QI focus leads to IT solutions systemwide

Hospital system improves patient care through IT

Technology never was the starting point for quality improvement projects at Community Health Network in Indianapolis. But in making good use of it as a tool, the four-hospital network has ended up with a bang-up patient database, improved safety programs, and the potential for many of its units to end up being paperless.

It all started in 1995, when brainstorming sessions focused on how to get clinical staff back to the bedside. "One part of that is technology," explains Lynne Royer, RN, MSN, clinical director and medical informatics specialist for the network. "The goal isn’t to be paperless, or even to be safe, but to provide better care. One way to do that is to provide information to a lot of people simultaneously. But you aren’t going to buy 50,000 printers to do that. You have to find a different approach."

Getting chart information to people’s fingertips became a stated goal, says Jeff Rush, director of information for the network. That meant putting it on-line. That created an additional opportunity to build a data repository, he notes. "We wanted to capture all admissions, all lab results, all X-rays, and all dictations. We wanted nursing documentation to become a component of that information."

A small Ann Arbor, MI, company, SEC, came in as a consultant to help develop the idea. It so happened that the company had a software tool that could help. "We decided to see what they could do rather than look at the big players right away," explains Rush. "We ended up developing software together and now have data on every admit since 1996."

General Electric (GE) bought SEC two years ago, and the product, Centricity, is available for others to use. But Community Health Network remains a development partner, and is working to expand its capabilities with a dose charting and physician-order entry. At a new heart center being built for the network, the program will allow for a completely paperless and filmless environment.

The program has been rolled out to various nursing units one at a time. According to Royer, nurses and physicians either use wireless laptop computers on carts or in-room, wall-mounted devices to make their notations and access patient data. For some units, the needs are different. For instance, on the med-surg unit, physicians didn’t want to be in the room with a walking, talking patient while taking notes on a computer. So they tend to use PCs stationed elsewhere. However, in an intensive care unit, there are more devices in the room for providers to use.

The only time patient data aren’t automatically entered into the system is if the patient leaves a unit for some procedure or therapy. "If a respiratory therapist comes to the floor, then it’s documented on the computer," Rush says. "But if the patient leaves the unit for physical therapy, then it’s documented on paper."

To a nurse or physician, the need to change the way information is presented is obvious, says Royer. Episodes of care are great for billing purposes, but not for clinical care, she notes. "If I’m taking care of a renal patient, I don’t care when he came in and left. I care about his creatine levels over time." Now, clinical records can be optimized to view patient information over time.

"Nurses are often shift thinkers, and physicians think across time," she continues. "Now, I can put relevant data side by side so that clinicians can always easily see important information." For example, a patient’s neurological status might note that he was normal at 8 a.m. and 9 a.m., but at 10 a.m., pupil size was unequal. "In a paper world, that information is disjointed and hard to find. Assessments in this program are next to each other — noting that he was normal, normal, and then different. That can help clinicians determine what the problem is and what to do about it."

IOM report reinforces need

Rush notes that the administration at the network has been willing to spend money on this system from the start. "This was never about the cost, but about having to do this, about being where we wanted to be. We never wanted to be on the leading edge, but we have ended up being just that."

When the Institute of Medicine report on medical errors came out two years ago, Rush, Royer, and others involved in the program felt an even greater sense of justification for the new system. "We did a video presentation that documented true stories of mistakes made," says Rush. "We displayed physician orders and asked people if they could read them. That really helped to drive it home to people. They would start to wonder what if that was their mom, dad, wife, or child."

More than 60 presentations were made, says Royer. "We never concentrated on what we were doing, but on why we were doing it — on what it meant to patients, families, and clinical staff to have information at the ready, to not have to wait for a chart. The goal was to create enough cognizant dissonance in a room that when we left, people believed we had to do something. And if we had that, we knew that this kind of program provided the answer."

All of that made providing a return-on-investment report or defining a cost center where the benefit would accrue less important to the network’s board. "We know that we can reduce overall costs in the operating budget by using this kind of technology," Rush says. "And it also didn’t hurt that the cost of PCs was going down. But we haven’t had to show them where the savings were going to come from."

The capital investment is just one part of the cost, Royer notes. "There are soft dollars you spend through training, through having to teach some people basic skills like typing. And when you have nurses in training, you have to back-fill the staff and pay another nurse to be on the floor. But in the end, you improve the quality of care."

Figuring out how well the new system is working is based on both quantitative measurements — how many physicians are accessing the system to view patient records, how much faster it is to get a patient record — and qualitative goals. "It’s easy to figure out if people are spending less time on documentation and how many fewer people you need now that no one is always running off to find a chart," says Royer. "But qualitative goals are harder. We want to know if nurses perceive that they are spending less time on documentation. We want to look at whether consistency is improved because you have some standardized definitions of conditions. On paper, a rale to one person may be a rasp to another. But on the computer, it’s standardized."

Not-so-official signs of improvement

Rush says he still is interested in the quantitative goals, and is paying attention to whether usage of on-line documentation is increasing. However, the increased usage brings another set of problems. "If a patient comes into an emergency room and you needed his chart in the old system, that was a minimum of an hour before you would get it," he says. "Now it’s 15 seconds. But I find I’m getting voicemail and e-mail messages that it’s taking the system 16 seconds, and can’t I fix it. That 15 seconds becomes the benchmark."

Patients seem to recognize the benefits the technological improvements are making, too. Within two weeks of implementing the system in an obstetrical unit, patient satisfaction scores went up two percentage points, says Royer. Other less-scientific evidence of success includes voicemail messages from physicians telling Rush how great the system is because they can get a page about a patient in one facility, look up his or her record while at another facility, and call back with instructions without having to leave the patients to which they are currently tending.

Royer says she got a letter from the wife of a patient in one of the intensive care units "telling me how much difference it made in her husband’s care that staff knew so much about him, his condition, and were so ready for him when he was transferred from a local hospital. In a paper world, you just don’t have that immediacy."

So much to do with the data

In the future, the system will include medication-dose charting and will be able to measure near-miss and error rates, Rush says. He hopes that initial rates will be high, indicating the system helps providers at all levels to avoid mistakes. When physician-order-entry systems are on-line, Rush and Royer will look at delivery time for drugs, and turnaround time for orders.

"There is so much we can and want to do with the data," Royer says. "Maybe I’ll be able to be alerted automatically to the nine of my 17 patients that have pain scores of nine or more," she says. "I’m sure that there is a lot more mining of this great data that we can do."

It all ends up providing better and safer care. Royer gives another example about chemotherapy. "We know that chemotherapy patients are at higher risk if they are on floors that don’t regularly give it. Now we have a system where if a patient has chemotherapy ordered, the information goes into a pager so a clinical nurse specialist knows if the treatment is ordered for another floor and can monitor that situation."

The new heart hospital, which should open in the next year, will feature some technology perks that eventually will expand to other units. For instance, when they log into the system every morning, managers will be given a single screen of items they need to handle that day. "When they log on, there will be a specific view that will guide them through what they need to do that day, based on what the system is telling them has happened or is due to happen," Rush says.

And the computer program doesn’t just provide opportunities for new uses of data, either. It also can give managers data that the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) requires facilities to collect. "We can look at issues like patient restraints," he notes. "But I can do more than what they ask, too. I can look at who does it and what kinds patients are being restrained."

Before, a lot of the data JCAHO, the federal government, and other agencies required the network to collect were labor-intensive projects. Now, says Rush, it can be done quickly, and even daily, providing real-time information.

All of the nursing units are due to be on-line by 2005, by which time Rush says the system will include physician order entry, physician documentation, most charting, and possibly even specimen and blood collection.

He says such systems may be expensive — although as a development partner with GE, Community Health Network is getting something of a break on the price. But in the end, it pays off. There is no additional full-time staff needed to run it, and the benefits are obvious to the providers. "If one person is looking at a chart, another person in another unit, in another hospital can still access it if necessary."

As long as a system or facility is "rooted in the why and has a fair amount of quality improvement work as a foundation," such a program can be successful, says Royer. "You also have to make sure it isn’t driven by any one group –— either nursing, medicine, or information technology. You all have to sit around the table together." She says that when she makes site visits to other facilities and sees groups without a single nurse at the table, or where only the IT staff are present, she can be pretty sure it won’t be a successful program. "You have to create a common vision across functions. And you have to design for the different needs that different providers have. Nurses use the data one way, physicians another."

Rush has one other piece of advice: Don’t hook such a program on something such as the Institute of Medicine report on patient safety or what The Leapfrog Group says is important. "You might be successful, but you might not," he says. "That won’t necessarily hook your people, and they have to feel the incentives, too."

Both Rush and Royer also warn that such a project is not a sprint, but a marathon. "This takes a long, long time to create, and you don’t always feel like you are making progress," Rush says. "We had to sit down and actually write down what we did in 2001 to realize its impact: we brought eight units live, started on the physician order entry project with 10 physicians actively helping the program, and put through a four-year plan with a budget that didn’t include a return on investment. That is very successful. And if you don’t take time to recognize your successes, the team will dry up."

[For more information, contact:

  • Lynne Royer, RN, MSN, clinical director and medical informatics specialist, and Jeff Rush, director of information technology. Community Health Network, 1500 N. Ritter Ave., Indianapolis, IN 46219. Telephone: (317) 355-5737.]