Many of us know someone who went to the hospital for one thing and ended up sick from something else entirely. 

Nationwide about 250,000 people die every year from what doctors call preventable harm.

In recent years, hospitals have made progress setting standards and improving hygiene that reduces infection rates. But now, a team at Johns Hopkins Hospital in Baltimore has set its sights on creating a safer ICU for the future. 

Hopkins has eight ICUs, and they are among the world's safest. Doctors and nurses follow all the best practices. Equipment is state-of-the-art. Infection rates are among the lowest in the country.

But even with all that, Hopkins ICU Nurse Rhonda Wyskiel says she’s not even sure about having her own father there. “I would never let my father be a patient without being here almost 100percent of the time,” she says.

Why would a nurse who’s worked on the unit since it opened 13 years ago feel like that?

Because mistakes are easy to make especially in the ICU.

Spend any time there and it’s pretty easy to see, actually hear, the chaos all around. There are beeps … and whirring … blasts of static … just a mix of sounds. And every one of those sounds is connected to something that is connected to someone.

A nurse monitors a critically ill patient.

But you know what …. it’s not happening.

Dr. Peter Pronovost is the Vice President of Patient Safety at Johns Hopkins, and a nationally known safety expert. “On average patients receive a little more than half of the therapies that they are supposed to. Half,” says the doctor. 

Pronovost says part of the problem is that ICUs have so many machines, and doctors and nurses have so much to keep track of, they simply can’t keep up.

Just some of the many machines used to monitor patients in the ICU.

“I need a technology that synthesizes alarms so I’m not answering a false one every 92 seconds. I need a technology that makes it easy and efficient for me to do the right thing in medicine. Not think I’m going to remember 200 things,” says Pronovost.

Instead, clinicians are stuck with individual machines that do a great job gathering data but there’s no way to connect the dots. So things get missed and people die.

There’s the famous case of Rory Staunton, a 12-year-old boy from Queens who skinned himself playing basketball and died three days later.

Joe Kiani, who runs the medical device firm Masimo Corporation, says the pediatrician and the emergency room in New York couldn’t put the puzzle together. “They collected vital signs data, like his pulse rate, like his white blood cell count. But all that data was sitting there, but only unfortunately after the boy passed away did people understand, ‘oh, my god, this boy got septic,'” he says.

Preventable errors like that kill an estimated 250,000 people every year.  Only cancer and heart disease are more deadly. Then there’s the cost, medical mistakes add up to more than $30 billion dollars a year. 

The solution seems straight-forward -- create machines that talk to each and make it easier for ICU staff to access all that information in one place.

Kiani says that’s been a tough sell in his industry. “Frankly, I think for the most part it’s been greed,” he says.

Kiani says device makers have been mainly interested in selling more machines, rather than machines that do more.

Robert Wachter is a Professor of Medicine at the University of California, San Francisco. He says companies can no longer afford that approach. “I think all of the manufacturers have begun to recognize that they can’t continue to do what they are doing if the end result is unsafe care,” he says.

Which is to say they’ve gotten the message from guys like Hopkins’ Peter Pronovost.

“We won’t buy monitors and technologies that won’t link and share data. We need technologies that will help productivity rather than hurt it,” says Pronovost.

Last month, nearly two dozen hospitals and nine manufactures including industry giants GE and Draeger pledged to start working together.

And in the meantime the Johns Hopkins Applied Physics Laboratory is getting ready to roll out what’s known as EMERGE.  Think of it as the iPad for the ICU -- a central location for patient information. With the tap of a finger medical staff can track what’s happening with patients and answer the most important questions.

“Who’s at risk for bad outcomes. What therapies should they get? Did they get them? And did they actually get well,” says Pronovost.

To bring EMERGE to life, Hopkins has a team with everyone from hospital executives to submarine engineers to front line ICU nurses like Rhonda Wyskiel.

Wyskiel says the whole goal is to let machines do what they do best -- organize information -- so doctors and nurses can do what they do best -- take care of patients. 

Nurse Wyskiel understands perfectly what’s at stake. 10 years ago her mother died from an infection that she got in the ICU.

“What if I had gotten in there and touched there and made sure she was turned every two hours. And made sure her head abed was up 30 degrees? Maybe she wouldn’t have gotten infected if I had been there more,” she said.

Wyskiel says if EMERGE works -- it would give her back all the hours she now spends running from machine to machine, responding to non-stop alarms. All the more time to make sure her patients are safe.