Whenever you enter the healthcare system, technology is all around.
And you have to trust it.
The highly trained medical staff have all been through rigorous testing to ensure they are able to deliver the best possible care. And you probably expect the same from the huge range of technology they use to treat you, monitor your progress, and update your records.
But as that technology permeates every aspect of healthcare and the pace of change increases, it’s getting harder to ensure that everything works as intended.
Dividends and dangers of digital transformation
Digital transformation involves embedding digital technology in an organization’s systems and processes in such a way that it improves end user experiences.
For those involved in developing software for healthcare providers, processes become more agile and more responsive to feedback. As a consequence, release cycles are accelerated, and the pressure on devops and software testing grows.
Those tasked with deploying that software arguably face an even greater challenge as they try to ensure that multiple applications both work as intended and are able to work together. Digital transformation introduces a lot of new software into hospitals at a fast pace, and that software needs to work seamlessly with medical equipment, both to improve patient care and reduce costs. And if interfaces aren’t properly tested, patients could be exposed to significant risk.
The opportunities in the healthcare sector are huge in terms of better patient outcomes and efficiency, with the adoption of digital health records reducing bureaucracy and freeing up more resources for patient care. But there are plenty of hurdles along the way, not least of which is interoperability—ensuring that patient information can be shared between systems and staff.
With a growing need to release and deploy software faster, there is a risk that testing becomes the bottleneck. This is particularly true when you have to rely on teams of manual testers, especially when those testers are borrowed from their real jobs as clinicians, so they can sit in front of a workstation and ensure that everything works. This can result in reducing the number of clinicians who are available to help patients. Bandwidth is limited, and if you depend on manual testing, you either have to hire (and train) new staff or try to squeeze more out of the people who are doing it at the moment.
That brings us to the next point. Pushing manual testers to work harder or faster is fraught with risk. People are fallible, and when they’re pushed to the limit, mistakes are more likely, especially if those testers are also already overworked clinicians. At the same time, more software and greater requirements for interoperability can mean that the number of possible workflows increases exponentially, beyond the capabilities of manual testing altogether.
And even if capacity isn’t the issue, simply having to run similar tests more frequently could lead to boredom and complacency.
In any other setting, this could mean poorer app reviews, lower net promoter scores, or falling conversion rates. In healthcare, if the right information isn’t shared between the right systems at the right times, it could mean that lives are lost.
Every organization is under pressure to minimize or reduce costs, and in the healthcare space, there are clearly additional tensions. Every additional penny spent on paying a person to test a system manually means you spend less on something else that could lead to better patient outcomes. And, as we’ve already noted, the other option—trying to achieve more with less—comes with its own risks.
The upshot of all this is that there is a clear risk: digital transformation, for all its benefits, could have unintended consequences of spiraling costs or poorer patient outcomes.
Intelligent test automation
If manual testing isn’t up to the task, the obvious question is: can it be automated?
The ease with which automated testing can be scaled up means that testing is far less likely to be a bottleneck than an overworked team of manual testers.
At the same time, it may simply be impossible to test increasingly complex systems manually, as the number of routes a user could take through a given piece of software increases exponentially.
Testing only the happy paths without some form of automated AI to hunt down bugs could mean that catastrophic failures are left undiscovered until the perfect storm of unlikely coincidences conspire to cause a tragedy.
AI-driven test automation gets over this by testing far more routes through applications than a whole army of manual testers could ever hope to achieve.
Human versus machine
Some aspects of testing are easier to automate than others, and it’s important to keep in mind that even if software passes functional tests with flying colors, end user experience is still critical. Even here, though, the ground that test automation cannot cover is rapidly shrinking, thanks to intelligent image recognition, screen comparison and automated usability testing.
Ultimately, there is much to be gained from digital transformation in healthcare in terms of better patient outcomes, but for the benefits to outweigh the risks, intelligent test automation is essential to ensure that testing is able to keep up with the pace of change.
See why Cerner chose Eggplant for intelligent test automation