Baylor Medical Center Garland: HF Program Works, But Cost Questions Remain

Heart failure (HF) care programs frequently are disjointed and uncoordinated, resulting in higher hospital readmissions and longer length of stay. To prevent unnecessary readmissions, staff at the Baylor Medical Center Garland (BMCG) in Garland, Texas, studied the effectiveness of a nurse-led transitional care pilot program designed to synchronize management for this patient population and help avoid readmissions. The program did just that by reducing 30-day HF readmissions almost 50 percent. While programs such as these achieve quality outcomes, they may not be cost-effective.

Choosing the right pilot

Research has shown that HF readmission rates range between 10 to 19 percent two weeks after discharge and skyrocket to nearly 50 percent within three months of discharge (Arch Intern Med 2011:171[14];1238-1243). U.S. hospital readmissions contributed to the $33 billion bill for HF costs in 2006 alone.

Due to the 23 percent readmission rate at BMCG, a high-volume hospital that sees approximately 300 HF admissions per year, staff at Baylor searched for a cost-effective, practical transitional care model to curb the potentially preventable obstacle.

"We were not at our desired level as a health system so we needed a different approach," says Brett D. Stauffer, MD, of the internal medicine/hospitalist department at the Baylor Health Care Center in Dallas. "To do so, we began looking at different care models to help reduce hospital readmission rates."

After much searching and planning, staff at BMCG, a 263-bed hospital, decided to test an Advanced Practice Nurse (APN)-led transitional care program in an attempt to reduce 30-day readmission rates, costs and length of stay. If the pilot worked, the system might later roll out the program to the rest of the 27-hospital system.

The intervention consisted of a three-month APN-led transitional care program where the initial APN home visit occurred within 72 hours of the index hospital admission. In addition to a full-time APN, the program had to be run in conjunction with a licensed physician to oversee the APN's work. (According to Texas law, APNs cannot act independently.) The study population included 140 Medicare HF patients who were 65 years or older and discharged from BMCG between Aug. 24, 2009, and April 30, 2010.

Along with overseeing the APN, the licensed physician managed back office billing and support for documentation in the EHR. The program, implemented in collaboration with the HealthTexas Provider Network, also required remote access equipment to facilitate house calls and hospital-to-home transition of care.

However, even more important than hardware and software was finding a dedicated group of staff—nurse practitioners, physicians and administrative staff—to make transitional care models such as this one work, Stauffer says. "The success of our program had the most to do with the staff who ran the project."

The $1 million question: What will it cost?

While the hospital realized a 48 percent drop in 30-day HF readmission rates, the impact of the program on length of stay and 60-day direct costs was rather minimal. Therefore, support from senior leadership is imperative to these programs because while they help patients, says Stauffer, they also may not be profitable.

Stauffer says that the facility realized a cost improvement after the first hospitalization because of a reduction in length of stay, but the overall system lost money. "Payors, whether Medicare or others, need to reimburse and support hospitals in a way that health systems can be rewarded for implementing these types of interventions," Stauffer says.

The costs are manageable with this program, says Neil S. Fleming, PhD, vice president and chief operating officer of the STEEEP (safe, timely, effective, efficient, equitable and patient-centered service) Global Institute at Baylor in Dallas, which disseminates best practices to external organizations. The fact that the hospital receives reimbursement for home visits offsets some of the direct costs of patient care.

"In addition, there are one-time fixed costs to set up this type of program," says Fleming. "Going forward, we will move past this initial phase when evaluating the impact the program has on the patient."

Ultimately, costs for each Medicare patient were reduced by $227 due to the dip in readmission rates seen with the program. "The first time you roll one of these programs out, it will always be more expensive," he says. This is why it is imperative to perform a pilot test of the program to identify challenges and solutions and outline a strategic plan to move forward.

Baylor Health Care System (BHCS) as a whole saw higher 60-day direct costs for an HF index admission prior to intervention compared with BHCG, which saw $1,251 less on average than the system average for BHCS. However, Fleming says that post-intervention costs were less at BMCG. Additionally, total 60-day mean direct costs decreased nearly 10 percent at BMCG, $5,729 to $5,176.

But what is the bigger picture? Currently, health systems are compensated very little for the transitional care period, Stauffer says. "These sort of siloed payments that only reimburse hospitals when you have a patient in your bed are not effective. Rather than pay for these episodes of care, payments should be robust enough so that health systems can continue working on the larger picture and implement these types of programs."

In fact, whether the program can be expanded to other hospitals within the health system will depend on the future of the reimbursement model and whether hospitals will begin being compensated for value rather than volume, he says.

Hospitals should be poised to address health issues, including HF transitions, Stauffer says; however, they must simultaneously be able to cover their costs and maintain adequate margins.

Long, winding road

BMCG realizes major benefits from the program—a large drop in readmissions, increased satisfaction of medical staff and patients, increased resources and a cost savings per beneficiary—but there also are many hurdles.

Fleming views different payment systems as potentially more effective. "Bundled payments may be a more efficient reward system," he says. "Existing acute care models, such as ours, are not currently incentivized. And also there are additional costs involved—the costs of the advanced practice nurse."

One challenge is getting administration on board. However, Cliff Fullerton, MD, vice president of chronic disease and care redesign at BHCS, says the 48 percent drop in readmission rates and increased patient satisfaction will help.

But nothing comes easy. "You can't turn these types of programs on in a few months, as it takes a long time to embed them," says Fullerton. The entire process took almost a year to roll out and the hospital won't have data for at least another year past implementation.

Since the pilot, Baylor has rolled out the program to two hospitals and is working on a third. Additionally, Fullerton says the next step will be to deploy a risk stratification model that includes various types of interventions and diseases. The interventions will depend on a patient's risk of readmission, says Fullerton. For example, low-risk patients would receive follow-up phone calls after discharge, while medium-risk patients would receive telemonitoring and nurse-led follow-up calls. High-risk patients would be enrolled in APN-led transition programs.

"The effectiveness of the APN-led TCP [transitional care program] provides further evidence that hospitals have the tools to reduce readmissions for HF patients, but payment policy must be amended to align incentives for hospitals to produce the highest value by reducing problems during transitions in care," Stauffer concludes.