Home  >  Analytics   >   Analytics Algorithms as Microservices

Analytics Algorithms as Microservices

by Brian Murphy | May 12, 2017

This is a preview of premium Chilmark Advisory Service subscriber content. If you are a CAS client and would like to read the rest of this monitor, please login here to access the file.

Key Takeaways
  • Microservices architectures have enabled significant reuse in many industries yet remain unknown in healthcare.
  • HCOs and HIT vendors want to increase user access to analytics and make broader use of existing analytics in other applications.
  • Microservices are one way to implement the emerging analytics algorithms now being discussed by HIT analytics vendors.
  • Progress will depend on a better understanding of modern ideas about developing and maintaining distributed, cloud-based applications.

HIT analytics take a monolithic approach to development. The fruits of this labor – large catalogs of reports and dashboards – serve many valuable purposes for a range of users. They can be a revelation for clinical and administrative users who are relatively new to using applications day-to-day. On the other hand, the size of these catalogs can also be daunting for many users, restricting the utility of the underlying data and analytics. Moreover, the metrics and analytics that make up these reports are of little use to other applications.

Several vendors point out that analytics is great at identifying problems but that practices struggle to address them. It doesn’t help that most of these analytics are delivered through a portal, outside of day-to-day workflow. Today, microservices, DevOps, and open APIs are making it possible to integrate data from disparate silos into unified workflows. These ideas are potential solutions to the functional and usability ills of healthcare’s various systems of record and systems of engagement. HCOs want to change their application portfolios to derive value from the scale, flexibility, and cost benefits of modern distributed applications.

Microservices – distinct application modules responsible for a single operation within a larger application ecosystem – have become an important way for enterprise and consumer developers to think about applications.

They support code reusability in a way that monolithic and prior service-oriented approaches to application development did not. No formal standard or definition of microservices exists, but certain characteristics mark the development style. The approach delivers a collection of independently usable, small functions – microservices – distributed over loosely coupled organizations and computing environments. IT organizations access microservices through APIs wherever they may reside: on-premise, remote-hosted, cloud-hosted, or across hybrid environments.

Microservices are not well known in healthcare but most people have interacted with a business whose applications have been implemented as microservices. Netflix, as part of its transition to the cloud, rewrote its entire application portfolio as microservices. Amazon also has invested heavily in building microservices-based applications. While these companies were motivated by the need for web scale, they also found that microservices helped make their applications more maintainable.

If you are a CAS client and would like to read the rest of this monitor, please login here to access the file.

Stay up to the minute.

“As biometric data becomes cheaper and easier to collect through smart sensors, devices, and mobile apps, expect to see more innovations in consumer health.”

-Alicia Vergaras