News Stay informed about the latest enterprise technology news and product updates.

SaaS-based clinical data analytics seen as next wave of automation

Clinical data analysis based on Software as a Service can help providers see which patients need attention most, provided that Web security issues are addressed.

Hospitals measure quality of care, and mark improvements by conducting patient and clinical data analysis. In recent years, software applications from such vendors as IBM and Oracle Corp. have made that time- and paper-intensive process more efficient. With the American Recovery and Reinvestment Act of 2009 (ARRA) offering incentives for provable quality improvement in a few selected health measures -- for example, diabetics’ A1c measurements -- software for clinical data analysis is becoming a more attractive purchase for decision makers, even in an economic climate forcing hospital budget cuts, layoffs and closures.

Boston software vendor Humedica Inc. has hit on a Software as a Service (SaaS) implementation that might open up clinical-data analytics automation to more hospitals that might not be able to sign up for the biggest-in-class systems.

“There’s great pressure in health care reform to improve patient quality, as well as reduce costs,” said Lynne Dunbrack, program director at Health Industry Insights, a market research and advisory services firm in Framingham, Mass. “There’s that old adage that you can’t manage what you don’t measure, and without these analytical tools … you can’t find out where your problem areas are and where to focus your quality improvement and operational efficiency initiatives."

Analysis tools enabling quality improvements at hospitals

Some hospitals keep track of quality initiatives, such as reducing hospital-acquired infections, with simple databases. They run custom reports to spreadsheets. That process is less time-consuming than paper, but it still requires intensive records review and data entry.

Hospitals collect far more usable data than they can analyze in this way “because it’s lying in a big pile somewhere, unstructured,” said A.G. Breitenstein, Humedica co-founder and general manager. “It’s a mess. It’s never really been attacked with technology that can help us make sense of that information.”

Humedica’s MinedShare helps leaders at Christus Health, a 40-hospital system based in Dallas, monitor quality-improvement data. In seconds its dashboard application gives them access to data that previously took hours or days to aggregate, said Catherine A. Mullins, Christus Health’s director of clinical informatics and patient safety.

“It’s really compressing the [time and] iterations to get answers you need,” Mullins said.

Christus is also rolling out Humedica’s MinedStream systemwide. This clinical-surveillance tool offers physicians and nurses real-time access via charts to many data sets, automating previously manual processes -- for example, determining which patients taking blood thinners are most vulnerable to or are showing signs of bleeding or dropping platelet counts -- and putting the data into a graphical interface.

“If I can come in, sit down at a computer, and say, ‘Gosh, these are all my patients who are receiving anticoagulation therapy today -- and of these 30 patients, there are three that need my attention,’ I’ve just put hours back in my day,” Mullins said of a process that previously involved manually reviewing each patient record and in many cases, comparing several sets of lab results.

Hospitals that are wired already for IBM and Oracle Corp. business intelligence software systems can plug into those vendors’ InfoSphere and Clinical Development Analytics packages, respectively. Government-mandated electronic health record (EHR) systems -- and a general drive for quality improvement -- are driving adoption of this fairly new class of health care software, Health Industry Insights’ Dunbrack said, even in a shaky economy where IT budgets are static or decreasing.

“Now, more [caregivers] are implementing electronic medical records, they now have clinical data in a digitized format that they didn’t before, so now they have data that they can electronically analyze,” Dunbrack said. “There’s great pressure through health care reform to improve patient quality, as well as to reduce costs.”

SaaS security a must for Web-based clinical data analysis

Humedica’s SaaS approach -- in which the vendor analyzes clinical data off-site and delivers the results to the customer via a secure virtual private network -- may give some hospitals pause in this era of high-profile HIPAA security breaches, when CIOs ponder the risk of letting a third-party vendor access protected health data.

Mullins said vendor access was a concern for Christus, but after her organization’s IT department vetted Humedica’s processes, the company was satisfied that patient data would remain secure. SaaS can be appealing because it requires less capital outlay for implementation and is cheaper to run and maintain, but hospitals must run any SaaS initiative through the IT staff, she said. Organizations should have IT staff members evaluate a vendor’s system, and don’t sign on until they have determined that it is secure, she added.

“I think you’re obligated to do that,” Mullins said. “You have patient data at risk, and you owe it to them to keep it secure.”

ARRA, meaningful use benefits to clinical data analysis

Systems for clinical data analysis and monitoring might not be listed chapter and verse in the ARRA, but they can contribute to streamlining a hospital’s efforts in achieving that legislation’s financial incentives for quality improvement. The ARRA also offers incentives for hospitals using clinical decision support systems. MinedStream offers some of the features those systems traditionally include, but not all of them.

If I can ... say, ‘Gosh, these are all my patients who are receiving anticoagulation therapy today -- and of these 30 patients, there are three that need my attention,’ I’ve just put hours back in my day.

Catherine A. Mullins, director of clinical informatics and patient safety, Christus Health

“I think we fall into the category [of clinical decision support] in the sense that we’re putting together clinical best-practice rules based around data flowing,” Humedica’s Breitenstein said. “Are we ‘put a bunch of symptoms in and get a diagnosis out’? No, we’re not.”

Humedica’s SaaS offerings for clinical data analysis may even have some ARRA compliance benefits toward achieving meaningful use of EHRs, Breitenstein said.

Although Christus expects to leverage its Humedica installation to satisfy meaningful use goals, it won’t be able to determine if satisfying those goals is possible until federal regulators spell out exactly how they will be implemented, Mullins said. The specific issue is whether analytics count toward meaningful use if they aren’t accessed within an EHR system and come from an associated piece of software, such as Humedica’s dashboard application.

Several traditional clinical decision support applications do connect to an EHR system, giving instant feedback on such matters as drug interactions, Health Industry Insights’ Dunbrack said.

“Decision support is often embedded directly into the electronic medical record,” Dunbrack said.”For example, a physician could be writing orders for a patient, and the embedded decision support would alert [him] that [he] prescribed a medication the patient is allergic to … or even [display a] gaps of care alert showing that a patient is overdue for an exam.”

Let us know what you think about the story; email Don Fluckinger, Features Writer.

Dig Deeper on Clinical data analytics software and systems

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.