s a very complicated problem without a magic bullet-type solution. So that says to me that you have to customize your solutions.

Obviously, you need to have a really strong IT foundation for this.

Yes, and what I talk about on page 9, and I italicize the things that meaningful use does, I try to talk about it at a fairly high level, on page 9.

I think the opening statement in that section says it well: “Tracking patients will be essential to ensuring planned post-discharge care and support actually occurs in time to address gaps. This will only be possible with the assistance of health IT to accomplish communication and close loops.”

Thank you. 

The Health IT Summits gather 250+ healthcare leaders in cities across the U.S. to present important new insights, collaborate on ideas, and to have a little fun - Find a Summit Near You!


See more on