PRODUCT DEVELOPMENT INSIGHT
Next year, AAMI is likely to release AAMI HE75:200X—an encyclopedic document containing more than a thousand human factors guidelines for the design of safe, effective, and user-friendly medical devices.1 The new document will tell you how to make a patient monitor's digital readouts legible, what features enhance a portable CT scanner's mobility, how to design alarms that draw attention in noisy environments, and myriad other ways to improve medical device safety, effectiveness, and usability.
The document promises to give user interface quality a boost just when the medical device industry needs as much guidance as possible to respond effectively to the requirements set forth by IEC 62366:2007, “Medical Devices—Application of Usability Engineering to Medical Devices.”2 In a nutshell, the IEC standard calls for manufacturers to follow good human factors processes when designing a medical device, but does not provide detailed design guidance. That's the forthcoming standard's job.
AAMI's Human Factors Engineering Committee, which includes a few dozen engineers, human factors specialists, and clinicians with a special interest in medical device safety and usability, spent the better part of the past decade creating the lengthy collection of design guidance—560 pages and counting. The document is currently undergoing editing in response to industry and committee member comments on a draft released for public comment in early 2007. It still could undergo another round of public comment and editing depending on the extent of the current round of edits, grammatical nitpicks notwithstanding. So, the exact release date remains uncertain.
AAMI's Teresa Zuraski, senior vice president of standards policy and programs, says, “The draft standard will eventually be submitted to American National Standards Institute (ANSI) for approval as an American National Standard.” Just as ANSI/AAMI HE74: 2001's influence has extended beyond the United States, the new standard should also affect medical device design worldwide.3 Although human factors guidelines are available in many other forms, none focus so specifically on medical devices (see the sidebar, “Popular Sources of Human Factors Guidance” for a list of those publications). Rather, they tend to focus on military equipment, computer software programs intended for personal and business use, industrial equipment, and consumer electronics.
The new standard succeeds a portion of ANSI/AAMI HE48:1993, which combined design process guidance with applied design guidance.4 In 1998, when ANSI/AAMI HE48:1993 became subject to recall after its fifth year in circulation, the Human Factors Engineering (HFE) Committee determined that there was a need for more in-depth process and design guidance than that found in the aging document. Based on research involving potential users—hardware and software developers working within medical device manufacturing companies—the committee decided to produce an updated version in two parts: one outlining good human factors methods and a second presenting detailed design guidance. This strategy enabled the committee to quickly publish the process-oriented guide and put more time into the applied guidelines.
An anesthesia workstation arranges flow tubes, flow controls, and associated gas cylinder pressure gauges in vertical columns to reinforce functional relationships.
Accordingly, AAMI published the first part (ANSI/AAMI HE74:2001, “Human Factors Design Process for Medical Devices”) in 2001. This document aims to help manufacturers respond properly to FDA's quality system regulation pertaining to human factors. In short, the process guide calls for manufacturers to conduct an appropriately scaled set of human factors research, design, and user testing activities en route to producing a final, validated medical device that resists use errors. But the document does not include applied design guidelines. For example, it will not help a mechanical engineer determine the proper size and spacing of control panel components to ensure ease of use and prevent accidental actuations. The forthcoming companion document does.
Notably, ANSI/AAMI HE-2001 formed the basis for the International Electrotechnical Commission's collateral standard, IEC 60601-1-6, “Medical Electrical Equipment—Part 1-6: General Requirements for Safety—Collateral Standard: Usability,” which was released in 2004.5 The current plan is to revise IEC 60601-1-6 to fully align it with IEC 62366 by replacing most of the current content of IEC 60601-1-6 with a normative reference to IEC 62366.
Eventually, the IEC committee plans to withdraw IEC 60601-1-6. That could happen by incorporating the reference to IEC 62366 directly into IEC 60601-1 or by amending IEC 62366 to add a normative annex that contains the necessary bridging requirements.
Return to Article:
The new standard addresses 22 topics (see the sidebar, “Breaking Down AAMI HE75:200X,” for a list of the topics included in the document). Readers will discover some inconsistency in how the standard addresses each topic. By intent, some topics cross the line between providing detailed design guidance and addressing design methodology issues, such as conducting usability tests. Some sections contain reference data, such as the dimensions of small and large hands that can inform the design of a surgical instrument. Other sections focus on user interface details, such as the proper size of characters appearing on a critical display (e.g., a patient monitor) that users will view from across a room. Readers should keep in mind that the standard includes design guidelines, not rigid requirements. Ed Israelski, the AAMI HFE committee cochair and a human factors program manager in Abbott Laboratories's quality and regulatory group, says, “Guidelines are suggestions for good design that are not absolute and, therefore, are not rigid requirements. They give one of many ways to design, but in the case of design principles from HE75, the guidelines reflect expert opinions that are commonly backed up by references to the research literature. [The HFE committee] tried to indicate when product-specific limitations might lead to other design alternatives.” He clarifies, “HE75 gives guidelines that, if followed, will make product approval easier. However, the user interface design still requires validation through the methods described in HE74—the HFE process standard.”
Vibrations in a moving ambulance make it harder to read displays. AAMI HE75 advises designers to make display characters (e.g., heart rate and blood pressure) larger than normal.
In contrast to best-selling novels, AAMI HE75 is hardly a page-turner, and there is little point in reading it cover to cover. Rather, the authors intended for users to jump directly to sections of interest. Therefore, for coherence sake, the 25 sections overlap. Hopefully, the industry vetting and AAMI editing process will eliminate any contradictory guidance by the time it is published. Although the new standard will be available in printed form, the electronic (PDF) version is likely to be most useful, enabling users to search for key words, such as handle, vibration, or legibility. AAMI typically sells standards of AAMI HE75's size for several hundred dollars, but it has yet to announce a price.
The new standard's prime audience is engineers and software developers who have a hand in shaping a medical device's user interface. It gives them a primary information source to support the design and specification of components such as handles, controls, displays, and computer screens. Industrial designers responsible for shaping and styling a product should also find the guidelines useful. Israelski says, “Many developers can read the design principles and apply them without being human factors experts. FDA has suggested that developers should be trained in HF to carry out the HF process, which is still required.”
Early in the design conceptualization process, product planners including marketers and the aforementioned engineers can review sections of AAMI HE75 to determine which guidelines would make suitable user requirements, or what FDA calls design inputs.
Of course, human factors engineers playing a user interface design and evaluation role on development projects can also draw heavily on the document, which essentially standardizes the design characteristics that they have championed for years without the support of such formalized and comprehensive guidelines. Who else might find the new standard useful? Certainly, risk managers will find valuable information that will help them judge a new device's vulnerability to use error. For example, a risk manager might ask the question, “Does our portable device have good handles, or is someone going to injure themselves or others by dropping it?” Turning to AAMI HE75, the risk manager would find the following guidelines:
- Handle grips should be made of a nonslip material.
- Handle grips should be relatively conspicuous to encourage proper handling.
- Handle grips should be visually distinct from controls and other elements.
- Handle shape should allow the user to maintain a neutral or natural and comfortable wrist position to reduce strain during transport.
- Handles will be used by either bare or gloved hands that may be covered with substances common to medical environments such as an antiseptic solution, blood, or talcum powder.
- Carrying handles should distribute weight evenly across the hand, including the palm, to prevent pinching from corners or edges.
- The handle size should be large enough to allow for a hand-to-hand exchange between users.
Such guidelines should help risk managers perform the necessary analysis to determine whether a design is vulnerable to use error and thus requires modification. Please note that sample guidelines such as those presented here are subject to change based on the AAMI HFE committee's review of industry comments on the draft standard.
The list of potential standard users goes on. Consumers, including clinicians and buyers involved in medical device procurements, could draw upon the standard to evaluate candidate devices, such as portable ventilators used in ambulances. In such cases, they would want to pay close attention to the standard's section on Environment of Use, which states, “High-vibration environments include ground and air emergency transport vehicles, for example. For medical devices used in these types of environments, consider large display characters, minimize display clutter, and provide large enough manual controls to counteract tracking and reach errors that can occur above accelerations of 2.0 m/sec2.”
Effect on Industry
Undoubtedly, AAMI HE75:200X will place greater pressure on medical device manufacturers to conform to good human factors practice because it contains far more guidelines than its predecessor. The section authors adapted many guidelines from reference documents such as other AAMI, ANSI, IEC, ISO, and military standards, as well as authoritative textbooks. However, to fill voids, they also created new guidelines based on their professional knowledge and experience. As such, readers will find guidance on topics never before addressed in a medical-related design standard.
AAMI HE75:200X should accelerate several medical device design tasks, including establishing user requirements, developing design solutions, determining testing criteria, and writing final design specifications. Until now, developers had to conduct a wide and often frustrating search for human factors guidelines that they could convert into requirements, criteria, and specifications. Often, they took what they could from the old AAMI standard (AAMI HE48-1993) and from MIL-STD-1472, the venerable human factors standard for military equipment.6 Using the latter, medical device designers often had to translate guidance intended to help weapon design into something applicable to a diagnostic device, for example. Medical device designers lucked out if their new instrument, a surgical stapler for example, incorporated a trigger. The new standard promises to be a one-stop shop for such guidelines, although it cites many other resources.
This infusion pump is equipped with a touch screen that spaces touch targets apart to avoid accidental actuation. (Courtesy of HOSPIRA (Lake Forest, IL))
One challenge facing AAMI HE75:200X readers is determining which guidelines apply to their medical device—not a simple task given the large number of guidelines. For example, an infusion pump manufacturer might need to follow guidelines found in the controls; visual displays; alarm design; connectors and connections; software user interface; and signs, symbols, and markings sections. But then, other medical device standards, such as IEC 60601-1, pose a similar challenge.
Richard Stein, a principal engineer with St. Jude Medical and a member of the AAMI HFE committee, expects that “rank and file engineers will react to the HE75 standard in various ways. On the positive side, engineers who value published and accepted guidance will use the standard as a definitive source of design guidance and test methodologies. Engineers experienced in user-centric design will use the standard as an additional source of guidance, complementing their accumulated knowledge and judgment. Less-experienced engineers or their organizations may see the standard as learning tool. On the negative side, organizations that have already institutionalized good human factors practices might see the new standard as yet another set of design criteria to fulfill and one that makes more work without clear benefit.” So, according to Stein, the new standard is likely to be welcomed and provide value to many medical device developers, but not all, which is probably the case with similar guidance documents.
Here are some sample guidelines from the new document:
- General/consider worst-case scenarios—In the normal course of product development, mechanical engineers purposely drop products to the floor from table height or shake them for hours in a test chamber to see if and how they break. User interface designers need to perform equivalent tests of their designs. In other words, user interfaces must be subjected to bad or worst-case scenarios to see if and how they fail…The goal is to see how well an untrained or minimally trained user performs when asked to operate an unfamiliar device; to observe people using a device under harsh environmental conditions, such as during the nighttime in a lurching and vibrating helicopter; to see what happens when people have little time to perform a critical and difficult task. By stressing the user interface, designers can learn how to make it better suited to real-world use. Note that minimization of the risk of faulty design or use errors causing human injury requires an integrated approach that includes rigorous design practices, usability testing, and risk analysis.
- General/limit user workload—Caregivers are often overworked, enduring vigorous 12-hour or longer shifts, sometimes several days in a row. This explains why some caregivers actively or passively reject medical devices that create too much mental or physical work, as discussed earlier. Caregivers will seek shortcuts and workarounds if a device distracts them from more-important tasks, even if the work-saving strategies are strongly discouraged by their institution. Therefore, designers should pursue opportunities to reduce the time required for caregivers to learn how to use and operate devices. For example, a high-quality video complemented by a quick reference card might shorten the time required to learn to operate a ventilator. As another example, a point-of-care blood gas analyzer might allow a clinician to commence with data entry tasks while it completes a calibration check or prepares to analyze a sample.
- Software user interface/menu depth— Generally, people prefer menu systems that are relatively shallow, requiring the user to navigate no more than two or three levels deep in a menu hierarchy to reach the desired content/options. This approach reduces the chance that users will consider certain features to be “buried” in the software user interface. It also reduces the time required to select a menu option.
- Software user interface/target size— Touch screen targets should be sufficiently large to facilitate rapid, error-free inputs by individuals with large fingers. Similar to physical keys on a keypad, a target size of no less than 0.5 in. (1.3 cm) is preferred. Sometimes, it is advantageous to oversize the touch area so that the associated target is actuated if the user touches anywhere close to it. This also helps to prevent parallax problems.
- Workstations/authorized operation— Workstations should prevent operation by unauthorized individuals. For example, therapeutic devices used in critical-care environments might require caregivers to enter a special code or press an enabling button before they can change the device's settings, thereby preventing hospital visitors from making adjustments.
- Workstations/functional relationships—Displays and their associated controls should be arranged in a manner that emphasizes their functional relationship. For example, gas control valves can be placed directly below the associated flow rate indicators (flow tubes) to emphasize their functional relationship.
Again, such guidelines are subject to change based on the AAMI HFE committee's review of industry comments on the draft standard. Applying such guidelines in the course of a comprehensive human factors design program promises to produce medical devices that are more effective, less prone to use errors that could injure patients, and perceptibly easy and satisfying to use. Accordingly, AAMI HE75:200X is likely to benefit people who interact directly with medical devices, including clinicians, laypersons, and the healthcare system in general.
In his book, The Tipping Point, Malcolm Gladwell discusses events that have shifted momentum in societies and industries.7 In the near future, the medical device industry may look upon the release of IEC 62366 as a tipping point. Currently, the document appears to be driving the majority of medical device companies to invest heavily in human factors for the first time. The IEC standard calls for a comprehensive approach to the design of medical device user interfaces. Manufacturers that disregard the standard will be placing in jeopardy their regulatory approval and device certification (obtaining a CE mark, for example). But, like its root document AAMI HE75:2001, IEC 62366 does not provide user interface design specifics.
AAMI HE75:200X fills the gap between good design process and good user interface design. Looking forward, there will be no excuse for illegible displays, confusing controls, and incoherent software menus. So, medical device manufacturers, prepare yourselves. Our industry is well past the point at which human factors may be regarded as newfangled. The rigorous application of human factors in medical device design is now conventional, and most likely starting in 2009, AAMI HE75:200X will help define the conventions.
Indeed, AAMI HE75's release should cause more relief than pain. Israelski, who spends much of his time infusing human factors into development projects, believes, “HE75 will be a great help to manufacturers that do not have trained HF staff, as well as a good tool for those that do have onboard staff. The goal is that the standard provides a lot of guidance collected in one convenient place. HE75 will save developers time in searching out the HF literature for design guidance. Each of the 25 sections has extensive references for those who need to dig deeper into design guidance for particular areas such as alarms, displays, documentation, workstations, software, etc.” He adds, “It should help all manufacturers make better design decisions. But, it is not to be seen as a cookbook or a shortcut for doing the complete HF process during development.”
AAMI has granted permission to the author to quote from AAMI's draft standard, AAMI HE75:200X. This article represents the view of the author only and not the views of AAMI or the AAMI Human Factors Engineering Committee. Neither AAMI nor the committee has reviewed this article for completeness or accuracy. AAMI HE75:200X is currently in draft form and may change substantially in the future based on the committee's review of ballot and public comments on the draft standard.
2. IEC 62366:2007, “Medical Devices—Medical Devices (Geneva: International Electrotechnical Commission, 2007).
3. ANSI/AAMI HE74:2001, “Human Factors Design Process for Medical Devices” (Arlington, VA: AAMI, 2001).
4. ANSI/AAMI HE48:1993, “Human Factors Engineering Guidelines and Preferred Practices for the Design of Medical Devices” (Arlington, VA: AAMI, 1993).
5. IEC 60601-1-6, “Medical Electrical Equipment—Part 1-6: General Requirements for Safety—Collateral Standard: Usability” (Geneva: International Electrotechnical Commission, 2006).
6. MIL-STD-1472F, “Department of Defense Design Criteria Standard—Human Engineering” (Washington, DC: Department of Defense, 1999).
7. Malcolm Gladwell, The Tipping Point: How Little Things Can Make a Big Difference (New York: Little, 2001).