Hospitals' Medicare readmisson rates aren't improving

On Oct. 1, 2012, Medicare will start penalizing hospitals that have excessive readmission rates for heart attack, heart failure, and pneumonia. Yet despite several years of planning and effort, readmission rates remain stubbornly high. What's going on? Are the incentives not strong enough, or are the wrong things being measured?

September 24, 2012 04:00
J. Duncan Moore Jr./For The Medicare NewsGroup

On October 1, Medicare is going to start hitting hospitals where it hurts. Those institutions that have worse-than-average rates of readmission for three common diagnoses are going to lose as much as 1 percent of their government reimbursement.

Some 278 hospitals, including some of the nation’s premier teaching centers, will suffer the maximum penalty, and another 1,933 will be docked something less than 1 percent, according to an analysis by Kaiser Health News. A more recent analysis by MedPAC found that although all-cause readmissions have not declined as much as hoped, readmissions for these three diagnosis groups dropped more than overall reductions.

The readmission penalties were put into place by the Affordable Care Act (ACA) and were intended to prod the health care industry to take quality improvement seriously. Nearly 20 percent of Medicare patients return to the hospital within 30 days of discharge, costing the program $17 billion yearly. The idea is that if hospitals are charged for not making people healthy enough to heal at home after discharge, taxpayer money should be saved and senior patients should have better outcomes.

That’s the theory. But does it hold true?

In individual cases, there’s some evidence that it may. Individual hospitals or hospital associations have had success stories. In the universe of all American hospitals, however, the results are decidedly mixed.

In July 2012, the Centers for Medicare & Medicaid Services (CMS) published the most recent readmission rates, from July 2008 through June 2011, for the three diagnosis groups being targeted. The data showed that 24.7 percent of heart failure patients were readmitted, an improvement of 0.1 percent over the 3-year period. For heart attack patients, 19.7 percent were readmitted, also a 0.1 percent decrease.  But for those diagnosed with pneumonia, readmissions rose 0.1 percent, to 18.5 percent, according to the KHN analysis.

“It is not as easy to fix this problem as some people initially believed,“ said Robert A. Berenson, M.D., senior fellow at the Urban Institute and a former member of the Medicare Payment Advisory Commission (MedPAC), the board that advises Congress on Medicare issues. “You hear all these stories that every hospital has gotten religion on this topic and is working on it. Yet the national rate doesn’t change. I don’t know how to reconcile that.”

There is growing sentiment that the CMS may be on the wrong track. Critics of the new plan complain that the wrong things are being measured, that the reimbursement penalties are either too steep or not steep enough, or that people and institutions are being held accountable for events over which they have no control. Some say readmissions rates aren’t a proper measure of quality in the first place, and are a distraction from more meaningful interventions.

And nearly all parties agree that the remedies are expensive, difficult to implement, and might not even pay off in the long run.

The MedicareNewsGroup spoke with several hospitals that have reported solid progress on their readmission rates.

  • Banner Good Samaritan Medical Center, in Phoenix, AZ., has assigned an internal medicine physician to do nothing more than drive around to check on patients in their homes to make sure they understand their medication regimens and are following discharge instructions. The hospital anticipates this activity will save more than $500,000 annually.
  • OSF St. Francis Medical Center, in Peoria, Ill., has reduced readmissions for all Medicare patients by 2.25 percent, which translates to 234 fewer patients readmitted after discharge.
  • St. Luke’s Hospital, in Cedar Rapids, Iowa, trimmed its heart failure readmission rate from 32 percent in 2007 to 14 percent in 2011. The Brookings Institution has even profiled the hospital in a study.

The work at St. Luke’s “has been pioneering and transformative,“ said Saranyi Kurapati, M.D., of the Institute for Healthcare Improvement. Best practices in care transition are being propagated across the country by hospital associations working in collaboration with one another.

Nearly three-quarters of the members in the Illinois Hospital Association are undertaking projects to reduce readmissions, said Derek Robinson, M.D., a former CMS regional official who now leads a quality improvement initiative at the IHA. Rates are actually getting better in Illinois, he said. Last year, about 30 percent of Illinois hospitals would have been subject to the maximum penalty; this year, only 14.8 percent would, he said. The percentage of hospitals paying no penalty has improved, too.

But hospital spokesmen and quality-improvement vendors don’t volunteer the information that on a national basis, the rate of improvement is nil.

“To be honest, this has been a fiasco,” said Patrick Romano, M.D., a professor of medicine who does outcomes research at the University of California, Davis.” We have had, too often, situations where hospitals discharge patients to a black hole, with inadequate education and inadequate resources at home to provide the follow-up attention needed.” He added, “To the extent this policy is forcing hospitals to rethink the discharge process, I think it’s good.

But it’s not all good. This reform was driven primarily to achieve financial goals and to drive the health care system toward better performance. Unfortunately, since the quality movement started in the 1990s, the health care community has made only fitful and uneven progress toward guaranteeing patient safety. It was thought that financial incentives —or in this case, financial penalties—would be required to motivate hospitals to move the dial forward. The penalties also act to remind the hospitals that resources are limited.

The total penalty that hospitals will forfeit in the coming fiscal year is $270 million, a drop in the bucket.

“Most financial people would say that the hospital is much better off losing 0.3 percent than they are reducing actual readmissions,” Berenson said. “Most hospitals will probably pay lip service to this kind of incentive.”

The maximum penalty for higher-than-expected 30-day readmissions rate gradually increases to 3 percent over the next two years. Even so, the costs of the process improvement interventions might exceed what the hospital could lose in reduced payments.

As the diagnosis-related group (DRG) payment system works now, a hospital receives a predetermined lump sum to treat a patient for a specific diagnosis. Obviously, there is an incentive to discharge quickly, and there’s no penalty if the same patient returns, because a new DRG payment is triggered.

Patients sometimes complain of being discharged too quickly. “They come back and say, ‘I was still sick, I told them I was going to come back, they discharged me anyway,’” Romano said.

  • And for many patients and their families, not every readmission is a bad thing. Family members may get a respite from the burden of caring for a loved one. The hospital may re-educate them on what they need to do, or try a new therapy in place of one that failed before.

    “Sometimes it takes a readmission for the patient to realize, “I can’t go home, I have to go to a skilled nursing facility,’ and they weren’t ready for that yet,” said Peg M. Bradke, director of heart care services at St. Luke’s. Patients and families are often working on a different timetable than health professionals. “They don’t always want to hear about palliative care and hospice right out of the chute,” she said.

    A lot of the recent work in U.S. hospitals is geared toward improving the transitions from one care modality to another. This is expensive and involves coordinating activities among a host of peripheral players.

    This is one of many concepts, some of which have been tried and abandoned. “All you have to do is look systematically at the literature,” Romano said. “There’s more that hasn’t worked than has.” For example, commercial disease management as having nurses call patients on the phone, has not been shown to work. Neither did check-in phone calls from nurses to patients. “What has worked has been expensive intervention,” he said.

    Many hospitals, especially those serving low-income communities, find their patient base frequently lacks the means or education to follow the measures in a discharge plan. Patients don’t fill prescriptions or keep appointments with primary-care physicians, if they even have one. Or perhaps they don’t have transportation to the doctor’s office. Or their health literacy is low to begin with. There just isn’t the support structure in the community for the patient to stay well.

    Prestigious hospitals such as Barnes-Jewish in St. Louis have been embarrassed by media exposés of poor readmission rates, and have had to institute expensive remedial plans to help low-income patients.

    The problem for the health system as a whole is the variability among institutions, Romano said. “To penalize the safety net hospitals for taking care of the most vulnerable patients is worrisome,” he said. “It is precisely these hospitals that need more resources.”

    Berenson has proposed an alternative to the current CMS approach: in essence, a warranty. Hospitals would receive an increase in their base payment for each original admission but not get paid for readmissions. Internationally, Germany and the United Kingdom, and here at home Maryland, have all moved in this direction, and it seems to be working, he said.

    He published his ideas in the New England Journal of Medicine in April. “We don’t pay you for your readmission, but we make it up to you by raising your rates,” Berenson said.

    By making the process hospital-specific, Berenson avoids the complicating factors of the case mix, risk selection, and socioeconomic status differentials. He’d like to try it out in a single state before rolling it out nationally, to make sure it’s effective.

    “It’s a positive reinforcement rather than a penalty,” he said. Further, it will likely have a spillover effect and reduce admissions.

    The current CMS effort, by contrast, is “a blunt tool,” Robinson said. “It’s not as sharp as it might be.“