By Erik Swain
In healthcare, human factors principles are generally associated with the design of medical devices. But Abbott (Abbott Park, IL) used them to redesign drug labels, attendees at a recent conference learned.
Abbott, which makes medical devices and drug-delivery systems as well as pharmaceuticals, has an internal human factors staff that performs functions such as usability testing, task analysis, and use-error risk analysis. When the company decided to redesign its drug labels to reduce dispensing errors, it put its human factors team on the case.
“Our techniques can be applied to a lot of things, and we got a champion within the corporation who saw a problem with drug labels,” said Edmond Israelski, PhD, program manager for human factors in Abbott’s corporate regulatory and quality science organization. He spoke at an AAMI conference on human factors held in June in Washington, DC.
First the team gathered relevant data from literature, adverse-event reports, and customer complaints. Then it observed and conducted interviews with pharmacists, clinicians, and patients. After getting a sense of how users found information on the old labels and analyzing what needed to be improved, the team designed and tested new labels. “In the testing, we used fictitious drug names to reduce conformation bias,” Israelski said.
“We measured how quickly people could find the information on the label for both the old and new designs,” he said. “We recorded eye movements to pick the most efficient label layouts.” Measures included response time, response accuracy, number and duration of visual fixations, and “scan path,” or sequence of fixations. The resulting designs were geared to require less time to find information, produce a shorter and more direct scan path with lessfrequent backtracking, to require fewer and shorter fixations, and, of course, to reduce errors.
The team found that users expected consistency and recommended that label templates for all Abbott drugs be standardized, and that they share similar font sizes and layouts.
These specific recommendations resulted from the study:
• Information critical to identifying drugs should be in the same order from top to bottom.
• Only information critical to drug identification and discrimination should be on the front panel.
• Each word should have a text size reflecting its importance.
• Text should be in dark letters on a light background, except when used for dosage strength color-coding.
• If competitors had look-alike or sound-alike names, the product needed to be discriminable.
• Left alignment of text is best for enhancing scanning speed and clarity and reducing clutter.
• If custom fonts for brand names must be used, they should not interfere with readability.
• Upper- and lower-case lettering should be used instead of all caps.
• Text on side panels should be horizontally oriented, so the bottle does not have to be rotated to be read.