APIs in Healthcare – Magic in the Air

Our modern lives are powered by APIs — software tools that allow different systems to talk to each other — and they’re so embedded in our routines that we rarely think about them. We check our bank balances, flight times, music queues, and weather forecasts in real time, all from our phones. Behind the scenes, countless software programs and servers work together via the internet to fetch the information we need and deliver it instantly. These interactions typically rely on RESTful APIs (a common web service design style) and transmit data using JSON, a lightweight format for structuring information.

Healthcare, however — arguably the most vital (pun intended) of all industries — doesn’t function with the same real-time, seamless functionality. While there are pieces in place — such as APIs that pull lab and imaging results into EHRs, claims processing pipelines for financial data, and custom APIs that enable cardiac and vital sign monitoring in hardwired intensive care environments — we’re missing rich integration between our personal health (which we can increasingly monitor ourselves, thanks to the quantified self-movement) and provider systems that can digitally monitor and respond to that data on a continuous basis.  

Most of us could use a little more support when it comes to managing our health. One example of more advanced, API-enabled care is the integration of continuous glucose monitors, insulin pumps, and, at times, the EHR for patients with insulin-dependent diabetes. Other chronic conditions such as obesity, hypertension, coronary artery disease, asthma, COPD, depression, and vascular disease could all benefit from APIs that enable seamless data sharing between individuals and smart servers.  

So, why hasn’t this kind of seamless, API-powered experience extended more broadly across the US healthcare system? It’s not a question of technical capability. The issue is that our healthcare infrastructure is still based on “bricks and mortar” / in-person visits and procedures. The entire business model incentivizes services delivered within the walls of clinics and hospitals, not the kind of continuous digital care that APIs make possible.

While value-based care is often championed as the future, it remains largely theoretical. And futures don’t come with billing codes. René Descartes famously said, “Cogito, ergo sum” — I think, therefore I am. In American healthcare, it might be more accurate to say, “CPTo, ergo sum” — I code, therefore I am. (Your mileage on translation may vary.)

More importantly, our digital anchor point in healthcare is the EHR, a tool purpose built for documentation, not automation. When one thinks of the need to monitor “state” in computer science terms, that is pretty much exactly the type of computation needed to understand and respond to a patient’s health status in real-time. Arguably, the entire cloud computing world is about bringing such sophisticated computing directly to APIs — an API-first strategy.  

Ultimately, to unlock the benefits that APIs have brought to every other industry, we’ll need to rethink the incentive structure that governs how care is delivered and paid for. Right now, many health systems and EHR vendors operate like monolithic fortresses where connectivity with the outside world is feared as a door for “patient leakage.” But, APIs shouldn’t be seen as a threat to healthcare delivery, they should be viewed as a gateway to delivering smarter, more responsive care. We should rethink APIs.  Healthcare would not be the first industry to have its “brick and mortar” existence upended by competitors providing a different mix of digital services.

What might an API-first world look like for providers and payers? Picture each patient with a computational model of their health where each treatable parameter (like weight, blood pressure, heart rate, exercise, inflammation, and blood chemistry) serves both as a marker for diagnosis and as an opportunity for treatment. Many of these metrics, and the chronic conditions associated with them, could be monitored and managed via smartphones and connected devices.

While this may sound futuristic, it’s worth noting that most large online retailers use the same type of approach to interacting with their customers. The HL7 FHIR clinical data transmission standard finally allows us to have a uniform, computable format for representing medical records, making this kind of intelligent data exchange possible at scale.  

There are also near-term opportunities for APIs to improve care, especially in acute settings where costs and stakes are high. This is where the money is being spent right now. Think of a patient with congestive heart failure, on dialysis, or undergoing chemotherapy. These are people whose conditions could be monitored and managed more effectively in between visits, avoiding costly ER visits or hospitalizations.   

Payers, too, have access to much of the same patient data (or could obtain it via FHIR APIs) and are increasingly incentivized to act on it, especially those who are operating under capitated risk models. The new HHS and CMS leadership has already said multiple times that they will be moving to more digital models of care — and APIs are at the heart of that. As healthcare moves toward modern tech stacks and smarter infrastructure, care management is poised to look very different.

Photo: chombosan, Getty Images

This post appears through the MedCity Influencers program. Anyone can publish their perspective on business and innovation in healthcare on MedCity News through MedCity Influencers. Click here to find out how.