Manufacturers should pay closer attention to user interfaces as they design and test devices, an FDA official told attendees at a recent workshop on infusion pump improvements.

"Better design can prevent use error," said Ron Kaye, Human Factors and Device-Use Safety Team leader in the CDRH's Office of Device Evaluation, at the FDA workshop May 26 in Silver Spring, Md. "To make a better infusion pump, we need to optimize the design of the user interface."

The workshop elaborated on recent FDA draft guidance that recommends new pre-market testing activities for companies seeking 510(k) clearance for a new infusion pump or changes to a product or labeling (¹ Also see "FDA, Industry Discuss Regulatory Impacts Of Infusion Pump Safety Steps" - Medtech Insight, 31 May, 2010.).

Under the guidance, FDA would look for two forms of validation testing - simulated use and clinical evaluation - even when a substantially equivalent infusion pump has been shown to be clinically effective.

"FDA views use errors as serious risks for medical devices, and particularly for infusion pumps right now," said Kaye.

"Use error is equally dangerous as device failure problems. If a device fails and somehow users have killed somebody, or if somebody makes an error when using a device, they can be just as injured or just as dead as a result of that."

Kaye wants device makers to consider how user interface design can reduce residual risk.

"Anytime anybody interacts with a device, there's going to be a chance of some kind of error, just like anytime one of us gets behind the wheel of a car, there's some residual chance of failure," Kaye acknowledged.

"But there are pumps on the market where you don't see a certain type of error, and if another pump has that [kind of error], you have to wonder why," Kaye said.

He noted that FDA has seen recalls attributed to use error when in fact the design interface was at fault.

"We know that a better design will help because when those designs were fixed, the errors went away," he said.

FDA wants firms submitting an infusion pump 510(k) to use pre-market validation testing to help support an assurance case on the product's safety and effectiveness (² Also see "FDA To Impose New Pre-market Requirements On Infusion Pump Makers" - Medtech Insight, 3 May, 2010.).

According to Kaye, validation testing can help uncover the kinds of use errors that tend to be found only after the typical user population interacts with the device under normal conditions.

Use-related hazards are "an entity [un]to themselves" that must be evaluated separately from device failure hazards, he explained.

"Human factors evaluation is not about whether or not the device is going to work, whether it's - in the case of the infusion pump - going to deliver drugs and deliver that accurately."

"The question is that when you get the user in the loop ... is that combination of user and device going to be effective at doing that consistently and safely, or are there going to be problems?"

During human factors testing, "we're out there looking for trouble," Kaye said.

Agency Focusing On User Interface Attributes

Variability among user interfaces on different infusion pumps can contribute to use error, he suggested.

"You don't want to have a design that does something backwards compared to all the other devices the user has ever used."

He encouraged device makers to design user interfaces that are intuitive and informative.

"It shouldn't be so complex that it will unnecessarily delay or prevent operators from using the pump at all," he said. "It should be designed to optimize the arrangement of displays, controls, labeling and accessories to facilitate user interaction and error prevention."

Controls on the user interface should respond appropriately to user commands and should also be resistant to wear from fingernails, the FDA official added.

For pumps with computer-based interfaces, an important attribute is the logic of the software operation, Kaye said.

Software designers should test what graphics are displayed to the user, when, and in what sequence, to see how they fit with what users are doing at the time and with "what they're thinking and what they're understanding while they're using the infusion pump."

Alarms on the pumps should successfully alert users in noisy and distracting hospital environments, but should not sound repeatedly for trivial reasons, Kaye described.

"And that goes for warning screens as well," Kaye added. "We heard repeatedly about instances where warning screens appeared so much that the typical way the device is used becomes 'See the warning screen, exit out, proceed, don't read it.'"

Tests Should Seek Subjective User Input

Subjective user evaluations are an important part of the validation testing, Kaye said, to help uncover close calls that did not lead to outright device failure but might have if not for the last-moment vigilance of the user.

"You need to ask your users what they think about what they just did with the device," Kaye advised. "Were there problems? Was anything particularly confusing? Did you think that you might make an error at some point in interacting with the device?"

"It's very important to get the user perspective on what's going on and not just some tally of performance measures," he added.

When conducting validation testing, device makers need to be proactive in soliciting assessments from study participants, Kaye said.

"You need to actively go out and engage the user," Kaye said. "You don't leave a form sitting there and say they can comment if they want."

Study participants should not include employees of device manufacturers, Kaye instructed.

The extent to which users are representative is very important, he said.

Kaye explained the main purpose of validation testing is to find cases of unanticipated use that would draw such comments as "I never thought they would use it like that, or in this way, under these conditions."

To help tests uncover potential use errors, training on how to operate the infusion pump should closely mimic what users are likely to experience in real-world settings, Kaye suggested. In some studies, training is fresh in the participants' minds and does not represent what can happen to users over time, he added.

Because clinical evaluation studies tend to be longer-term than simulated-use studies, participants are more likely to experience the realistic effects of training decay, he added.

Simulated-use studies should be done prior to clinical evaluation, Kaye said, in order to demonstrate that the pump is safe enough to be used on real people.

"You can load up your simulated-use testing session with a lot of very important tasks that users might not typically do in that amount of time but because they're important, we want to test them," Kaye explained. "You can script the scenario, if you will, so that you are evaluating that intensely."

He complained that some prior human factors tests submitted to FDA have incorporated trivial measurements, such as "How long does it take the user to unpack the device from the box, or how long does it take them to plug it in?"

"Typically those aren't the types of things that we want," he said. "We want to see [whether users are] going to misprogram it or are they going to set it up so it fails to deliver what they think it is, that sort of thing."

When a validation test is conducted on a fix for a particular problem, Kaye emphasized, "you've got to focus on that. You have to prioritize that in your testing."

- Monica Hogan ( ³ m.hogan@elsevier.com )

Better Interfaces, User Testing Could Improve Infusion Pumps, Says FDA

More from Archive

More from Medtech Insight