Welcome to Part Two of a three-part blog series about the state of EHR, third-party, and enterprise application testing in healthcare. Part One provided an overview of the enduring strain that today’s manual testing demands are putting on hospitals and health systems who want nothing more than to increase efficiency and optimize the time and skills of their IT team with higher-level tasks. The constant drumbeat of script changes makes doing both a monumental undertaking.
In Part Two, we’ll examine UNC Health’s quest to improve its testing protocols by transitioning from a manual to an in-house, automated system and why certain realizations led them to seek third-party assistance from SureTest, a leading test automation solution provider.
UNC Health’s Journey
THE UNC HEALTH CHALLENGE | |
The Situation | UNC Health experienced unprecedented growth after it expanded facilities and acquired regional hospitals, requiring the UNC team to integrate Epic and third-party testing for new facilities, annual upgrades, and system updates. |
Initial Efforts | The goal was to improve test quality and free up time for internal analysts and clinicians to focus on strategic projects related to expansion and Epic optimization. Initially, UNC Health invested in test automation software and trained internal staff to build and maintain automated scripts to assist in the testing effort. |
What Happened | The reality was that the complexity and sheer volume of scripts were overwhelming and required significant additional capacity to complete, support, and maintain an automation library. After 18 months of effort and less than 20% of scripts automated, UNC Health realized that the project scale demanded more resources. |
Building an Intelligent Test Automation Machine
At SureTest, what we hear most often from health system IT executives is:
“I know the constant wave of testing demands will grow exponentially, but I also know that I can’t continue to expand my department to match it. So how do I build an infrastructure that:
- Automatically tests throughout all enterprise-wide workflows;
- Taps existing script testing knowledge that makes my testing faster and smarter so I don’t have to start from scratch every time; and
- Recognizes the unique complexities of my health system?
Those are the right questions to ask and the perfect springboard to demonstrate how to build a best-practice automated EHR, third-party, and enterprise application testing machine that is built around people, processes, and technology. It’s a “work smarter, not harder” strategy that significantly reduces demand on health system resources while paying incredible dividends, by using the following five steps:
BUILD A MODULAR FRAMEWORK | It might seem oxymoronic that best practice testing automation requires a rock-solid foundation plus nimble flexibility across thousands of moving parts, but that’s exactly what needs to be built from the start and maintained over time. The reality is that the stakes are so high in testing that any system needs to be built on bedrock. At the same time, however, it is tough to find a more dynamically changing environment than healthcare at the macro and micro levels, which requires near-real time flexibility to adapt. Care settings within a health system and each specific workflow can vary greatly, so the testing infrastructure needs to be able to adapt to current and changing workflows and even seamlessly absorb the workflows of new entities joining the system. The automation system must provide continuous testing, robust script development and maintenance, real-time logs and audit trails, reporting and documentation. |
INTEGRATE A TEST SCRIPT LIBRARY | Question: You can have the most amazing testing framework, but how do you make it scale so that it outpaces the constantly growing demand for testing throughout every aspect of the health system operation? Answer: Have a great library that taps into the testing experiences of other health systems and that can update in real time so that you have the freshest data possible in your library. You need to either develop or leverage a foundational collection of extensive automated test scripts that you can automatically leverage, deploy and schedule at scale for system-wide testing across each and every workflow, which also powers end-to-end smoke and regression testing of key clinical and revenue cycle workflows. |
CUSTOMIZED ENGINE FOR EVERY WORKFLOW | That general library of automation is the speed pass for developing a customized testing system with the appropriate test modules that emulates exact functions for each use case across workflows. As a result, you can target workflows that need testing and automate them with much less overhead, executing them either scheduled or on demand, and workflow capture recordings can be leveraged for eLearning, making the system and the team smarter and supporting the decision-making process on an ongoing basis. |
FUEL IT WITH A POWERFUL TECHSTACK | To power testing automation, you are going to need a proven enterprise software tool, which should have the following core criteria: Meets health system requirements for enterprise solution testing Handles the complexities of patient journeys and patient care intricacies, such as complex EHR systems like Epic or Cerner Supports automation testing for any system with a test environment and a mouse/keyboard/touch-screen interface Has AI capabilities that can model and apply patterns, workflows and tasks, and analyze historical testing data, user behaviors and system interactions to continually improve overall testing performance |
MANAGE WITH EXPERTISE | While most of your IT staff who are currently focused on testing will be able to pivot to higher priority projects, you’ll need some oversight of the testing engine, whether that is in-house or with an external partner. Frankly, if someone tries to tell you that machines and complex processes run themselves, walk away (quickly). That internal or external core team should understand the healthcare ecosystem in general and the health system’s operational infrastructure in particular and should include test engineers with multi-year Epic and Cerner build and implementation expertise. The team should be able to perform all the continual development and daily maintenance (testing scripts, validating changes, and updating as appropriate) required to keep the health system’s test automation engine evergreen. |
After UNC Health’s experience building an in-house automation testing program, they elected to partner with SureTest, which checked all the above boxes and helped springboard UNC’s testing operations to the level needed to support its strategic objectives.
“Leveraging the SureTest partnership, we have been able to get back thousands of hours annually, freeing up time for our analysts to focus on more strategic build and optimization projects,” said Lynn Nichols, PMO, Healthcare System Manager for UNC Health. “The accuracy and volume of testing have significantly increased the quality of the changes going into production during upgrades and go-lives, resulting in greater productivity across our teams that support our EHR and third-party application testing.”
Up Next: UNC Health’s Testing Transformation
This second installment of our three-part series examined UNC Health’s ambitious attempt to build an automated, in-house testing system, the challenges it brought about, and UNC Health’s decision to change course and seek help from an experienced solution provider in SureTest.
In the final installment of our series, we’ll look more closely at UNC Health’s partnership with SureTest and how SureTest’s customized, multi-phase solution delivered the testing efficiency, accuracy, and employee time savings that UNC Health was looking for.