As William Gibson famously opined, “the future is already here-- it’s just not very evenly distributed yet.”
Progressive healthcare institutions have pivoted remarkably from healthcare’s traditional mantra that it is uniquely complex and therefore unable to meaningfully apply advances from other sectors. Instead, leading institutions are actively seeking out such learnings and adapting them to healthcare. From leveraging data assets to make better decisions to more effectively engaging patients (and their support network) as customers, the results are powering rapid and impressive gains as these enterprises avoid the arduous and unnecessary process of reinventing the proverbial wheel.
In particular, the information revolution is transforming the last, greatest cottage industry into a data-driven, mission-based enterprise. Decisions based on limited information, augmented by gut, intuition, and ‘tribal knowledge’ are giving way to far more quantitative decisions based on ever-better data from real world outcomes evidence. Measured data leading to insight, leading to impact is the virtuous cycle that will power the transformation of healthcare as we begin to use what previously amounted to data exhaust to leverage ‘institutional memory’ to measure, understand and improve care. Establishing this progression could not be more important as it serves as the foundation for delivering value-based care.
Of course, as with any complex transformation, there is maturation along the crawl, walk, and run continuum. The starting point, simply stated, is integrating clinical, financial, operational and claims information at the service line level. This creates the foundation for identifying and studying care variation. Much easier said than done, but, advances in interoperability, new data exchange standards such as Fast Healthcare Interoperability Resources (FHIR) and maturation in data governance are breaking down the barriers of old.
From this baseline, additional data sources including omics, environmental, behavior and biometric can be pulled in to explain variance for a broader swath of the population. In the past, traditional disk-based database technologies were simply too slow, cumbersome and people-intensive to keep pace with explosive growth of data size and sources. This is where new technologies such as in-memory computing, Hadoop, machine learning and artificial intelligence allow us to go deeper into the data, across more sources in a truly scalable and agile fashion.
Running these new technologies requires the ability to perform broad analysis at the point-of-decision across the enterprise, essentially democratizing analytics to drive better decisions for every key clinical and administrative decision across the enterprise. This presupposes the right processes that are able to present contextually relevant information at the point-of-decision, as well as a workforce engaged and on-board with the importance of care transformation to drive patient value.
The ultimate end-state isn’t assembly-line medicine. Much of medicine is, and will remain a chimera of science and art. But, where data can lead to a better decision, whether it be applying known best practice supported by medical literature or leveraging institutional memory by looking at local real world evidence we can, and have to, do much better in applying it in a reproducible fashion across the enterprise.
About the author: As CMO for SAP Health, Dr Delaney is engaged in driving co-innovation projects focused on cutting-edge applications of advanced analytics and in-memory technologies. He is a board certified critical care physician with 14 years of practice as an intensivist at Boston’s Beth Israel Deaconess Medical Center.