Home Cover Story Have a problem with readmissions?

Have a problem with readmissions?

June 2009

Published in the June 2009 issue of Today’s Hospitalist

Ask Kimberly Bell, MD, what her hospitalist group is doing to reduce unnecessary readmissions, and she rattles off a long list of initiatives, from talking to primary care physicians about their patients’ discharge and checking on follow-up appointments to moving most of the group’s patients to a dedicated floor so physicians can spend more time on discharges.

While most of those initiatives were designed to improve continuity of care and boost satisfaction scores, they are now being viewed as weapons in the war on reducing unnecessary readmissions. That’s putting hospitalists like Dr. Bell, who directs the hospitalist program at Centennial Medical Center in Nashville, Tenn., at the center of one of the hottest topics in American health care: reducing hospital readmission rates.

It wasn’t that long ago that the issue of hospital readmissions was the province of CMOs, not CEOs. But in the last few months, the subject has taken on a much higher profile, with the Obama administration hoping to save billions by targeting hospital readmissions.

As a signal of just how serious the issue has become, the Centers for Medicare and Medicaid Services (CMS) is planning this month to begin posting readmission rates for patients initially hospitalized with three conditions “heart failure, pneumonia and acute MI “on its Hospital Compare Web site. And last year, the Medicare Payment Advisory Commission (MedPAC) recommended that the CMS reduce payments to hospitals with a relatively high number of readmissions.

Those developments have hospitalists everywhere scrambling to reassess what they can do to prevent unnecessary bouncebacks. The good news is that many hospitalists, like Dr. Bell, have been able to repurpose initiatives that have been part of hospital medicine quality improvement efforts for years.

But while hospitalists may have a head start, that doesn’t mean they can expect smooth sailing as the executive suite and insurers seek ways to reduce readmissions. Although the specialty has earned the respect of administrators over the years based in no small part on its ability to reduce length of stay, hospitalists will now be expected to shift their energies to tackle readmission rates.

“The issue of readmissions shows that focusing on one metric “like LOS “to gauge success or failure is potentially near-sighted,” says Christopher Frost, MD, vice president of hospital medicine and clinical excellence for TeamHealth Hospital Medicine, a national hospitalist management company in Knoxville, Tenn. “Achieving a balance between optimal LOS and optimal readmission rates is really the goal.”

New evidence
The attention being paid to hospital readmissions heated up earlier this year, when a widely publicized study found that nearly one in five Medicare patients was readmitted within 30 days. The study, which appeared in the April 2, 2009, issue of New England Journal of Medicine (NEJM), looked at more than 13 million Medicare patients discharged over 15 months between 2003 and 2004.

Researchers found that 19.6% of those patients were readmitted within 30 days, and that nearly two-thirds were either rehospitalized or died within a year. Most of the rehospitalizations were for medical diagnoses, including 70.5% of patients discharged after surgery who were readmitted within 30 days.

Half of the patients initially hospitalized for a medical condition and then readmitted within a month had no bill for a follow-up visit between the time of discharge and readmission. Researchers estimated that the cost to Medicare of unplanned rehospitalizations over the study was $17.4 billion.

The five DRGs that the study found associated with the highest rates of 30-day readmissions will come as no surprise to hospitalists: heart failure, psychoses, vascular surgery, COPD and pneumonia. But while these five DRGs might seem like an obvious place to start to cut readmission rates, one of the study’s authors says that would be a mistake.

“Each of those populations is a much smaller percentage than you’d think,” says Mark Williams, MD, chief of hospital medicine at Chicago’s Northwestern University Feinberg School of Medicine. He notes that rehospitalizations for heart failure, for example, accounted for less than 8% of all readmissions in the study.

“In addition, a lot of research is beginning to show that the bulk of these rehospitalizations aren’t necessarily due to heart failure,” Dr. Williams explains, “but to other, chronic medical conditions patients may have.”

Instead, Dr. Williams adds, the study highlights the many areas in which hospitalists may need to form “a better handshake with the outpatient setting and optimize how we educate patients.”

One program designed to meet those goals is Project BOOST, a Society of Hospital Medicine initiative for which Dr. Williams is principal investigator. The initiative promotes a host of interventions to improve transitional care including a readmission-risk assessment tool, discharge checklist and patient education materials.

Toward “a better handshake”
Hospitalist leaders agree that the study highlights the need to improve transitions of care to cut readmissions. They also note that in many instances, hospitalists already target issues that can drive readmission rates up.

Sound Inpatient Physicians, a national hospitalist practice based in Tacoma, Wash., for instance, is one of several large hospitalist companies that has built a centralized call center to contact both discharged patients and primary care offices. Sound also has created software that automatically transmits patient data to outpatient practices.

According to Robert Bessler, MD, president and CEO of Sound, contacting patients post-discharge allows hospitalists to pinpoint problems at particular hospitals, such as a high percentage of patients confused about their post-discharge medications. Enhanced communication and monitoring, he adds, have cut readmission rates in half in some hospitals where the company operates hospitalist programs.

Another huge driver of readmissions “flawed medication reconciliation “has been on hospitalists’ radar screens for years. Eric Rice, MD, MMM, a hospitalist at Methodist Hospital in Omaha, Neb., says that data from his hospital revealed an 8% medication reconciliation error rate. The hospital is now working with a vendor to create software that will allow outpatient offices to enter their patients’ medication lists into the same system that the hospitalists use.

Bigger obstacles
But even as hospitalists acknowledge that much can be done to improve transitional care, some point out that there are formidable obstacles to preventing hospital readmissions, which was borne out in the NEJM study.

TeamHealth’s Dr. Frost, for example, says the fact that two-thirds of Medicare patients were either rehospitalized or dead within a year of discharge points to the need for both inpatient and outpatient physicians to engage more in end-of-life discussions.

Dr. Bessler says that the fact that one-half of study patients apparently had no primary care follow-up underscores problems with primary access. “Hospitalists need to make sure that expectations of how soon patients will be seen after discharge are put in their service agreements with referring physicians,” he explains.

And Omaha’s Dr. Rice adds that lack of follow-up points to the critical role that patients sometimes fail to play in their own transitional care.

“For six months, we were calling outpatient offices and setting up appointments before people left,” Dr. Rice says. “We stopped because we got feedback from primary care physicians that patients weren’t showing up.”

The issue of incentives
To address some of the bigger obstacles they’ll face in driving down readmissions, hospitalists are getting involved in bigger-picture initiatives. Dr. Rice, for example, notes that his hospital is piloting “and paying for “another transition-of-care project, one that targets pneumonia patients.

The hospital is working with two home health agencies to ensure that patients diagnosed with pneumonia receive at least one post-discharge home visit, with nurses listening to patients’ lungs and double-checking their medications. The goal, says Dr. Rice, is to bring down the readmission rate for pneumonia, one of the DRGs that results in the highest number of bouncebacks for the hospital.

In Florida, Krishan Nagda, MD, CEO of Central Florida Inpatient Medicine, is lobbying the health plans he contracts with to pay for a similar service. He’d like discharged patients, at least those with chronic conditions like heart failure, to get one home health visit to monitor their weight and medications. Readmission rates in Orlando, where his group has nearly 40 full- and part-time hospitalists working in four hospital systems, are running at 21% for discharges to skilled nursing centers.

While insurers have a great deal to gain from preventing readmissions, Dr. Nagda says, most don’t pay for that kind of patient follow-up. “Hospitals and health plans are coming back with, ‘How do we really drive these readmission rates down?’ ” says Dr. Nagda.

His group is well aware, he adds, that post-discharge phone calls, home visits and follow-up in skilled nursing facilities all yield lower readmissions and better care. “But without support from the hospital or payers,” he points out, “most independent groups like ours cannot provide these additional services.”

In fact, experts point out that under the current state of hospital incentives, readmissions are often good for revenue. The April 2 NEJM study, for instance, found that at one-quarter of the hospitals studied, 25% of all admissions were actually 30-day readmissions. The authors noted that hospitals that have excess capacity may derive as much financial benefit from readmissions as from first-time admissions.

In Omaha, Dr. Rice says that his hospital’s 30-day readmission rate in 2007 was 15.9%, up slightly from its 2005 rate of 15.6%. In 2005, he adds, Medicare reimbursed Methodist Hospital $8 million for its share of those readmissions.

Testing what works
Two pilot projects now being sponsored by Medicare show that the CMS is getting serious about both covering interventions that can reduce readmissions and removing financial incentives that promote bouncebacks.

The first project “the three-year Care Transitions pilot, designed to test interventions that can reduce readmissions ” was launched this year in more than a dozen communities.

Virtua Health, a four-hospital system based in Marlton, N.J., is serving as one site for the project. Gregory Busch, DO, medical director of Virtua’s post-acute services and medical director for one of Virtua’s four hospitals, says the project plans to take a much more rigorous approach to medication reconciliation and is instituting a personal health record, which is paper-based, for its patients.

“The idea is to get patients to write down all their discharge information with the help of a hospital staff member,” Dr. Busch says. “Studies show that when patients write it down themselves, they understand it better and there’s better follow-up.”

The project is also testing two different models of post-discharge coaching. In one, nurse practitioners provide a concierge service that would make home visits and go with patients to appointments. The other model features a “transition coach” who follows patients for several weeks and teaches them self-management techniques.

Still another project arm targets reducing readmissions from nursing home and skilled nursing facilities by teaching nurses in those facilities to use SBAR communication techniques (situation/background/assessment/ recommendation) when relaying information about medical problems to on-call physicians.

“These are medical facilities,” says Dr. Busch, “and we can treat medical problems in these facilities a lot more frequently than we do now, rather than sending them back to the hospital.”

The prospect of bundled payments
Another CMS pilot project launched this year is much more controversial: testing the use of bundled payments across episodes of care. Medicare is experimenting with this payment methodology in large part to determine how bundled payments might reduce readmissions.

The idea behind bundled payments is simple: Give hospitals one lump sum that hospitals would then split among all the providers and facilities involved in an episode of care. Because bouncebacks would dilute everyone’s piece of the pie, the thinking goes, physicians and facilities alike would have an incentive to better coordinate care.

It remains to be seen, however, whether the CMS will eventually apply bundled payments to patients with strictly medical conditions. The demonstration project launched this year is bundling payments only for orthopedic and cardiac surgeries, and only at a handful of sites.

While bundled payments are still an experiment, the concept has drawn significant attention from hospitalists.

“It appears that it is not a matter of ‘if’ but ‘when’ the CMS will institute episode-of-care payments,” says Dr. Bessler of Sound Inpatient Physicians. “All of a sudden, the hospital will be accountable for the cost of care through the rehab or nursing home process.”

While Dr. Bessler, along with many others, predicts that bundled payments may be a major business opportunity for hospital medicine, others are concerned. They worry that even more of the work related to improving transitional care and reducing readmissions might fall to hospitalists ” but that the same inequities in pay that now exist between medical and surgical specialties may be preserved.

In Florida, Dr. Nagda’s hospitalist group began covering skilled nursing facilities four years ago when it realized that some facilities had readmission rates as high as 45%.

“At this point, we have almost six physicians in 11 facilities covering about 800 patients,” Dr. Nagda says. Dedicating providers to skilled nursing facilities has significantly improved communication with the hospitalists. As a result, his group has seen readmission rates from some of those facilities cut in half.

But while Dr. Nagda’s private group would be able to take advantage of bundled payments for orthopedic and cardiac surgeries, he remains uneasy about the prospect. He worries whether hospitals will want to work only with their own employed physicians to make it easier to administer payments.

“Our concern is what will our relationship be to the hospitals,” Dr. Nagda says. “Will hospitals say, ‘We need to have every provider under one network’? Are they going to be happy with just a contractual relationship, and are we going to remain independent?”

Phyllis Maguire is Executive Editor of Today’s Hospitalist.

How does comanagement affect readmissions?

A STUDY THAT APPEARED in the April 2, 2009, issue of New England Journal of Medicine found that in 15 months between 2003 and 2004, almost three-fourths of surgical patients readmitted within 30 days came back for medical, not surgery-related, problems. According to Christopher Frost, MD, vice president of hospital medicine and clinical excellence for TeamHealth Hospital Medicine, a national hospitalist management company, those data highlight the need for better surgical comanagement.

“Often, surgical comanagement focuses on preop preparation or clearance,” Dr. Frost says. Study findings, he adds, point to the need for better perioperative care, including enhanced postop management.

And while hospitalists who act as consultants in comanagement arrangements typically leave postoperative communication to surgeons, they may want to move toward communicating medical recommendations directly to primary care providers. “This is one opportunity,” says Dr. Frost, “for significant improvement.”

For Krishan Nagda, MD, CEO of the private hospitalist group Central Florida Inpatient Medicine, high readmission rates for patients post-surgery also point to the need for better standardization and a more equal footing for hospitalists in comanagement arrangements. He points out that his group works in a large orthopedic hospital that does as many as 40 joint replacements a week.

Currently, however, there is no standardization of responsibilities between the hospitalist and the specialist when it comes to comanagement. Some surgeons, Dr. Nagda points out, like to manage many aspects of medical postop care, while others hand that management off to hospitalists.

“Comanagement has made for better outcomes, but we still see a few complications post-discharge,” Dr. Nagda says. “We need to continue to work toward standardization and developing clear roles in this model to improve outcomes and reduce readmissions.”