As more healthcare systems look to embrace big data and the benefits analytics-driven care strategies could bring, the need to build EHRs that can communicate patients’ health data without hassle is growing with each day.

But despite growing awareness and billions in expenses, the issue of inadequate interoperability remains the albatross around healthcare IT’s neck, and one that is having a substantial impact on those working the front lines of care.

“At this point, clinicians are really getting overwhelmed with the fruits of our labors in the interoperability space,” Dr. Steven Lane, clinical informatics director of privacy, information security, and interoperability at Sutter Health, said. “There’s data coming at them from multiple sources in different formats, it’s unreliable, it’s inconsistent. It needs to be normalized and reconciled and analyzed, and then presented in a way that physicians can actually make use of it, because clinicians are either disconnected, the old lament, or they’re overwhelmed, which is the new lament. When you’re overwhelmed, the reason is because the data isn’t being curated and refined and presented in your workflow in a way that you can really make use of it. And that’s where I think AI and machine learning and APIs can really help.”

At the Health 2.0 Annual Fall Conference in September, Lane will be joined by Matt Menning, director of the American Medical Association (AMA)’s Integrated Health Model Initiative (IHMI); Dr. Anthony C. Chang, chief intelligence and innovation officer at Children’s Hospital of Orange County; and Eric Louie, chief medical officer at Healthbox, to discuss how interest in AI and other data-driven technologies are reaffirming the need for health system interoperability.

Speaking firsthand as a clinician, Lane emphasized the benefits well-implemented technologies such as AI or machine learning could bring to beleaguered health workers. Taking that idea one step further, he also had thoughts on how novel consumer-focused records platforms, such as Apple’s personal health record feature, could interact with providers’ EHRs and provide a more complete care ecosystem.

“Apple Health is figuring out how to grab a subset of clinical data based on the Argonaut profiles from EHRs — multivendor, across the country — and pull it into Apple Health. Well, once it’s there, let’s get to work on it,” Lane said. “Let’s solve the problem of medication reconciliation. “Let’s do care gap management and advice for patients. Let’s not only figure it out to display it to the patient, but turn it around and push it back to the provider. If a patient downloads their data from five different organizations and then they’ve got an app that lets them do their own [medication reconciliation] — for goodness sakes, let me take their reconciled med list, pull it into my EHR, and just replace the med list that I have.”

But while flexible platforms and new technologies certainly have the capacity to improve on-the-ground care, they’re also a necessity for more ambitious care efforts. For instance, integrated health models such as those Menning focus on using data collected throughout the continuum of care to guide research and targeted interventions. However, being able to use AI to analyze these data is only possible with data that is both clinically relevant and uniform.

“The data is not standardized. It’s not represented consistently, and if it’s not narrowed to the meaningful data, you can’t really trust the information that you’re analyzing or that you’re running algorithms or AI on,” Menning said. “You can have a wide variability in the data, and that’s going to lead to a wide variety in your analytics. I guess another was to say it is that big data is great, but the challenge is moving from that big data to small data that actually makes a difference to individuals’ health.”

Menning, Lane, Chang, and Louie will be discussing these challenges and others at the “AI and New Initiatives Driving the Need for Interoperability” Provider Symposium Sept. 16 at 12:30 p.m.