How use of medical data will change the future of healthcare
The intelligent use of medical data is a foundation stone for the dramatic changes coming to healthcare. Patient treatment, doctors’ and hospitals’ business models, and the role of government in regulation and reimbursement are all undergoing profound changes, and heated debate – but the underpinning to all of this is the availability of aggregated, interoperable medical data.
According to a recent article in Information Week, healthcare industry Business Intelligence (BI) was a $600 million market in 2009, and will grow faster than any other BI vertical industry in the next 5 years. Such analytical tools working on large data sets will allow healthcare providers to be able to analyze treatment outcomes based on who provided them, what treatment options were chosen, and where they were given, among other factors. In other words – tailored, individualized best-practices medical advice can be given by healthcare providers to patients.
This is a bold vision of the future. “Expert systems” that help physicians decide on what tests to order, or what medications to preferentially prescribe, have been around for decades. They have never really been taken up very broadly, fueling skepticism about such achievements in the future. However, experience-based insights that come from analysis of very large data sets may change that.
Large healthcare organizations have been where most of the effort has occurred to date – after all, that is where the budgets to undertake such things are found. The St. Joseph Health System (a large regional not-for-profit provider) has gathered patient data from multiple Electronic Health Records (EHR) – Allscripts, GE and Meditech – along with financial, staffing and other data into Microsoft’s Amalga Unified Intelligence System.
The Amalga system contains about 43 TB of data derived from 2.3 million patients. The analysis from this is used to track trends in hospitals that might indicate a need to change how they assign caregivers, supplies, or budgets.
Will such research on very large systems ever trickle down to front-line physicians in ambulatory practice? It’s all well-and-good that such sophisticated BI analytics are used in multi-hospital systems, but what about the independent community doc just struggling to adopt EHR technology in the first place?
Clinical Decision Support (CDS) is a Meaningful Use element, and EHR technology that physicians use (either modular or full-spectrum) must address this. Currently, the CDS requirements are fairly rudimentary – they dovetail with measurements of Clinical Quality Measures (CQM – another Meaningful Use element), and adherence to certain thresholds of performance around these measures – things like getting certain lab tests regularly on diabetics, or achieving good control of diabetes based on these lab findings, for example – is part of Stage 1 Meaningful Use.
This is a good place to start, and is available to rank-and-file physicians with the emergence of modern EHR technology. Prompting to make sure that the CQM measures are achieved is a form of Clinical Decision Support: when Mrs. Smith comes in for a cold, the CQM system might note that she is due for a mammogram, and prompts the clinician to order such a study.
More sophisticated CDS will likely be built on this foundation. Appropriate medication management, for example, in diabetics with gradually diminishing kidney function based on longitudinal lab tests, could be prompted and suggested. “Best practices” based on actual experience, of course, is what drives such “expert systems of the future.”
In order for these kinds of tools to be in the hands of rank-and-file physicians, medical data needs to be interoperable, amalgamated and analyzed. The St. Joseph’s experience is a step in this direction, though it is confined to a finite, specific hospital system. Another step in this direction is Practice Fusion teaming up with Microsoft’s Windows Azure MarketPlace to support health research. The kinds of insights that result from such ventures – though currently only in their infancy – can potentially find patterns and associations that can save lives and guide treatment.
Healthcare BI is an emerging field, looking for patterns in very large data sets across many settings of care. To date, much of the work has been in large multi-hospital systems. In the era of Meaningful Use, such activity can also embrace small-sized ambulatory practices outside of such systems. And the potential offered by newly emergent web-based EHR technology may make the expansion of such research not only possible, but directly available to healthcare providers through increasingly-sophisticated Clinical Decision Support.
Robert Rowley, MD
Chief Medical Officer
Practice Fusion EMR
