Based in Grand Junction, Colorado, Quality Health Network’s (QHN) mission is to connect people for better health. Using their award-winning health information exchange (HIE), they help hospitals and healthcare providers in western Colorado and southeast Utah securely share patient health information to improve coordination throughout the continuum of patient care. By streamlining this communication, reducing duplication of services and focusing on quality care, QHN can help those providers effect better patient outcomes. In addition to their HIE, QHN also operates their Community Resource Network (CRN)—a social information exchange (SIE) designed to fill gaps and improve the health of communities overall.

Making the Switch to Tableau Software

Seeing as how QHN is in the business of electronic health data, analytics has always been a key component of their business. With over 4,900 users within their HIE serving upwards of one million patients, the amount of data QHN interacts with on a daily basis is massive. Add to that the unique standards and guidelines intrinsic to the healthcare industry, and you have a complex analytics environment that needs to operate within very specific parameters.

Of course, QHN has always been keen to keep their analytics stack as modern as possible. Even with their existing cloud-based BI platform, however, they found that there were limitations to what they could do. For example, they had long relied on sending PDF reports to their users but knew they eventually wanted to send reports directly to those people instead. To do this on their existing platform would be an expensive addition.

So, QHN set out to look for an analytics platform that could better meet their needs. This effort was led by their Director of Analytics, Jason McRoy, who had considerable experience with different analytics platforms. Under his guidance, the conversation around Tableau Software began, and Tableau was chosen as the new analytics anchor for their stack. From there, QHN would figure out what Tableau could integrate with to provide more robust analytics. It was in this phase that Tableau recommended InterWorks as an adept services partner to help integrate Tableau into QHN’s broader analytics stack.

Compounding Success with Matillion and Snowflake

Initially, QHN had intended to use an all-in-one data warehouse and BI platform to set up a traditional environment where they would host their own SQL Server and then plug Tableau into that. After conversations with InterWorks Data Lead Michael Treadwell and Account Executive Karlee Cline, a different path forward emerged.

Michael brought two new players into the conversations that would make life for QHN much easier: Snowflake and Matillion. Snowflake would give QHN a cloud data platform that was lighter to implement and more scalable than a traditional data warehouse. Plus, with Snowflake, there would be no hardware QHN would have to support. When the conversation shifted to ETL, Matillion was a natural fit. Also based in the cloud, Matillion’s real strength—beyond its user-friendly GUI—is its ability to easily ingest and build workflows around variant data. This was especially attractive to QHN because most of their data flows include variant data like HL7, which has historically been a challenge to work with using traditional tools and processes.

Eager to learn more and see Snowflake and Matillion in action, QHN requested a proof of concept (POC) from InterWorks, and Michael Treadwell was happy to oblige. In delivering this POC, there were specific things that really sold both Snowflake and Matillion as the right fit for QHN’s needs.

The big takeaway of Snowflake was that it blended multiple products and a lot of strong functionality into one platform. You have Oracle-based functions, Postgres and tSQL functions in the same place, as well as the ability to work with JSON and XML datasets. Because QHN extracts a lot of data from semi-structured content, Snowflake’s robust functionality was a huge plus. At a higher level, the POC showed just how easy it was to scale Snowflake while providing full cost transparency.

Where Matillion really shone was with its wide variety of data connectors, working with MongoDB, Postgres DB and flat files. Because QHN works with all these variations to some degree, those data connectors were a huge selling point. Of course, Matillion’s ability to streamline and automate workflows also made a solid case for the tool’s utility.

“The staff we worked with at InterWorks – Tim Rhymer, Michael Treadwell, Justin Lemmon – are all really talented, know the tools and are down to earth. They understood our circumstances as a company and went above and beyond to get the work done and get it done right, which gave us a lot of confidence that these efforts would end up where we wanted them to.”

– Jason McRoy, Director of Analytics, QHN

Making Progress with a Modern Analytics Stack

More often than not, progress with data architecture projects takes considerable time, but the Snowflake and Matillion POC took only three months to establish a strong foundation and produce results. At the end of the project, QHN had imported most of their data sources, built an initial data warehousing layer and changed their data-capture and history-capture methods for every connected source.

Building this foundation has enabled QHN to pivot their focus to uncovering new streams of data from providers, subjects and diagnoses with the goal of quickly generating reports within Tableau. The ability to rapidly prototype and test out new reports to see if they are accurate and resonate with stakeholders is an incredible leap for QHN; as a small team, they don’t have the bandwidth to sink time into massive projects. On top of all that, their new technology stack provides these advantages while being budget-conscious, flexible and scalable.

For QHN’s customers, the streamlined data architecture and agile reporting translates to more meaningful insights that can help them run more efficiently and influence better patient experiences. By exposing more streams of data and deeper insights, QHN is able to provide customers with a more useful product on several different levels.

“When we first estimated what we’d need out of a legacy environment, it was clear that it couldn’t handle the full scope of data we wanted to feed into it. With Snowflake and Matillion, however, we were able to load it all and now have more data than ever before. This helps the picture we paint for our customers to be that much richer.”

– Jason McRoy, Director of Analytics, QHN