Health/Science
6:00 am
Wed May 29, 2013

Software That’s Hard on Hospital Readmissions

When you’re leaving the hospital, the last thing you want to think about is being readmitted in a couple weeks. The odds of that happening are surprisingly high. Starting in October, as part of the Affordable Care Act, more than 2,000 hospitals will be penalized for high readmission rates. Two hospitals in North Texas are trying to tackle the problem of high readmissions, with technology.

The numbers are shocking. After surgery, one in eight elderly patients ends up returning to the hospital within a month. Anne Weiss directs the the Robert Wood Johnson Foundation’s Quality/Equality Health Care Team. She says repeat trips are not only a sign of poor quality but are expensive.

“One estimate is that these unnecessary readmissions cost the government about 17 billion in Medicare alone.”

Inside the Cardiac Step-Down Unit at Texas Health Resources Hurst-Euless-Bedford, nurses wheel around mostly elderly patients with heart monitors attached their chests. Dr. Velasco is a heart and lung surgeon who is chief health information officer at THR. He says some patients need extra care after they leave the hospital to make sure they don’t, well, do a U-turn.

“Maybe you’ve heard of house calls, we do that, they visit them at home,” Velasco says, “make sure they’re taking meds, weighing themselves, all the things they were prescribed as their home regimen.”

Here’s the thing, they can’t afford to do that for everyone.  

“Sometimes what happens is there are just too many patients for us to cover,” Velasco says.  “And so there may be patients that aren’t seen and they end up back in the hospital.”

Velasco says nurses and doctors are focused on trying to keep patients alive, not predicting what could possibly derail their recovery at home. That’s all about to change. Texas Health Resources in Bedford is about to start using a new technology developed across town, at the Parkland Center for Clinical Innovation in Dallas.

Building The Model

Dr. Ruben Amarasingham has put the final touches on a mathematical model that predicts which patients with cardiac failure are at the highest risk of readmission.

The alert appears on the computers throughout the hospital unit, so at any time a nurse or doctor can see a list of patients that need special care. This predictive model, called PIECES, uses data from patients electronic medical records, sifting through dozens of risk factors. The model looks at the obvious -- like blood glucose levels, age, heart rate, but it also considers more subtle variables: how a patient is paying for care – whether there’s a history of depression – even how many home addresses a person has had in the past year.

“The health and wellness of a population and individual is driven in many cases by what we call social determinants,” Amarasingham explains. “That could be your educational status, your income level, your family environment. If you don’t have some of those fundamental building blocks it makes it very difficult to take care of your illness.”

This isn’t the first time someone’s tried to come up with a model to predict readmissions, but in a review of The Journal of the American Medical Association this one was ranked as one of the best. After Parkland Hospital started using the software in 2009, it cut its Medicare heart failure readmissions by nearly one-third, and saved an estimated $500,000. Dr. Amarasingham says more than 200 hospitals have expressed interest in using the software.

The First Partnership 

Amarasingham is first collaborating with Dr. Velasco, at the Cardiac Step-Down Unit at Texas Health Resources HEB. And the potential is enormous. Texas Health Resources is one of the largest health care providers in North Texas – with a total of more than 4,000 beds. If the software reduces readmissions in Bedford, it could be adapted to work in Fort Worth, Dallas, and Arlington.

Velasco says the key is having a shared electronic medical record system.

“All the work we’ve done to implement it here it can be easily scaled into the other hospitals at our system. It’s just a matter of flipping a switch.”

Of course, the algorithm by itself won’t solve the readmissions problem. But it will help coordinate care, and keep some patients from making a return trip to the hospital bed.

Don't End Up Back In The Hospital:

Advice from the Robert Wood Johnson Foundation “Care About Your Care” program: 

  • Ask and ask again
  • Say it back
  • Have a discharge plan
  • Manage your medications
  • Keep appointments
  • Know what to do if you don't feel well