What should you do with your aging laboratory information system?
That’s not a simple question, and it grows more pertinent all the time, says consultant Hal Weiner, president of Weiner Services LLC, who spoke at this year’s Lab InfoTech Summit and by phone with CAP TODAY. A study he conducted showed more than 1,800 laboratories, or about 25 percent of the marketplace, have had their LIS for at least a decade.
“And they’re going to get older because two things have happened,” Weiner says. “One, hospitals are not getting funding to replace systems on the cycles they have in the past, and two, vendors have upgraded their systems to keep them current.”
But during the past two years, a noticeably larger number of vendors have been sunsetting older systems, and many of them have not kept up with the “latest and greatest” technology to continue to meet clients’ requirements, Weiner says. This has created a burgeoning market for third-party middleware products that extend the life of an LIS.
“Do you need to replace your LIS, or can you just enhance it?” he says. The answer lies in knowing your organizational structure.
If your information systems department is setting the strategies and pointing out that you have an old system that isn’t Web-based, client-servered, object-related, or written in a current programming language, and your organization wants a common vendor, that most likely means you may be replacing your LIS, Weiner says.
But if upper management encourages the lab to find the most suitable products, “you can, if you want to, go out and find other products that can integrate with your existing system and solve many of your problems,” he says.
Replacing an LIS solely because it’s “deficient in technology” typically cannot be justified, Weiner adds. “Many institutions are trying to reduce the number of disparate systems they have to support, the number of interfaces they have to put together… Economies of scale can be met using common databases.”
Also driving the need for change is the movement toward an electronic medical record system. You need to be able to extract data quickly and easily from the LIS into the EMR, Weiner says. “Older [LISs] may have a problem getting data into the format the EMR wants,” he says. “If your old LIS is incompatible with these standards, it may complicate your ability to create a comprehensive electronic medical record. But you may not have to replace it to meet those standards.”
A lab might also want to improve quality, reduce the error rate, or speed up turnaround. “This may require solutions to mitigate the potential for errors as well as enhance your existing quality assurance system,” he says.
Other questions to consider: Can the system provide adequate turnaround times to meet client needs? How often is it down? Does this have an impact on servicing clients? “On the other hand,” Weiner says, “going out and buying a new system that may be down an hour a month to replace a system that’s down only six hours a month may not be adequate justification.”
External factors can also prompt a lab to make a change. Is the lab moving toward an active outreach business to boost revenue? “Many of the older LIS systems do not have a good outreach module. So the alternative, then, is to look for third-party systems that could fill those holes,” Weiner says.
Competition too plays a role. “Many, many people are being forced by pressures from their clients to do things because their competitors are saying, ‘Gee, I can do the same thing,’ ” he says. “They want Web access to your results. Do you need a system to manage your new couriers suddenly—because that gas money adds up fast? Do you have adequate management reports to determine whether your business is profitable?”
A lab could also be facing new regulatory requirements that its current LIS isn’t suited to, such as the revised Medicare rules, HIPAA provisions, or other state, local, or federal compliance issues. “You may need to start producing file surveillance data for your local health agency,” Weiner says. “In order to do that, you’ve got to standardize and codify that data so it’s consistent. That’s going to require implementing new standards, and there may be solutions other than replacing your whole system that enable you to do that.”
Many vendors don’t care about these issues, Weiner adds. “They say, ‘Why should I pay money on something that’s not going to give me a return?’ And they really aren’t interested in making those changes… So you’ve got to look at other sources.”
Other questions to ask about external issues: How responsive is your vendor to system problems and requests for new features? Has the system been sunsetted? “That’s one way to force you to change,” Weiner says. “Some vendors have found this is the largest source of revenue opportunity for them… they’ll sunset a system and force you to buy a new one. Hopefully, you’ll buy theirs.”
With answers in hand, a lab should move forward by first drawing up a detailed project plan. Project planning, Weiner says, should include developing a team, executing a communication process, finding a way to manage and share all the documentation that’s going to get produced, and implementing a tracking mechanism to evaluate your progress. “There are plenty of tools,” he adds.
Next, a lab should take inventory of how it’s using its current system and how it would like to use it, and investigate whether the existing system can perform the new functions. Weiner extimates that about 80 percent of systems are underused. “I’m just amazed at how little functionality of existing systems [is] actually being used.” That’s because “whoever installed the system 10 years ago forgot it was there, or the vendor never bothered telling you it was there in the first place,” he says.
Labs should consider having their vendor come back in and re-demonstrate the system to uncover some of these hidden features. “You’ll be flabbergasted by what you really have already,” he says. Labs also might want to determine if they have the most current system. “Find out what the next version of your product is and what new features are out there.”
When you’ve completed the needs inventory, make sure you don’t have 17 top priorities, Weiner says. “Once you’ve got a list—what I’m trying to do, and what my needs are, no matter what the task is, whether it’s replacing the entire LIS, setting up an outreach program, or just creating a new report format—you and your staff can start doing some prioritization,” he says. More important, you can start looking beyond “What do I need to do that my current system doesn’t do now?”
The lab should set goals for each area that needs improvement. “Those objectives should be able to be evaluated in an objective manner, with metrics,” Weiner says. “You shouldn’t say, ‘I want to improve physician satisfaction.’ You should say, ‘I want to improve physician satisfaction X percent.’” And have a means to document what that level is today and how you met that goal and when.
After analyzing its needs and setting metrics, a laboratory should consider alternative solutions to meet its goals. “I believe that replacing your system with a new LIS should not be your first option,” Weiner says. The current LIS vendor or third-party software vendor may have supplemental hardware or application software modules, or both, that can expand the capabilities of the lab’s existing system. Such products can help with quality control and quality assurance, as can third-party products, like the CAP’s eLab Solutions, Weiner says.
“You can specify your quality objectives and be alerted in real time when these objectives are not being met, so you don’t have to look at each individual chart, for each individual site, for each individual instrument,” he says. “The LIS vendors don’t really have much incentive to do much in that arena,” which means that market “will continue to grow.” Many of these products provide for peer comparisons, so you can tell “how you are doing compared to the hospital down the street.”
Another arena in which third-party products have exploded is positive patient identification, which many smaller vendors have not attempted to tackle and for which they use third-party products instead, Weiner says.
Lab portal software, perhaps the most significant offering from third-party vendors from about 1998 to 2002, is “still big,” Weiner adds. “These products are now becoming a requirement for many people to be able to market their services in an outreach environment. The vendors are finding they can add additional capabilities to these products that make them more attractive to the laboratory.”
For example, Quest recently added the ability to correlate not only test results but also medication data, he says. “They can also order drugs from their local pharmacies for their patients through Quest’s lab portal.”
Probably the most significant current contributions third-party vendors have developed are add-on solutions to meet laboratories’ business management needs and increase return on investment, Weiner says. These include tools to handle courier and fleet management, specimen management and tracking, and other client services.
Middleware products also can extend the life of your legacy LIS by creating the capability to set rules for when tests should be ordered or canceled. Many systems have that capability, but for those that don’t, “those are arenas where immediate ROI is there. The day you install it, you start making money,” he says. “One of the most common examples is reflexive testing, whereby abnormal results on one test automatically trigger another test, sometimes more expensive.”
Consider as well how many hours you spend every month updating local medical necessity rules. “There are vendors out there who can provide you files… that can be imported directly into most LISs, totally removing the need for any manual intervention,” Weiner says.
Assess too whether your LIS can insert images into market reports. “A lot of the pathology systems have these capabilities, but there’s also many third-party solutions that can enable you to add diagrams, pictures, to your existing reports, or take a report from your pathology systems, add a picture to it, transmit it out, print it, however you want, to deliver it to your client, without having to replace your existing pathology system because your existing pathology system doesn’t know how to spell ‘image.’”
Vendor products also can extract reports from your system through a process called report scraping. “What if you need to get a report out of your system that your system can’t produce, but the data needed exists only in several different reports?” he says. Non-LIS vendors provide new tools that can “scrape” data from those existing reports into a formatted Access database, he says, from which a final report can be produced.
Some third-party vendors take that one step further and allow labs to extract data from one system and input it into another, already converted to the new format. “Companies can take data out of any source-file structure and stick it into any other one. What this finally does is “free the end user from being under the gigantic foot of the vendor when it comes to converting archive or current data,” Weiner says.
Finally, labs might consider replacing or enhancing their LIS because the current system is running slow, Weiner says, or because the vendor is no longer making the associated hardware. New products developed in the past five years handle those issues. But, he warns, “Never underestimate the time and resources needed to integrate any third-party product.”
The final project plan should include a realistic estimate of the resources required, timelines, and priorities, whether you’re integrating third-party products into your existing LIS or replacing it with a new one. Often, however, these steps are done after a lab has contracted for services. “And suddenly, it’s the ‘oops’ factor,” says Weiner, as in “Oops, I didn’t know that.”
To Weiner, laboratory information systems are a lot like wine. “Sometimes they improve with age, sometimes they turn to vinegar. So do you make a salad dressing?”
Yes, he says, that’s one of the solutions.
Ed Finkel is a writer in Evanston, Ill.