Interaction design reaches deep into a product. It explicates system constraints, supports user goals and tasks, facilitates cognition, enables perception, and requires human action. The resulting user experience also extends beyond skin deep: It directly involves human musculoskeletal and nervous systems. As such, it can do them damage.
We rely on "official standards," such as the definition of usability from ISO 9241, Part 11: "The extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use."
But how easily can we test such usability? We need the highest return on our testing resources, so we mostly test with neophytes in sessions of an hour or two. We make sure our designs match users’ goals, tasks, and mental models. We verify that our interfaces are explorable, effective, efficient, and even pleasing to our test participants.
Yet our specified users and contexts of use include people exercising our designs over long periods of time. Will our testing find the consequences of long-term use? How much satisfaction has our design provided a "specified user" if using it causes a musculoskeletal disorder (MSD)? In addition to providing effectiveness, does our design efficiently strain connective tissue and crush nerves?
We know about MSDs. We all know someone who has carpal-tunnel syndrome or other damage that involves chronic pain and impairment. The four factors of MSDrepetition, force, posture, and vibrationconstitute the potentially hazardous attributes of every physical interaction we design, in software or hardware (with the possible exception of vibration… while haptic feedback remains scarce).
If interactions damage users, the designs of those interactions bear partial responsibility. Few interaction designs have proved as deadly as the Therac-25 radiation-therapy machine interface, which killed three patients in the mid-1980s, or Multidata radiation machines, which apparently killed five more in the 21st century. However, less dramatic user experiences deliver far more victims. In 2003, according to the U.S. Bureau of Labor Statistics, MSDs of all types and causes accounted for 435,180 injuries and illnesses with days away from work33 percent of all such injuries and illnesses .
But what about MSDs and human-computer interfaces? Countless studies have targeted that question. One such study  observed, "More than 50 percent of computer users reported musculoskeletal symptoms during the first year after starting a new job." As devices and interactions proliferate, our creation of MSDs constitutes neither a small nor a shrinking problem.
Neither can we dismiss it as a problem of only low-profile organizations. Take the Apple mouse, for example. This device shipped with my beloved Macintosh G5. Certainly it was designed. But was it designed well? The present design ignores readily available knowledge about optimal hand and wrist postures and the associated forces. (At least it doesn’t vibrate!) Granted, it reduces support costsbecause large numbers of Macintosh owners promptly purchase third-party pointing devices and toss the Apple mouse into a drawer.
Of course, that’s an easy target and only one example. But the point generalizes: Organizations with good reputations for design perpetrate risky interfacesrisky to users, and risky to those organizations. In this era of short design and manufacturing cycles, with generations of technology coming ever faster, products and companies have reputations. Those reputations can change as quickly as users can disseminate dissatisfaction via the Internet.
Consider the discussion about computer security and data privacy. The twin specters of government regulation and legal liability arise frequently in that debate about improving industry practices. How long will our users absorb the costs in lost productivity, healthcare dollars, and suffering inflicted by demonstrably poor designs before we hear talk of regulation and liability? How long ought we to escape unscathed if we choose to practice irresponsible design?
Does your current design projectthe one into which you are pouring expertise, energy, heart, and soulincorporate the best available knowledge about causes and avoidance of MSDs? How many unnecessarily harmful designs can we afford to add to our portfolios? Shall we wait, as did makers of steam boilers during the 19th century, until the body count takes matters out of our hands?
We don’t have complete freedom in design. Hardware components, operating systems, myriad factors lie outside the scopes of our current projects. Evolution of a more resilient user also remains out of scope. But we can question presumed constraints where better possibilities may exist. We can inform ourselves about risks to our users; we can consciously reduce those risks.
Some of us can do the basic research to extend that knowledge. Some can study research results and share the implications for design. Via usability testing we’ve generated design guidelines for the short-term benefit of our users. We must also promulgate research-based guidelines for long-term avoidance of MSDs. As interaction designers, we cannot honorably fail to attend to physical consequences to our users.
Most of us love the multidisciplinary challenge of melding elements into a successful whole. If "risk of MSD" does not reside among the design criteria in your methodology, add it. Still in doubt? Just ask that MSD sufferer you know, whose carpal-tunnel syndrome presents challenges raising a glass to toast interaction designers.
2. Gerr, F., Marcus, M., Ensor, C., Kleinbaun, D., Cohen, S., Edwards, A., Gentry, E., Ortiz, D., & Monteilh, C. (2002): A prospective study of computer users: I. Study design and incidence of musculoskeletal symptoms and disorders. American Journal of Industrial Medicine, 41, 221-235.
Douglas W. Anderson
The Mayo Clinic
Computer Sciences Corporation
15245 Shady Grove Road
Rockville, MD 20850, USA
©2005 ACM 1072-5220/05/0700 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2005 ACM, Inc.