Three Must-Haves for Turning Healthcare Data into Healthcare Policy
We’re awash in data.
It’s on our phones, it’s saved on our computers, our networks. It’s in the cloud.
In a relatively short period of time, we’ve gone from an analog world where data was exceedingly rare to an era where there’s literally more data available than humans could possibly process. There’s not much we do that isn’t transmitted, catalogued and scraped for insights.
We’ve arrived at a point where the collection of data is so intrinsic to our society that we’re left to wonder what we’re supposed to do with all this information. How do we act on it? What is acceptable information to collect, and what should remain private? How can we pull actionable insights from the deluge of digital details when trying to do so is a bit like sticking a tablespoon under a waterfall?
Nowhere is this dilemma more apparent than in healthcare. Our challenge right now isn’t collecting data. Hospitals track their patients’ vital signs at every turn. They (hopefully) plug details into an Electronic Health Record so there’s a running tally of improvements and declines. Wearable devices can be deployed to track health metrics outside the hospital, thanks in no small part to the recent CMS rule allowing for reimbursement of this technology.
The bigger problem lies in harnessing this data. At Ensocare, we’ve taken note of three specific ways to achieve not just accurate data collection, but to create applicable insights from the deluge and turn that into actual health policy for patients at both the micro and the macro level.
The first step in making sense of the myriad data at your disposal is to make sure you’ve developed a complete picture, and interoperability is essential in this regard.
It’s not going to be easy. Some of healthcare’s top minds are still trying to get disparate EHR systems to work together. On top of that, you have newcomers such as Garmin, FitBit, Apple and Amazon entering the health IT ecosystem. If we’re going to fully analyze the grand totality of patient health data, ALL of these systems have to play nice with one another.
We also mustn’t forget the array of data that still lies within physical media. Yes, it still exists, and for rural areas and smaller clinics whose economics haven’t made switching to an EHR a viable option, it’s just as important as data that’s on the cloud or backed up in an electronic filing system.
Not only do we have to simplify the process of getting these records digitized, but we have to make sure other healthcare institutions can interact with these legacy systems for the benefit of the patient. That’s why, for instance, we take great pains at Ensocare to make sure hospitals can communicate with even those members of our post-acute provider network who haven’t transitioned to fully digital systems.
Gathering and aligning all this data is a tall task, but we’ve already taken some pretty good steps in the right direction. The HL7 FHIR standards developed by leading health IT experts provide an excellent framework with which to approach interoperability at a large scale.
So what exactly can you do to make sure you’re set up for maximum interoperability? The more of your processes you can get online, the better. Staff members should get used to updating medical records in an online system. Setting up an easily accessible patient portal is a must, and it’s a good idea to provide portability through that portal so that your patient can take their medical record with them to other facilities if needed.
Leading EHR vendors like Cerner and Epic are doing some incredible things with interoperability right now, so keeping up with the latest developments on their blogs (as well as others in that space) can ensure you’re up to speed on what’s going on. I also encourage you to attend health IT conferences and keep your systems updated to the latest software iterations.
Whether or not we reach true interoperability any time soon, the fact remains we still have plenty of data to bide our time with. Too much for any human to sift through. And that’s where artificial intelligence comes in.
You can’t just plug the data generated by the healthcare ecosystem into a spreadsheet and peruse it at your leisure. It takes sophisticated algorithms backed by platforms like IBM’s Watson (among others) to pore through the vast ocean of data and pull out something resembling an actionable insight.
We’re finally starting to see the fruits of AI labor coming to market. Millions of outcomes can be put into an application to take note of historic trends and issue guidance, for instance, on the best potential therapy, the optimum time to hold a surgical procedure, the ideal post-acute setting for a patient and more.
Artificial intelligence is still in its relative infancy, and it will be some time before we see it deployed with any kind of frequency on the population health level, let alone the individual level. Still, the early results are very promising, providing everyone from health executives to front-of-line team members with guidance they can use to create an improved patient experience.
Finally, for all of this data to lead to changes in the way we approach patient care, you need people qualified to interpret what machine learning software provides and then put those takeaways into action.
Artificial intelligence is certainly wonderful, but you still have to marry the information it gives you with the human, compassionate side of healthcare. You need people who can take a look at a given workflow from a managerial perspective and address the unique human dynamic within any given facility or department.
An informaticist, ideally one with a clinical background, needs to turn the theoretical insights of an AI-powered algorithm into something practical. An AI may help you identify areas of the workflow that could be streamlined in order to reduce length of stay for a given diagnosis, thus creating savings for the patient and a revenue opportunity for the hospital, but it takes someone with administrative prowess to put those changes into action.
This requires an objective analysis of the data itself and careful consideration of how to approach the situation given the interpersonal dynamics of the facility. It means working with department heads, physicians, nursing staff, case managers and anyone else who may need to alter their daily regimen to accommodate what the data suggests.
Data to the Rescue
Data is everywhere around us, and it has the potential to transform healthcare.
But to get there, we have work to do. We need interoperability to ensure all crucial points of information are gathered together, we need artificial intelligence to provide a mechanism for analyzing that information, and we need informaticists to analyze the analysis and subsequently turn all those ones and zeroes into real healthcare policy.
It’s quite a challenge, but it’s one I’ve no doubt today’s healthcare IT leaders are capable of confronting head-on.
This article was originally published on Ensocare and is republished here with permission.