The Health Insurance Microservices Revolution

By Nathan Hackley, Co-founder, Sana Benefits And Will Young, Co-founder, Sana Benefits

Nathan Hackley, Co-founder, Sana Benefits

In 2001, Amazon had a big problem. Their core application was getting larger and more complicated as they scaled. Launching new features required increasingly extensive code review and their momentum as a company was slowing significantly. How, they wondered, could they maintain a fast pace of innovation while safeguarding application reliability and security?

Amazon’s answer was microservices. Instead of maintaining one “monolith” application, Amazon engineers broke the site down into small single-purpose applications with well-designed interfaces that communicated with each other. For instance, one micro-application rendered the buy button, another calculated sales tax, etc. Product teams could add features quickly without worrying about crashing the whole site. It worked so well that Amazon eventually turned their solution into one of their most successful products - Amazon Web Services.

Health plans today largely function like Amazon’s web application in 2001 - as inflexible monoliths. The different “services” of a health plan—benefit design, underwriting, claims administration, pharmacy benefits, care management, etc—are so deeply interdependent that it is difficult to innovate in one domain without breaking another. The result is innovation gridlock.

This stasis is immediately apparent to any web developer that touches health tech infrastructure. COBOL (designed in 1959) and EDI over FTP (developed in the early 70s) are common technologies used in daily operations across health insurance. These technologies are so obviously inferior to their modern counterparts (Ruby, JSON, APIs) that the most developers react with revulsion when they have to work with them. The most popular open source tool to interpret EDI for Ruby applications is called “StupidEDI”, making clear the developer’s feelings.

In order to accelerate the pace of innovation, the industry’s technology needs a total overhaul. As with Amazon, the solution is a microservice-like plan architecture. The monoliths of today are simply too big and complicated to change all at once, but if you carve out individual services, carefully define their interfaces and iterate on them independently, you can systematically improve the product piece by piece. This is how the health plans of the future will work.

Will Young, Co-founder, Sana Benefits

We are talking about more than just technology, of course. A successful microservice architecture requires alignment on the business activities across service areas. With health insurance that means aligning operating processes (underwriting, claims processing, etc) and vendor contracts with a pseudo-microservice architecture. This is a difficult organizational challenge.

Fortunately, there is an environment that is well suited to enable this kind of innovation: self-insured health plans. The self-insured ecosystem already has a wide range of vendors across different service areas with well-defined contractual and operational interfaces between each other. We are already seeing the result: a Cambrian explosion of innovation in self-insurance.

At the core of the revolution is enabling infrastructure. Amazon and others have made the implementation of HIPAA-compliant web applications relatively straightforward. On top of that infrastructure is an array of generic HIPAA-compliant web services for chat, fax, mail scanning, etc. More specialized enabling technologies include the likes of Noyo, which helps modern web frameworks integrate with legacy infrastructure, and Ribbon Health, which provides provider search and analytics as a service.

Supported by those fundamental building blocks, entrepreneurs are free to build fully featured plan services. Tech-forward TPAs like Sana Benefits (our company) handle claims processing and member services. There are transparent PBMs like SmithRx, bundled payments vendors like Remedy Partners, underwriting tools like Verikai, telehealth-first primary care offerings like Eden Health, fertility benefits from Carrot, mental health from Lyra, musculoskeletal solutions from Motion. The list goes on, with several startups in each microservice category.

This is just a tiny sample of a truly explosive number of startups in this space. They provide very different services, but they share a common technological and cultural DNA that makes it easy to integrate with one another. None of them have COBOL in their codebase (that we’ve seen anyway). As a collective, they represent a new wave of health insurance innovation.

Whether or not the companies listed here are the ones that ultimately prevail, it’s clear from the sheer number of them that the revolution has arrived. And it doesn’t stop at self-insurance. Once these platforms get built and tested with self-insured plans, they will migrate to the fully insured world. As a technology manager at a health insurance company, the question to ask yourself is not whether the revolution is here, but whether you want to be a part of it.

Read Also

Insurtech's Real Promise: Growing the Pie

Insurtech's Real Promise: Growing the Pie

Alexandre Lazarow, Investment Director, Cathay Innovation
Compliance-Not Only For Banks

Compliance-Not Only For Banks

Bassam Alousi, Director-Technology, White Clarke Group
Technology Broker of Choice

Technology Broker of Choice

Michael F. Jackson, SVP & CTO, UMB Financial Corp
Outsourcing Demands Strategic Flexibility

Outsourcing Demands Strategic Flexibility

Marc J Firenze, CTO, Eagle Investment Systems