Information technology (IT) is playing an ever greater role in healthcare, from helping doctors improve outcomes by tailoring treatment to an individual patient through DNA sequencing to developing new and more effective drugs with fewer side effects.
But there is another important, though perhaps less visible, way in which IT is being used to improve the efficiency and quality of healthcare. Healthcare providers desperately need tools that will enable them to sift through the huge mountains of patient-related data in order to discover insights, make better informed decisions and, on occasion, detect fraud.
Healthcare is the classic ‘data-rich, information-poor’ business. It is a Big Data problem that impacts almost every type of organization — including the hospitals, insurance companies, medical groups and dental practices that make up our healthcare systems.
Meanwhile, healthcare providers everywhere are struggling with rising costs, an aging patient population and the need to deliver continuous improvement in the quality of care. Like other businesses, healthcare providers are being asked to do more with less. Harnessing the power of Big Data analytics and the speed of in-memory processing may be the only way to deliver the insights that medical professionals and administrators need to bridge the gap between healthcare demand and supply.
Ultimately, the competitiveness and viability of healthcare institutions may well depend on their successful adoption of a data-driven strategy.
Many of the latest innovations in the application of IT to healthcare are being driven by small, entrepreneurial companies including those involved in the SAP Startup Focus program. Some, like PHEMI Health Systems and GaussSoft are using the power of SAP HANA to deliver innovative predictive analytics systems targeting healthcare providers specifically.
“As healthcare organizations move to value-based care to lower costs, drive quality and improve outcomes, they become critically dependent upon access to high quality information,” explains Adam Lorant, PHEMI’s VP of Marketing & Product Management.
“An information chasm exists in healthcare organizations (HCOs) today,” he adds. “It’s time for HCOs to move beyond analyzing claims data alone, and truly embrace the depth and breadth of clinical data in a hospital to derive new insights.”
Vast amounts of data are generated as a result of medical tests and examinations ― including patient records, radiology images and digitalized lab results. The problem is that typically this information is locked away in information silos. Some healthcare organizations store data in literally hundreds of different and incompatible information systems.
The Vancouver-based startup teamed up with SAP to address this problem and build what it calls the PHEMI Central Big Data Warehouse. The system uses Big Data techniques to unlock information trapped in non-relational and unstructured data like x-ray images and doctor’s notes, across an organization.
“More than 70 percent of healthcare information is unstructured and difficult to mine for relevant insight using traditional methods,” says Paul Terry PHEMI’s CEO.
PHEMI’s software converts this unstructured data into structured data while managing information privacy, security and governance. PHEMI Central can handle hundreds of petabytes of historical data, converting Big Data into what it calls “Small Data” and feeding the most frequently accessed information into SAP HANA. Then, using predictive analytics tools, clinicians can calculate risk scores, develop a patient care plan and minimize the costly need to re-admit a patient.
PHEMI claims the system is up to 60 percent cheaper to run than a traditional data warehouse system and that new data can be added 35 percent faster while minimizing the risk of data leakage – crucial in the highly regulated healthcare sector – because privacy, security and governance features are baked into its design from the outset.
GaussSoft also uses SAP HANA as its primary database and warehouse, tapping into the mass of financial data across a healthcare enterprise to understand what drives profitability and enable users to make decisions that improve the bottom line. Instead of relying on averages or aggregated costs, GaussSoft’s technology enables healthcare providers to examine the actual costs incurred in delivering patient services and the profit or loss associated with a particular patient outcome.
This is particularly significant when you consider that a hospital with perhaps $1 billion in annual revenues may have 25,000 different charge items and perform two or three hundred different procedures.
“Today a hospital’s top-line revenue generation emphasis is giving way to a bottom line focus driven by healthcare value, and reduced reimbursements,” explains GaussSoft’s Hal Daseking. “Increase the number of procedures to increase revenue and the bottom line seemed to follow. Average and aggregate costs based on revenue models were good enough. Now a healthy bottom line requires a painless way to determine more precise margins.”
Using the company’s Profit Analysis and Sythesis (PAS) for Healthcare software, Gauss Soft compared the realistic cost (NFC) for each of 200 typical hospital procedures, to the traditional ‘average RVU’ cost. (Relative value units (RVUs) are a measure of value used in the US Medicare reimbursement formula for physician services.)
It revealed that the traditional method for calculating costs masked losses in 40 percent of procedures and was unable to precisely measure margins (off by typically 35 percent) making managing the bottom line next to impossible. By using GaussSoft PAS, the company claims one large hospital group increased their margin by 40 percent over three years while and another hospital achieved a positive total margin (from a loss) in six months.
Such results may not capture the headlines in quite the same way as the development of a new breakthrough drug. But the health of our healthcare systems and our ability to continue to improve patient treatment may well depend on harnessing IT and Big Data analytics in particular to detect hidden patterns in data, predict healthcare outcomes and improve efficiency so that more patients can be treated.