Physician assistant intervention helps reduce readmissions and costs following cardiac surgery

Following up with patients at their homes soon after cardiac surgery and providing them with individualized care may help patients improve their health—and help hospitals save money at the same time.

An intervention at Staten Island University Hospital in New York in which physician assistants visited patients’ homes on day two and day five following hospital discharge for cardiac surgery decreased the rate of readmissions by 41 percent. The hospital also saved $39 for every $1 spent for the intervention.

John P. Nabagiez, MD, PhD, a thoracic surgeon at Staten Island University Hospital, presented the results at the Society of Thoracic Surgeons annual meeting in Phoenix on Jan. 25.

The hospital implemented the program in 2010 for all adults following cardiac surgery. Physician assistants who worked with patients during their hospital stay have a checklist of things to do during the home visits, including checking blood pressure and heart rate, listening to patients’ lungs, examining for edema and making sure patients are doing deep breathing exercises and staying active. Before the hospital implemented the program, patients would be sent home and have a nurse visit them.

“This cannot be outsourced to physician extenders who do not understand the critical nature of these patients,” Nabagiez told Cardiovascular Business in a telephone interview. “That’s the key caveat to making this work. I’ll be honest. I was very skeptical at the beginning. I didn’t buy into it. But I’ve officially drank the Kool Aid. I believe in it. It works.”

This analysis examined 1,185 patients who underwent cardiac surgery at the hospital between September 2008 and August 2012. The researchers compared the two years before the program began with the two years afterward.

Previous research found that between 8 percent and 21 percent of patients undergoing cardiac surgery are readmitted within the first 30 days of discharge. Of those patients, half are readmitted within three weeks of the procedure.

The current study examined transitional care, which Nabagiez defined as the care given between discharge and the first office visit. During that time period, physician assistants or nurse practitioners would usually call patients and attempt to coordinate care, ensure timely follow-up and improve medication compliance. Nabagiez said other hospitals have also had physician assistants visit coronary heart failure patients at their homes, however, nobody visited cardiac surgery patients before Staten Island University implemented its program.

“A question that often comes up is, ‘Why not just call [patients]?,’” Nabagiez said. “It’s a big difference when you go into their native habit. The [physician assistants] see things that we would never expect. There are patients that you know when you discharge, they really don’t have control of what’s going on and they’re very shaky. But then there are the ones who you think have everything together, but you show up and they don’t know where their medications are, they didn’t fill some scripts, they’re confused about what to do, they’re not doing the exercises you’re telling them to do. Reinforcing that was very important. It’s so much more than you can get out of a simple phone call.”

For the current study, the researchers examined two years of home visits versus two years of no home visits. The patients in each group were not statistically identical, so the researchers performed propensity score matching and selected 17 covariates that are markers for issues following operations.

In all, there were 1,185 patients discharged to their homes during those four years. Of those patients, the researchers found 726 patients whom they could match. Thus, there were 363 patients in each group.

When examining the full cohort of 1,185 patients, the readmission rate was 16 percent in the first two years when physician assistants did not make home visits and 10 percent in the following two years when physician assistants visited patients’ homes. After propensity score matching, the readmission rates were 17 percent and 10 percent, respectively.

For the full cohort, the length of hospital stay was 5.7 days in the first two years and 6.2 days in the following two years.

The mean cost of readmission was $39,100 for the two years without physician assistant visits and $56,600 for the two years with visits. After eliminating the most costly and least costly readmissions, the mean cost of readmission was $21,100 and $31,100.

Despite the increased cost of each readmission and the cost of sending physicians assistants to patients’ homes, Nabagiez said the program saved the hospital $952,214 overall during the two years of the intervention thanks to the lower readmission rate.

“In our hospital in Staten Island, it’s a suburban community and it’s doable,” he said. “In a major medical center that covers an extremely large geographic area somewhere in rural America, it may not be feasible. It certainly would cost more in terms of time away from the hospital and gas and all that stuff. It’s less of a savings in a rural setting…It depends on the setting. It might be even cheaper in Manhattan where you’re going down the block or going up and down buildings. For our area, it works.”

Nabagiez said the main limitation was determining the cost of readmissions. He based his calculations on billings and not reimbursement, so he admitted that his numbers may not be completely accurate. However, even if Staten Island University's reimbursement was a small fraction of its billings, the program would still likely save the hospital money.

"There’s no math I could do that could not make this cost effective," he said. "In fact, I would take it one further by saying even if it’s cost neutral—if there’s no savings whatsoever—you are still doing the right thing for the patient. You’re giving them optimal care. By keeping them out of the hospital, we are optimizing their health and their satisfaction to such an extent that you could argue that it would be worth paying to do this.”