We have the ability to improve healthcare interoperability through the use of existing technology. But we are not using it, why?
When I started my healthcare IT career in 1998, most healthcare organizations selected vendor applications based on which one provided the best workflow solution for their staff at the best price and value. This “best of breed” approach gave very little consideration to EMR integration or the challenge of sharing information between the vendor solution and the core EMR. Then, in a rapid shift, the healthcare industry swung the pendulum to selecting a “vendor of choice” for not only the core EMR but for all related applications. But in this current environment, we find ourselves with the inability to easily integrate 3rd party vendor solutions with the EMR and therefore data is trapped and innovation is stifled.
Did we swing the pendulum too far?
The best of breed approach to purchasing software from vendors was based on need and fit. Integration with the EMR was a painful journey of mapping identifiers, scheduled extracts (automated or worse, manual), delicately negotiated HL7 inbound and outbound interfaces between two different interface developers, and if you’re extremely unlucky, all of the above would apply. Understandably, after suffering through the integration challenges, the industry collectively moved to vendor of choice. Even though the application provided by the EMR vendor may not be the best fit in terms of features or workflow optimization, the fact that it tightly integrated with the core EMR system made those issues bearable. In essence, there was a deliberate trade on functionality and value in favor of ease of integration. I recall sitting in on physician advisory committees where the group admitted and concluded that’s exactly the trade they were willing to make.
In our effort to simplify our technical integration we forgot our end goal: the seamless flow of data to enable clinicians to do their job and for patients to get the care they need. The best of breed approach to software selection was actually the better approach: doesn’t it make logical sense to select a vendor that has the best workflow solution because they dedicate their entire business to solving a very real and niche problem? Other industries use the best of breed approach and thrive today. Why not healthcare?
What if there was a solution to solve the EMR integration problem? What if vendor solutions had real-time access to reading and writing data to the EMR system to supply clinicians with the data needed for the benefit of the patient? This would allow the best of breed approach to software selection to come back. Health systems could continue to select one vendor as their core EMR system of choice. They could then select any vendor solution that best meets the department workflow challenge. While EMR vendors provide a wealth of applications for the enterprise, it is impossible for one company to develop the best application for every clinical department.
The simple solution to healthcare integration is APIs. APIs allow for secure, real-time access to the database storing the patient records. Only trusted applications will use APIs through secure keys and authentication. APIs turn closed systems into open platforms. APIs are precisely the reason why when I order a product from Amazon and the UPS driver delivers my package, I am notified by Amazon that my package is sitting at my front door before the driver returns to her vehicle. There is not some magical, tightly aligned formal partnership between Amazon and UPS that makes this possible. Just as the rest of the digital economy uses APIs, this is the solution for healthcare. I understand the challenges faced by healthcare in adopting new technology because of the complexity, many moving parts and security concerns. But, if a solution is out there that can work, is affordable, safe and can deliver value quickly, shouldn’t we take a look? I believe we are ready for the next generation of integration – APIs. And then we can, once again, select the best workflow solution for clinical needs without integration as a barrier.