By Florian Quarre, Chief Digital Officer, Ciox
The military, the VA, and the Federal Government at large, are together sitting on a mountain of historical health data that needs digitizing and abstraction. Once completed, our Federal Health data scientists will have access to generations of military and Federal health insights that could reshape public Healthcare. Here’s how:
The Power of Data
Within the various arms of military health alone, there are quite literally buildings and warehouses full of legacy paper medical records that today are saved as a matter of course, but that do not garner any value. Within the Healthcare space, Government agencies operate as both payers and providers and, with this care, create reams of legacy paper data dating back generations.
The data represents events that happened in the past, however, locked inside are opportunities to learn from historical trends that may benefit the future. With abstracted historical data sets, organizations can gain insights into longitudinal facts about broad populations, can study treatment efficacies across time and place, and can better understand how Healthcare has affected Veterans and public service workers across generations.
Imagine the insights that can be gained from abstracting 50-75 years of military paper records for research: One could study the mechanism of injury, the place of injury or the likelihood of outcome – many things a researcher could want to correlate to a hypothesis is there and backed by generations of history.
Consider for example the US Air Force’s management of Airwomen’s and Airmen’s Health information critical to confirm that they qualify for flight standards as per the Medical Standard Directory. In select cases, enlisted personnel have the opportunity to apply for a waiver to keep them active in flight when not meeting standards; yet this mission critical process of gaining a waiver sometimes locks crews on the ground longer than necessary.
The USAF should digitize its 80 years of flight waivers data. Today that information is locked away in non-transferrable storage systems sitting in a warehouse on microfiche, paper, and other proprietary systems. But by allowing for abstraction and predictive analysis of how past waivers have played out, the USAF could then predict the capability of prospective pilots at present and accelerate the Air Force’s critical decisions about who flies planes in the future.
For all of these smarter, better decisions to be informed by historical data, we must first create useable data sets. This is a task that the industry has found best served by automation tools like Artificial Intelligence (AI), Natural Language Processing (NLP), and Robotic Process Automation (RPA).
The AI-powered revolution has already begun within the private sector in Healthcare. Advanced technologies are being used more widely to break down information silos, and to take the once complex, manual task of retrieving, digitizing and delivering medical records for broader analysis to make the process clearer, and more straightforward.
To unlock the power of information, leading Health data companies first utilize RPA to help breakdown the storage silos, streamlining and accelerating the retrieval of health records into one place and form where they can be handled at scale.
Second, these organizations utilize NLP to extract, standardize and normalize the health and clinical information often contained in within these medical charts. Too often, that data is written in a narrative form unique to that physician, which generates a lot of variability of data from one record to another.
Finally, AI techniques such as machine-learning and neural-networks are used to cross-correlate datapoints across a large amount of information and identify the patterns necessary to make timely decisions across a population.
These technologies work together and augment the workforce to increase the quality of records digitization, and the continuous learning of the ecosystem, making every touchpoint a learning opportunity.
Bringing Federal Health into the Next Generation
There is no doubt that smarter, faster and more qualitative systems of information exchange will be the catalyst for paradigm-shifting improvements in the U.S. care ecosystem:
- Arming doctors with relevant information about patients
- Increasing claims accuracy and accelerating providers’ payments
- Empowering universities and research organizations with research-grade data sets
- Correlating epidemics with the preparedness of field teams
- Alerting pharmacists with counter-interaction warnings
The first step to arriving at such an era in Government Healthcare, in military data sets, and across Veteran services will be to apply a systematic redesign of many operations and systems, focusing on the opening of data silos and the transformation of data to information such that it can be cleaned, augmented and enriched to the points of usability in mission critical decision making.
Ultimately, improving information exchange will help Federal Healthcare professionals avoid unnecessary deaths each year while ensuring higher-quality patient care and better decisions beyond care. Through the abstraction of discrete clinical information from historically unstructured data locked in records, a first clinical research use case for AI and NLP is born for Federal Health IT workers. However, as these technologies become standard and more accessible, it’s only a matter of time until Government organizations begin to look seriously at their historical data and ask what insights can be mined from it.