“I’ll tell you a story,” says UX designer Judy Keeley. “Back in April, there was an emergency use ventilator that was rejected by the National Health Service in the UK. The only thing written about it was that it was too hard to change the settings when suctioning a patient.
“Clearly, the ventilator’s developers had missed the boat on something. We wanted to know what it was, and for a while we were just scratching our heads trying to figure it out. What’s needed when you suction a patient? What would make it easier or harder? And what is it about the settings that would make it so difficult they won’t even accept a ventilator that was technically capable of treating a patient during COVID-19?”
UX stands for “User Experience.” It’s a field of research and design geared towards making products more intuitive and easy to use, and it’s absolutely essential for product development. Any kind of quality product that people use and interact with—from a website to a monkey wrench—has likely seen its way past a UX researcher and designer.
Judy Keeley and Hilary Coolidge are the UX team leads at The Ventilator Project.
As the UX research lead, Hilary seeks out potential solutions for the ventilator’s development by looking into the needs and behaviors of product users and stakeholders. Judy, the Design lead, applies that understanding to create a Human-Machine Interface (HMI), or the interactive control screen that allows respiratory therapists, nurses, and physicians to interact with the ventilator and change settings.
Judy Keeley, Lead UX Designer
Hilary Coolidge, Lead UX Researcher
Hilary has extensive experience working with product developers and designers, while Judy has held design leadership roles at companies ranging from healthtech startups to software enterprises in the Fortune 500. It was Judy who joined the initiative first. She heard about a need for UX designers at TVP through her volunteer work for another COVID-related effort. She recruited Hilary soon after, when her friend and former coworker reached out to ask if she knew of any COVID-19 relief efforts that could make use of her UX skills.
“I think both of us were really just looking to help out in this situation, as we all are,” Hilary says. “It’s great to be contributing your knowledge and expertise towards something that’s hopefully saving people’s lives.”
During their first round of research, the UX team set out to gain foundational knowledge about ventilators; how they work, how they are used. Reaching out through personal and professional connections, they conducted interviews with their target users. Judy did a lot of research reading ventilators user manuals, and over time they’ve built up a massive collection of screenshots of different ventilators user interfaces and detailed notes on how they work.
“One of the things we do as a UX team is to understand who the kinds of users for a ventilator are. Respiratory therapists and nurses have different needs when they’re using these things, so we specifically target both kinds of users for research sessions,” says Judy.
The UX team conducts user research sessions with an interactive prototype of the HMI. This way, they receive direct feedback on the controls from those who would be using them in a hospital setting.
“I think it’s pretty user friendly. I like the layout of it,” said one respiratory therapist who interacted with AIRA’s interface.
“Both of us were really just looking to help out in this situation… It’s great to be contributing your knowledge and expertise towards something that’s hopefully saving people’s lives.”
“Any respiratory therapist within 10 minutes of touching this machine will be an expert on it. GE came out with a vent that was just… so hard to use. We had two of them and donated them to the aquarium.”
Judy had to do a double take.
“Yeah, they were ventilating a lot of the turtles. We gave them our GE vents, because they were so not user friendly. This here… I think it’s fairly easy to use.”
As it turns out, UX design can make all the difference for whether a device is used on a human being or a sea turtle. It’s no wonder the team is so thoroughly integrated into the company-wide structure of TVP, working in conjunction with Engineering, Quality and Regulatory affairs, Product Management, and Sales.
As Judy and Hilary continued to conduct user research sessions, they arrived at a key discovery.
Every patient on a ventilator goes through the process of suctioning. Ordinarily, the upper airway cleans and moistens the air we breathe on the way to our lungs, but the endotracheal tube of a ventilator bypasses this mechanism. The air moving through the tube, while it keeps the patient supplied with oxygen, is cooler, dryer, and less clean than it should be. The body’s response to produce more mucus, and it’s necessary for nurses and doctors to suction the patient’s airway from time to time to prevent buildup.
People with severe cases of COVID-19, however, may develop Acute Respiratory Distress Syndrome (ARDS) which means the lungs are so inflamed that tiny blood vessels rupture and flood the lungs with a liquid. These patients must be suctioned very often. ARDS usually takes time to develop, but COVID-19 patients have been known to progress to this dire stage practically overnight. A ventilator designed to respond to the pandemic must be able to accommodate the need to suction a patient quickly and frequently—in the worst cases, almost constantly.
To suction a patient, the practitioner must first activate a mode called “100% O2.” This mode floods the patient with the amount of oxygen they’ll need to carry them through the suctioning procedure.
“After hearing about this from everyone we talked to,” Judy says, “we suddenly connected the dots for this story that we’d heard a long time ago, when we were wondering what settings would be so difficult to get this UK ventilator rejected. Now, we’ve made an automatically-timed 100% O2 button that’s easy to reach on the home screen, as opposed to an interface that requires someone to go into a settings menu to turn the mode on and off. It’s getting great feedback.”
This is an artist rendering of the device. Final product may differ.
“Suctioning isn’t technically a ventilator mechanical function, but through our interviews, we discovered there was an optimal way to support the process via the user interface,” says Hilary. “Then, the engineering team was able to incorporate it into the next version of the prototype.”
With the first few rounds of UX research completed, AIRA’s EUA submission has been sent to the FDA to be processed. The next round of research will be aimed at a market-ready product.
“A big part of the product’s success is usability,” says Judy. “If something isn’t intuitive, or if there’s too many steps, or you can’t find where something is, or if you’re not seeing the alarms where you need to be… that can lead to critical patient safety concerns.”
One thing is certain: by focusing on the practical needs of ventilator administrators and COVID-19 patients (not turtles), the UX team isn’t about to let that happen.
“Any respiratory therapist within 10 minutes of touching this machine will be an expert on it.”