Reducing Heart Failure Readmissions & Rethinking Their Meaning
Since CMS began publicly reporting hospital readmission rates for heart failure in 2009, a greater emphasis has been placed on quality initiatives that can help curb them. But are readmission rates the best metric by which to judge hospital quality?

When to follow-up?

Heart failure (HF) is the No. 1 readmission diagnosis, which is related to a strong disconnect between hospitals and private practice settings, ultimately resulting in a lack of coordinated care, says Adrian Hernandez, MD, of the Duke University School of Medicine in Durham, N.C. “There is no systematic protocol for early follow-up of heart failure patients. But it’s sometimes as simple as making a phone call when a patient misses an appointment,” he says.

Early follow-up is integral to decreasing readmissions, says Hernandez. He and colleagues found that patients were readmitted less often if they were followed up within seven days by any type of physician—primary care, general internist or cardiologist. Interestingly, those patients followed up by cardiologists had a lower 30-day mortality rate (JAMA 2010;303(17):1716-1722). “The ideal follow-up is within one week from discharge,” says Hernandez. “We saw an association with fewer readmissions at 14-day follow-up, which became weaker the further out from discharge.”

Hernandez calls the first week of discharge “crucial” for evaluating patient status. “Patients are no longer under 24-hour hospital supervision. They may not have picked up their medications or they may not have followed up on a test. Early evaluation should include a review of therapeutic changes and a thorough assessment of the patient’s clinical status outside of the highly structured hospital setting.”

While HF readmissions can be dangerous and costly—soaking up an estimated $17.4 billion, or 5 percent of total Medicare spending—a majority of them are preventable. Practitioners are now emphasizing a multidisciplinary approach, comprising cardiologists and other physicians, nurse practitioners, nutritionists/dieticians and pharmacists, among others. “Disease management programs are so effective because they are not focused on a one-size-fits-all solution. Instead, the team works to individualize a patient’s problems and examine why they lead to readmissions, along with how they can be alleviated,” says Barbara Riegel, MD, a professor at University of Pennsylvania School of Nursing in Philadelphia.

Patient education also is key to preventing rehospitalization, says Riegel. “We need to change behavior. We need to urge patients to follow a strict low-salt diet and guide them to take medications properly, in addition to using a team-based approach to care.”

In a 2009 meta-analysis published in Health Affairs, Riegel and colleagues found that patients in a disease management program had 25 percent less all-cause readmissions and 30 percent less readmission days. The team-based approach reduced HF readmissions specifically by 2.9 percent and readmission days by 6.4 percent per month. Riegel estimated that the 2.9 percent reduction per month in HF readmissions could lead to 14,700 to 29,140 fewer hospital stays per year.

A BOOST from Michigan

In May, the University of Michigan in Ann Arbor, Blue Cross Blue Shield of Michigan and the Society of Hospital Medicine (SHM) announced the Michigan Transitions of Care Collaborative, or MTC2, a training and mentoring program where physician groups and hospitals share best practices in an effort to reduce hospital readmissions and ER visits.

The initiative, based on SHM’s Project BOOST (Better Outcomes for Older Adults through Safer Transitions), involves 15 physician organizations working with 14 hospitals across the state. “The idea is to help patients get better prepared to move from the closely observed setting of the hospital back to their home,” says Christopher Kim, MD, an assistant professor of internal medicine and pediatrics at the University of Michigan and the program director of MTC2.

Project BOOST reports 30-day readmission rates, patient satisfaction, length of stay and other process metrics on a quarterly or monthly basis to help identify specific trends and provide transparency within hospitals.

Other strategies/programs utilized include:
  • Teach Back, which involves getting patients to describe important steps they need to take to ensure a smooth transition from hospital to home;
  • Preparations to Address Situations (after discharge) Successfully, or PASS, which helps patients understand potential problems that may signify a future event that could land a patient back in the hospital;
  • Call Back, an outreach effort to patients within the first 48 to 72 hours after discharge; and
  • Transition Back, a program to communicate the patient’s discharge to the cardiologist or primary care physician in a timely manner through direct phone or electronic communication, and sending the discharge summary report.

Road to reimbursement

Based on the new healthcare reform law, CMS will begin reducing payments to hospitals with high readmission rates on Oct. 1, 2012.  But there are concerns about incentivizing rapid discharge. In a recent letter to the editor, Gorodeski et al from the Cleveland Clinic questioned whether hospital readmissions are always predictors of poor quality of care (N Engl J Med 2010;363:297-298).

Gorodeski says that he and his colleagues had noticed in the CMS Hospital Compare data that their 30-day HF readmission rate was higher than the national average (28 vs. 24.7 percent), while their mortality rate was lower (8.8 vs. 11.2 percent). This prompted them to examine whether readmission rates should be the sole quality indicator. They then assessed data from 3,857 hospitals that reported to the Hospital Compare website, finding that a higher occurrence of readmissions after index admissions for HF was associated with a lower risk-adjusted 30-day mortality.

There are two possibilities for this finding, says Gorodeski, a HF and heart transplant cardiologist. “First, U.S. hospitals that appropriately treat patients during the index admission and have a lower mortality rate consequently have more patients discharged, which increases their potential for readmissions. Second, hospitals that use more resources—i.e., more procedures, surgeries or devices—that are indicated within 30 days post-discharge will have a higher readmission rate, even though they are providing evidence-based care.”

Other recent studies also have found a decrease in 30-day mortality with a concomitant increase in 30-day readmissions, including Krumholz et al who reviewed data on more than seven million Medicare patients over 13 years (JAMA 2010;303(21):2141-2147) and Heidenreich et al who looked at more than 50,000 patients in the Veterans Affairs Health Care System over five years (J Am Coll Cardiol 2010;56:362–8).

In an accompanying editorial to the Heidenreich study, Duke University’s Christopher M. O’Connor, MD, and Mona Fiuzat, PharmD, wrote that hospital readmissions are a poor marker for quality. They suggested that hospitals with low mortality rates in this population should not receive deductions on quality if there is an increase in rehospitalization rates. They also suggested that total hospital days alive over a 30-day period should be the marker of quality following heart failure hospitalization.

In the meantime, CMS also has considered bundling payments for 30-day periods of care. While cautiously optimistic that bundling could improve post-discharge care, Hernandez says that without a “truly accountable healthcare organization that brings all of the players together, it’s hard to know whether this kind of reimbursement plan would meet its full potential.”

And finally, the newly enacted Patient Protection and Affordable Care Act in the healthcare reform package  incorporates the Community-Based Care Transitions Program, which seeks to pair hospitals with community organizations to help implement evidence-based hospital-to-home transition strategies. Hospital readmissions for high-risk Medicare patients over a five-year period will be the metric of the program’s success.

“If we could get reimbursement for a transitional care model where a nurse or other staff would follow up with the patient for the month after discharge, this would be heaven,” says Riegel.