Basit Chaudhry, MD, PhD
Medical Scientist, IBM Research
The U.S. healthcare system is capable of producing breathtaking innovations that drive progress forward. New frontiers open up on an almost regular basis. This is the “miracle of medicine.” At the same time, however, advancements made at the leading edge of science are slow to diffuse through the system and enormous inefficiencies exist in how scarce resources are used. Our ability to generate new scientific knowledge and develop advanced medical technologies has never been greater. Our ability to apply those innovations rationally in practice has not kept pace, unfortunately.
One of the major reasons for this disconnect has been the limited integration of data into the care delivery enterprise. Too often clinical decisions need to be made based on intuition and opinion alone, leading to significant variations in care and waste. Medical knowledge continues to expand at dizzying rates that push practicing state of the art medicine beyond the cognitive capacity of any individual. The absence of data and knowledge at the point of care helps create a gulf in quality between the science of medicine and it’s application in practice.
To address these issues, the Robert H. Smith School of Business at the University of Maryland and IBM organized a workshop of thought leaders to address how data analytics can be harnessed to transform healthcare. The purpose of the workshop was to explore how quantitative approaches pioneered in other fields such as operations management, statistics, economics and information science can be leveraged to help create new models for service delivery in healthcare.
While the problems of ever growing costs and variable quality in healthcare delivery are daunting. The good news is that pathways for improvement are emerging. Historically, quantitative data has been difficult to access due to low levels of information technology adoption in clinical care. This is changing rapidly. Over 40% of physician offices and 80% of hospitals have now adopted electronic health records. Large scale clinical and financial data assets are emerging as never before, opening up the possibility for transforming healthcare like industries such as retail and manufacturing have already done.
Digitizing medical records isn’t enough however. For industry transformation to occur, that data needs to be analyzed and the results of that analysis integrated into the care delivery process; knowledge needs to diffuse into clinical workflows and back to patients in a continuous process. In this way, new insights into how best to deliver care are generated, shared and amplified on an ongoing basis that will make care more convenient, higher value and ultimately more humane.
Fashioning these new data driven approaches to care delivery will be challenging however given how clinical practice has traditionally been organized. The University of Maryland-IBM workshop explored how advanced data analytics can be applied to healthcare to spur innovation in service deliver at the point of care. An example of this kind of breakthrough is IBM Watson for Healthcare. This technology can analyze massive stores of structured and unstructured medical data and provide answers to clinical questions posed in natural human language.
Rising costs and deficits in quality have made transforming the healthcare system a national priority. While the problems are significant, promising solutions exist and are already being put into practice around the country. Innovation is needed but at the same time we don’t have to reinvent the proverbial wheel. Learning from how other fields and other industries can provide a vital map for changing the value proposition that the healthcare system offers the country.
Drawing insight from data has been fundamental to the advancement of industries around the globe, creating vibrant change such as increasing productivity and improving customer service. Similarly, large scale innovation is possible in how we deliver and pay for healthcare. Analytics will be vital to this transformation as well. The lessons are there. We now have the tools to learn in healthcare as well.