Total Recall: The Consequence of Ignoring Medical Device Usability

A nervous thirteen-year-old girl sat in a pre-operative room. As she spoke with the anesthesiologist about the impending knee surgery, a nurse came by with a clipboard, requiring that the eighth grader confirm she needed surgery on her left knee. She checked the box “Left,” and signed her name. Then the nurse handed her a giant permanent marker and asked the young patient to label her “good” leg, “NOT THIS ONE” and “NO SURGERY HERE.” Her finishing touch was a large, “NO!” on the top of her right kneecap (see Figure 1). The girl nervously laughed about why they might require her to do this.

Drawing of a patient's leg with No! No surgery here, no, no, no scrawled all over it in red marker.

Figure 1. n hospitals, patients are sometimes asked to label the part of their body where the doctor will perform surgery, or, in this case, they are asked to label the part of their body where the doctor should not operate.

Perhaps you have heard stories of a doctor operating on or amputating the wrong limb. Even though this is an age-old problem, some medical devices cause the user to confuse the sides of the body, consequently leading to recalls of the devices. Just last year, the U.S. Food and Drug Administration (FDA) recalled a software system because the interface led doctors to confuse the left and right sides of the brain when evaluating patients (see Figure 2). Imagine the consequences of this design flaw during brain surgery!

Left and right profiles of a human head labeled Left and Right.

Figure 2. This depiction of brain hemispheres is inherently confusing. The person’s right side is labeled “left” and vice-versa.

In 2008, the FDA recalled an ultrasound system because the graphics made users misunderstand the image orientations of the patient’s left and right sides. The users made the assumption that the patient’s right and left sides were oriented in the same direction as the transducer, but this assumption was incorrect. Usability testing reveals the assumptions that users make, which can prevent designers from overlooking medical device characteristics that may lead to a higher risk for users, and ultimately, a recall.

Unfortunately, evaluating the wrong side of a body is just one example among many usability issues challenging medical devices. In the medical field, problems with usability are referred to as “use errors” instead of “user errors” to defer blame away from the user. It is not the user’s fault when the interface sets them up for failure.

Use errors are defined as “something that the user does or fails to do that results in an outcome that the user or manufacturer does not expect.” Use errors can be reasonably predicted to occur. Their risk of happening can be minimized through contextual inquiry, risk analysis, and usability testing.

Users of medical devices fall into five general categories: the general public (patients and home caregivers), healthcare providers (clinicians, including physicians and nurses), laboratory and biomedical technicians, imaging specialists (such as radiologists), and pharmacy staff. Each user profile tends to have its own set of challenges for medical device development. Usability research can aid in predicting the use errors common within each user group and inform design.

Medical Devices for the Home

Use Errors Due to Displays

One glucose meter was recalled because its users incorrectly interpreted the numbers on the display. When it showed “2.2,” the square decimal point was so small and so close to the first “2” that users read the number as “22” (see Figure 3). As a result, the users thought their blood glucose levels were ten times higher than they really were. Patients would mistakenly believe they needed to adjust their insulin injections to a much higher level than necessary. The consequences of this mistake could range from severe hypoglycemia to diabetic coma or death.

A screen with exceedingly large digits two and two, and a very small decimal point between the digits.

Figure 3. This glucose meter screen makes it difficult to see the decimal point and may lead users to believe their blood glucose level is ten times higher than it actually is.

In another case, the FDA issued an industry-wide recall of glucose meters that led to similar consequences for patients. The meters allowed users to choose between units of measure during setup. This feature allowed the meter to be marketed in both the U.S. and Europe. Additionally, it accommodated users who travel and regularly visit doctors in multiple countries. In the United States, insulin units are displayed in mg/dL, but in Europe and other parts of the world, units are displayed in mmol/L. The units differ by a factor of eighteen (18), which could lead to patients mistaking the units and taking incorrect therapy actions.

Users originally toggled through different modes when setting up the time and units of the device. When they set the clock or changed it for daylight savings, they would sometimes inadvertently change the units of measure from mmol/L to mg/dL or vice versa. Many times, users did not realize they had changed the units of measure because the units were not always prominently displayed.

In response to this use error, the FDA required the manufacturers to remove the capability of changing units on all glucose meters. Now travelers are inconvenienced by having to own two meters, and suppliers must manage additional inventory, but the prospect of serious consequences is diminished.

In the case of glucose meters and many other medical devices used by patients at home, design teams should confirm that users understand labeling, warnings, and instructions for use. This can be accomplished through comprehension studies in which users rely on instructions, labels, or the device itself in order to use the device.

Use Errors Due to Ergonomics

Failure to test the usability of a device can lead to poor ergonomic design—another common cause of recalls of medical devices. In 2008, the FDA recalled a wheelchair that had the potential to pinch a user’s fingers between seat bars when the user was opening the wheelchair, resulting in injuries as severe as fracture and severing.

To prevent the hazard of wheelchairs clipping fingers, a design team should create a detailed list of tasks associated with using a wheelchair and steps within each task. They should then consider each way in which a task could be performed incorrectly, and the consequence of skipping a step.
For medical devices, the FDA requires a formal risk analysis as part of the design process (see Figure 4). In a risk analysis, a cross-functional team considers the probability of a use error occurring and harming the user. Then, they rate the severity of the resulting harm to create risk level ratings for each task. Tasks with the highest risk level are prioritized during the redesign process.

Sample chart.

Figure 4. This is an example of a Failure Modes and Effects Analysis (FMEA) for a hypothetical automatic external defibrillator. FMEA is used to evaluate the risk profiles of use errors.

Contextual inquiry is another useful tool for identifying potential use errors, since it reveals what users do during actual use. The design team must create a design that reduces the possibility of these use errors, and then ensure that the use errors have been minimized by performing an iterative set of usability tests.

Medical Devices in Hospitals

Use Errors Due to Misconnections

One hot topic in hospitals is medical device misconnections. There is often a maze of tubes, connectors, and cords surrounding patients in hospitals. Unfortunately, due to haste or momentary confusion, it’s not impossible to mistakenly connect a ventilator air supply tube to an IV line. The Association for the Advancement of Medical Instrumentation (AAMI) and ISO offers a standard for small bore connectors in healthcare (ISO 80369-1:2010). Currently, AAMI and ISO are developing enhanced universal standards for connectors used in medical devices to further minimize misconnections.

Correct practice of usability engineering dictates that mating parts should uniquely mate with their counterparts, with minimal possibility of an incorrect connection (see Figure 5). In 2010, a tool for cataract surgery was recalled because it was possible to mate two components incorrectly. This incorrect mating led to the generation of plastic dust during cataract surgery.

Illustration of two mating parts that are impossible to connect. Three prongs on one plug and two receptacles on the mating part.

Figure 5. Mating parts should uniquely mate with their counterparts with minimal possibility of an incorrect connection. The design in this picture prevents the user from making a misconnection.

Medical device manufacturers can prevent misconnections by performing usability studies in which target users (medical professionals) perform tasks in a simulated use environment, typically without instructions. The results of the usability study would reflect what a doctor might do in the worst-case scenario—a situation in which the instructions are forgotten, ignored, or misplaced.

Manufacturers should attempt to foresee use errors such as misconnections by performing a thorough task analysis and determining all possible misconnections. They must take into consideration all other devices in the user’s environment as a routine part of risk prediction.

Use Errors due to Alarms

Another major concern in hospitals is alarms. Medical device alarms and signals must be understandable and audible in their use environments. Usability testing with healthcare providers can reveal whether an alarm is effective in its various modes (for example, lowest volume, visual only, and sound only). Designers must test users’ responses to alarms in a simulated use setting that replicates actual noise levels in the hospital room, along with the effect of having multiple personnel, and the distractions of many other medical devices alarming simultaneously.

A monitoring device was recently recalled because of an ineffective alarm system. Device makers prefer to err on the side of overly-sensitive alarms, which lead to false alarms. To prevent alarm fatigue, healthcare providers sometimes turn audible alarms down or off, which is what happened in this case. However, the visual indication alone was insufficient in notifying personnel of a critical situation (see Figure 6). A usability study would have shown whether or not the visual alarm was attention grabbing and action inspiring. If the visual alarm alone was ineffective, designers might have chosen to prevent nurses from muting it.

A poorly designed visual alarm where the warning is unclear owing to the excess numerical information surrounding the warning

Figure 6. Visual alarms and warnings need to be noticeable, understandable, and informative. This user interface is crowded with information and has poor color contrast. The “Warning!” message is difficult to see and uninformative. A usability study might reveal that this warning is not effective in inspiring prompt and meaningful action.

Logically, users are less likely to respond to a quiet, unobtrusive alarm. Designers need to decrease the possibility that a nurse will turn off an alarm by reducing the number of false-positive alarms emitted by a device. Likewise, the designers of adjacent devices should reduce the volume or intensity of less safety-critical alarms in relation to more safety-critical alarms. Standards are available for guiding design teams in the development of effective alarms, including International Electrotechnical Commission (IEC) 60601-1-8.

Hospitals offer diverse use environments with wide-ranging levels of noise and activity. Medical device designers need to ensure that alarms can be heard under circumstances of normal operation. In one scenario, an alarm on a fluoroscopic imaging system could be heard by the technicians in the control room but not by the doctors in the procedure room. Since doctors sometimes base split-second decisions on the output of the imaging system, the consequence of an unnoticed frozen screen or an otherwise misleading image (which they would have been informed of by the alarm), could be fatal. As a result, the imaging system was recalled.

An understanding of how users interact with their environments is critical not only for designing effective and safe alarms, but also for medical devices in general. Designers must use contextual inquiry to unveil important characteristics of use environments and subsequently integrate these characteristics in their usability studies.

In sum, usability engineering offers numerous tools to avoid recalls due to use errors. The following three steps contain the key to reducing medical device recalls:

  1. Perform contextual inquiry to gain insight into potential use errors and to identify characteristics of the use environments that must be simulated in usability studies.
  2. Evaluate a detailed list of tasks to determine which tasks are the most risky. Then, design the device to reduce or eliminate risk.
  3. Test user interfaces in usability studies with representative users. Include tasks necessary for normal operation, along with the most safety-critical tasks.

Usability is a cornerstone of safety; use errors are frequently predictable and avoidable through proper contextual inquiry, task analysis, risk prediction, risk reduction, and usability testing.

Clark, S., Israelski, E. (2012). Total Recall: The Consequence of Ignoring Medical Device Usability. User Experience Magazine, 11(1).
Retrieved from http://uxpamagazine.org/total-recall/

Comments are closed.