Jeanne-Vida Douglas
The Australian Financial Review | 07 Aug 2007

Technology managers are called upon to solve a lot of problems, although most are associated with the shortcomings of information systems rather than the shortcomings of the labour market. Nonetheless, the principal focus of the amalgamation of a series of disparate databases and technology systems at the Friendly Society Private Hospital in Bundaberg was to reduce the administrative burden on nursing staff, giving them more time to spend with patients.

"We had a lot of different systems, including a patient management system, a nursing management system, as well as databases and spreadsheets running all over the place," explains the hospital's chief finance officer, James Waters. "We really needed a product which would get all this information together in one place."

And while the end goal was obvious, the solution had proved illusive. Since 1999, the hospital had tried two separate approaches to its data dilemma, but it had been unable to amalgamate the different repositories. "First we tried internally to create another spreadsheet, which would contain all the data held elsewhere," Waters says. "We tried to build it internally, but when we came to consolidate the data it just couldn't handle the volume."

Having failed to solve the problem in-house, the hospital contracted a local systems integrator in Bundaberg to scope out the challenge, and design and implement a solution. "He spent about six months working on the project, but by mid 2004, he just threw his hands up in the air and said he wouldn't be able to complete the build," Waters said. "There were too many variables; it was just too complex to handle."

Left with a series of half-built systems and no resolution, Waters was casting around for an alternative when systems integrator Datacom offered to help. "We were sourcing some of our hardware from Datacom, and they actually came to us and pitched their services," Waters says. "We spoke to them about some of the challenges we were facing and they decided to pick up the project and run with it."

Left with a series of half-built systems and no resolution, Waters was casting around for an alternative when systems integrator Datacom offered to help. "We were sourcing some of our hardware from Datacom, and they actually came to us and pitched their services," Waters says. "We spoke to them about some of the challenges we were facing and they decided to pick up the project and run with it."

Much to Waters' frustration, however, Datacom seemed to do nothing for the first 12 months, except observe the hospital's operations and ask questions. "It seemed to be a big waste of time," Waters says. "They spent so long trying to work out what we wanted, when we just wanted to get in there and get it sorted out."

Nonetheless, by October 2005, Datacom had a proposal on the table, and by February 2006 the hospital was running a trial of the software solution, amalgamating the disparate data in a single open-source enterprise resource management tool.

"We were able to use our existing infrastructure, a Windows NT server, and a Unix box, which ran the patient management program," Waters says. "We hadn't had much experience with Linux before, let alone anything open source, so there was a bit of a learning curve for the internal IT team."

While part of the project involved transferring some of the data into a common format and then amalgamating it in a central repository, other information, such as the patient management system, was simply left in its original format. Datacom then went about designing a centralised hub, referred to by Waters as a Key Performance Indicator (KPI) dashboard, which draws the data from different sources and presents it via a web-based interface.

"One of the challenges we had was to make it very easy, so that anyone with basic computer skills could find the information they needed with minimal training," Waters says. "Because the solution is web-based anyone who can use a browser can use it, and that change process was much easier to manage than it might have been."

Strict access controls determine who gets access to what data, and many of the hospital's management procedures were written into the identity authentication process. Initially skeptical about the 12-month planning and design phase, Waters concedes it was more than compensated for by a rapid, and relatively problem-free deployment.

"It may have been a lot cheaper and a lot quicker to pick up something off the shelf, but really the cost of the software was minimal because it's open source," Waters says. "The biggest cost came from the deployment, and that was made up within two years." Not only has the solution significantly cut back on the time nurses and clinicians spend gathering data for reports, it has also enabled hospital staff to drill down into data, turning it into a predictive resource rather than a burden.

"We knew there was data that we simply weren't reacting to because it was buried," Waters says. "Now we can react to things ahead of time that we may not even have known about previously."