Shift in CMS Policy Shines Spotlight on PCI Readmissions

In recent years, hospital readmission after percutaneous coronary intervention (PCI) has become a hot topic in interventional cardiology, with much talk about the economic and clinical consequences. Questions include whether readmissions should serve as a marker of hospital quality and what constitutes the ‘correct amount’ of in-hospital and follow-up care.

Reimbursement policies lie at the heart of the debate. Since the Centers for Medicare and Medicaid Services (CMS) reports risk-adjusted 30-day readmission rates, the metric has already effectively become a quality performance measure. And since 1983, CMS has reimbursed hospitals on a diagnosis-related group basis; the system is intended to encourage more efficient care, though some say it may incentivize premature discharge by paying for patients on a per stay and not a per day basis.

In 2013, the Patient Protection and Affordable Care Act and its Hospital Readmissions Reduction Program will bring concerns over readmission to a head by beginning to penalize hospitals that have excessive readmissions related to acute MI, heart failure, and pneumonia. The program will curtail reimbursement by reducing payment for the readmission if it occurs within 30 days of the index hospitalization. Two years later in 2015, CMS will expand the program to encompass additional conditions, with PCI said to be a likely target.

According to Joseph G. Cacchione, MD, of the Cleveland Clinic Foundation (Cleveland, OH), if everything goes as planned, hospitals will be “incentivized to say they don’t want any of these patients to be readmitted.” But the situation is not currently ideal, either, he told TCTMD in a telephone interview, noting that hospitals and physicians may recognize that readmissions “look bad statistically and are not right for the patient, but by the way we get paid again.”

In a telephone interview with TCTMD, Jeffrey W. Moses, MD, of Columbia University Medical Center/Weill Cornell Medical Center (New York, NY), said that CMS has no foundation or basis for its decision. “Somehow they decided [readmissions were] a quality indicator even though . . . no one is clear why people are being readmitted,” he countered.

Little Clarity but Some Hints in Literature

Three years ago, the first study looking solely at readmission after PCI found a 30-day rate of 14.6% among 315,241 PCI procedures at 1,108 US hospitals (Curtis JP, et al. J Am Coll Cardiol. 2009;54:903-907). Since then, other research has shown that such readmissions increase 1-year mortality but no one has identified modifiable risk factors that might curb repeat hospitalization. Patient age, sex, indication for PCI, and length of index hospitalization appear to be influential, however.

A study of 36,000 patients who survived to discharge after PCI in Massachusetts, meanwhile, offers additional insight (Circ Cardiovasc Interv. 2012;5:227-36). As the first to examine variation among hospitals, its findings suggest that only a small fraction of readmissions are due to actual measures of hospital quality. The 10 most common principal discharge diagnoses comprised 65% of all readmissions. Among these were ischemic heart disease, chest and respiratory symptoms, and heart failure.

Readmission Differs Elsewhere

The situation also varies around the world, with the United States being unique in its propensity toward readmission.

A study of 5,745 STEMI patients slated for PCI who were enrolled in the APEX AMI (Assessment of Pexelizumab in Acute Myocardial Infarction) trial, for example, looked at the United States, Canada, Australia, New Zealand, and 13 European countries (JAMA. 2012;307:66-74). Of the patients who survived to hospital discharge, 11.3% were readmitted within 30 days. Notably, the repeat hospitalization rate was 14.5% in the United States and 9.9% elsewhere. Once readmissions for elective PCI were excluded, 8.6% of patients from the overall cohort were readmitted, amounting to 10.5% in the United States and 7.7% outside the country.

In addition, the index hospitalization tended to last longer for patients outside the United States, with 60% of US patients staying 3 days or less vs. 15.9% of those treated outside the country (P < 0.001). On average, the duration ranged from 3 days in the United States to 8 days in Germany.

After adjustment for baseline characteristics, patient length of stay was no longer a significant predictor of 30-day readmission. The strongest predictors were multivessel disease, US enrollment, and baseline heart rate. When the researchers adjusted for length of stay, however, US enrollment was no longer associated with readmission.

These data lead to questions regarding the pros and cons of the US health-care system. Dr. Moses argued to TCTMD that, when it comes to PCI, the United States spends less than other countries such as Germany. Because there is no mortality difference related to readmission, he said, such hospitalizations are only deemed bad because they are erroneously considered more expensive.

Edward L. Hannan, PhD, of the University of Albany (Albany, NY), stressed in a telephone interview with TCTMD that CMS is not the enemy in this situation. “Certainly what CMS and all these organizations that are committed to reducing readmission rates are trumpeting—as they should be—is to try and improve the continuity of care and coordination with primary outpatient care,” he said.

“I think maybe the reason why readmission rates in places like Europe and other places are lower than ours is because they have a better coordinated health system than we do,” Dr. Hannan commented. “We have a fragmented health system in many respects.”

Too Many Exceptions for Blanket Rule

What makes this situation unique in terms of PCI, Dr. Hannan said, is that interventions are often staged, as opposed to other types of procedures. That aspect would, of course, give a sense that readmission is elevated when in fact the initial procedure is merely extending over several visits.

Furthermore, it is inappropriate to suggest a standard length of stay for PCI patients as complications change the math. “It’s just too simplistic to say, ‘Let’s just keep everyone in the hospital longer,’” he asserted. Ultimately, “you can’t expect to have no readmissions. Many of these readmissions are not due to problems that occurred in the hospital.”

Which brings the debate back to the new CMS policy and its pending effects on PCI. Until more is known about the reasons for readmissions, Dr. Moses argued, any decision to change reimbursement is premature.

CMS “only has a vague portrait of the patients’ reasons for readmission, much less whether we can determine [if these events are] preventable or not. That’s what they should be focusing on,” he said. “No one wants to acknowledge it, but they are struggling to make up rules that save them money.”

Dr. Moses emphasized the fact that many PCI patients are older and have multifaceted disease. As such, he said, they will inevitably be hospitalized because of other comorbidities.

A Possible Alternative?

TCTMD sought comment from CMS, but the agency did not reply to repeated attempts to obtain an interview.

Dr. Hannan proposed separate reimbursement plans for preventable and unavoidable readmissions, with some way to differentiate between them. That way readmission rates are not judged on an overall basis, but rather characterized by patient condition. However, “there are unfortunately going to be a whole bunch of cases in the middle where it’s not clear if [the readmission will be] avoidable or unavoidable, and those are difficult,” he said, suggesting that future research should focus on whether this subset is “shrinkable.”

Devising this payment plan would “be complicated and somewhat arbitrary,” Dr. Hannan predicted. But “you can take measures to try and reduce that arbitrariness by calculating risk adjusted rates, looking at the reason for readmission, separating out the staged patients, etc.”

Time to Embrace Change

Since CMS has already made its decision, the community has no choice but to embrace the policy, Stephen G. Ellis, MD, of the Cleveland Clinic (Cleveland, OH), told TCTMD in a telephone interview.

“From a physician’s standpoint, you can lobby CMS, but I think they are fairly set in what’s going to happen,” he said. “Most centers are going to have to . . . come up with a way of reducing the readmission rate.”

Ultimately, Dr. Cacchione said that “payment reform will accelerate the process of trying to prevent possible readmissions. We are going to be more incentivized to keep people out of the hospital if it is on our nickel.”

Dr. Ellis reported that his institution is currently piloting a program that requires 1-week follow-up for all PCI patients with a physician assistant. Although the program is too new to yield any results, he predicted it will help identify preventable complications and hopefully reduce readmission rates.

This is just one piece of the medical economic dilemma in the United States, where regulations intending to improve quality may actually increase costs until high-risk patients can be reliably separated from lower-risk patients and targeted for intervention, he concluded. “I don’t think this is going to go away any time soon.”

 

---

Source:
US Statutes at Large. Patient Protection and Affordable Care Act. http://www.healthcare.gov/law/full/index.html. Effective March 23, 2010. Accessed June 21, 2012.

 

 

Related Stories:

• Debate Continues on PCI Readmissions, Focus Shifts to Hospital Variation
• Study Shines Spotlight on US Readmissions After STEMI
• Hospital Readmission After PCI ‘High’

Click here for a listing of companies that provide support to the Cardiovascular Research Foundation, owner and operator of TCTMD.