Against the grain: University of Arizona Medical Center automates the organ transplant process in-house using rapid prototyping

Health Management Technology, Feb, 2005 by Sam Miller

Like agents in a Cold War novel, Kim Halloran and Dana O'Neill were practically handcuffed to their briefcases day and night, wherever they went. The two kidney transplant coordinators at the University of Arizona Medical Center (UMC) in Tucson never knew when they might get a call that could save a life as the result of a donor losing his. Each is an experienced R.N. and B.S.N, and each had to lug around a paper-filled briefcase because it was the only portable database they had of potential kidney recipients.

Today, they can access all that information from a portable, handheld pocket PC via software that is both comprehensive and intuitive--because they helped design it.

Customized to Meet Individual Needs

UMC automated organ-transplant services, one of its highest-profile clinical programs, as part of a strategy to launch a virtual electronic medical record and go paperless by 2006. Unable to find a commercially available product to meet the needs of this diverse area, which includes blood and marrow transplants (BMT), kidney/liver, and heart/lung transplants, UMC opted to develop its own system using a rapid prototyping model.

The result was TransTrack, a Web-based transplant system tailored to user needs and integrated with mainstream UMC clinical systems. TransTrack provides an architecture and visual integration platform that helps physicians and nurses manage patient demographic information, laboratory results, insurance information and patient scheduling while adding transplant-specific information that is not available in the core systems.

It is also customized for separate transplant programs, each of which includes different scheduling demands, patient acuity, experimental nature and treatment.

The new system also provides analytical tools, enabling clinicians and managers to view patients according to stage of transplant and other factors, It's a homegrown system for both the clinical care and research needs of a university.

"We're going against the grain of a healthcare industry that has moved away from in-house development because of the perception that it's too expensive and too difficult to support," says Mark Nahirny, a software engineer and chief architect of TransTrack. "But that view discounts a big advantage of in-house development: People get exactly what they want. It's not a generic solution, which is especially important with high-profile services like organ transplants," he says.

Using a rapid prototyping process that allows user involvement in the development process, UMC completed its BMT component at the end of 2003, its kidney component in 2004 and expects to complete heart/lung by the summer. User acceptance has been high, largely because users were involved in system design from the beginning.

Importance of Human Factors

"TransTrack really helps when you're on call at 2:30 a.m. and get an offer of a kidney from the Donor Network of Arizona," says Halloran, because it contains "every particular" necessary to track down potential kidney recipients and their physicians.

"You only have an hour to look up patients who are potential recipients, call the doctor and the transplant surgeon, give information on all the potential recipients and make calls back to the organ procurement organization," she says. The pressure arises from the fact that the "cold ischemic time"--the actual time a kidney can last from its removal from a cadaver and placement in a live patient--is only eight to 10 hours, and the window for aligning all the players is a tenth of that. It's critical in the first hour to get all the players on board.

Nahirny says the rapid prototyping model encourages user involvement in a way vendor-based development rarely does. At its basic, rapid prototyping is an interactive process that relies on getting a basic software prototype in front of users quickly for feedback with the goal of producing successively better versions. "The best way is to develop a system that's an approximation of the environment and let them try it out," he says.

Human factors are critical and that's a big reason why building a system in-house can be so effective, he explains. "Putting a system in place changes the environment. Doctors no longer interact the same way with nurses because they're interacting with the computer. You can document very closely how an assembly line works, but this is not like that at all.

"If you ask doctors or nurses what they want in a computer system, they don't know. It's an unfair question. They're focused on patient care and can't picture that kind of abstraction, it's not until you actually give them something to use that they can say, 'Oh, yeah, that's kind of what I want, but it needs this or that,'" he explains.

Lofty Goals

The primary goals of the TransTrack project were to:

* provide a transplant-specific view of patient data;

* capture data that could not be captured in traditional hospital information systems due to the transplant-specific nature of the data;

* create a system that parallels the existing processes followed in transplants;