When it comes to laboratory information systems, lab executives says it's been a few steps forward, a few steps back.
Advances in testing devices have expanded life-saving potential for molecular testing, mass spectrometry and flow cytometry, among others, and orders are flowing like a river for labs, especially those at academic medical centers and large health systems that cater to diverse patient populations and clinical staffs
That progress, however, is tempered by the struggles for information systems to analyze and distribute results from more complex tests.
It's a similar story with regard to interoperability and results distribution. More and more large health systems are taking an enterprise approach to information systems by adopting full suites of technology from large HIT vendors. While that helps large systems streamline their data and reach more users, experts say the move can a mixed blessing for labs, which have to forgo best-of-breed lab information systems for systems that sacrifice lab-specific functionality for broader compatibility throughout the enterprise. Lab technologists and lab IT staff tend to get lost in the shuffle.
The laboratories at Memorial Sloan Kettering Cancer Center are in the midst of this upheaval, says Melissa Pessin, M.D., chair of the department of laboratory medicine. The test complexity and the volume of data they generate are exploding, she says. In addition, clinicians and other lab "customers," such as public health agencies, expect these results in ever-shorter timeframes.
"The push is on," Pessin says. "Everyone is asking us to do everything faster to ensure the clinical relevance of the results." On top of that, laboratories are now required to give patients or their designated representatives the right to receive copies of medical test results, which adds to the burden of getting test results out the door quickly even as new privacy safeguards have to be put in place
All these new pressures are being brought to bear on laboratory information systems, which are having some problems keeping up. At MSKCC, the more advanced the tests, the more challenging the task of capturing the data in the LIS, Pessin says. The cancer center's current laboratory information system can't perform the more complex calculations and analytics required to interpret a more advanced mass spectrometry or flow cytometry test, she says. To get the results into a coherent format, lab techs have to use another application to calculate the test results, then manually enter strings of data into the LIS.
"It really feels like we've taken a step back," Pessin says. "Laboratory information systems haven't changed much since I was a resident-they're designed for relatively discrete answers, and since many tests don't yield those now, it's harder to work around the LIS limitations."
<p/>Sharing actionable info
The increasing complexity of tests and the speed with which they need to be delivered create a need for more sophisticated information systems. But another market driver is the need to put a premium on interconnectedness and communication between laboratory information systems and electronic health records systems, as well as health information exchanges and online portals. Communication has become a tipping point for many health systems that are deciding to standardize their infrastructure via an enterprise IT vendor and homogenize their information systems.
At OhioHealth MedCentral Hospitals, the decision to follow an enterprise approach was based on the need to "open a lot of doors" that had previously been closed to lab data, says Eugenio Zabaleta, Ph.D., a clinical chemist at the Mansfield-based delivery system, which is anchored by two acute care hospitals.
"Does it matter that you have the 'best' lab result you could ever provide if you can't get that information into the hands of a clinician to make a medical decision or save that patient's life?" he says. "In real estate the saying goes that it's all about location, location, location. In health care it's really about communication, communication, communication."
In Zabaleta's mind, it comes down to the difference between data and information. "We produce a lot of data in the lab, big strings of numbers, and instead of spitting that into the face of clinicians we have to provide information they can act on."
An example of that approach is embedding laboratory data into the clinical decision support system at Mansfield and Shelby hospitals. In 2009, the two hospitals implemented lab decision support software, RippleDown, from Pacific Knowledge Systems. In one of the projects, the software stratifies patients at risk for heart failure according to their troponin and B-type natriuretic peptide (BNP) test results. Since 2011, every inpatient who has a BNP test performed is stratified, the risk is calculated and the results are sent via e-mail-along with patient demographics and other lab data-to clinical staff overseeing MedCentral's "Heart Success" program aimed at decreasing readmission rates for heart failure patients.
"When we're planning IT initiatives now, we're not doing lab projects, we're doing enterprise projects," Zabaleta says.
While an enterprise approach does help get health data under one roof, it also raises a host of lab connectivity issues that those involved must address. Labs aren't secondary when it comes to enterprise data. "Sometimes you have to remind people that our data is 70 percent of the medical chart," says Melissa Tone, an LIS analyst at Tucson (Ariz.) Medical Center.
Tone has participated in multiple best-of-breed to enterprise transitions during her career. What often gets overlooked during such transitions is what to do with laboratory IT employees, she says. "Something that isn't taken into account is the expertise of lab IT staff. Sometimes during a switch to an enterprise approach, lab IT staff move over to the IT department. There are pros and cons to that approach: If they move to the IT department, they're aware of the various IT projects that will have an impact on lab data, but moving them to the IT department means they're farther away from the day-to-day lab operations.
"But whatever happens," she adds, "you have to hold on to that talent."
A primary concern when switching from a best-of-breed laboratory system is that connectivity has to be re-built for an enterprise LIS, Tone says. Lab information systems are tethered to numerous applications, such as blood banks, hematology and anatomic pathology systems, as well as software for centralized instrument management.
"All these little bridges that have been built disintegrate when an enterprise LIS is brought in, so HIT leaders have to be aware that building them out again is going to take a significant amount of time."
Getting lab data more widely distributed is a big step forward. But expanded connectivity can create chronic data quality and standardization problems. "As the saying goes, if you've seen one HL7 interface, you've seen one HL7 interface," Tone says. "In the real world, there's nothing standard about those connections."
A headache for lab staff is that they simply don't know about all the systems pulling lab data, says Pessin at MSKCC. "It used to be difficult getting data to everyone who needed to see it, but now the challenge is getting the data to them in a format they can view it," she says. "Some systems can't take tables; others will truncate text based on rules that limit the number of characters. The data from the LIS is going out, but not in any uniform fashion, though theoretically it should [be uniform], since it's all HL7 messaging."
And while data is going out of the LIS, another challenge confronting Pessin and her peers is getting data into the LIS. "We can send data to the EHR via feeds, but it doesn't work as well going the other direction," Pessin says.
"Again," she adds, "this is an issue that many HIT leaders might not understand very well; we need a lot of patient data for our work, such as demographic and pharmacy information. At this point we have our staff go into the EHR and other systems and search for that information using patient identifiers. And this at a time when we're being pushed to speed up our testing and results reporting."
Register or login for access to this item and much more
All Health Data Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access