HIMSS News

Introduction to Human Factors and Usability in Health IT Design

By Beth Meyer, PhD, CPE, Human Factors Engineer, McKesson Corporation

As healthcare work is increasingly computerized, there is increasing discussion around healthcare, human factors and usability. But – what exactly is human factors? And why is usability important? If it is so important, how do we encourage it?

What are human factors and usability?
Human factors is “the discipline that tries to optimize the relationship between technology and the human.”1 According to the Usability Professionals Association, “usability can be defined as “the degree to which something…is easy to use and a good fit for the people who use it.”

Why is usability important?
The design of tools affects how well work is executed. Imagine a carpenter with a toolbox that is missing a hammer. While the carpenter could use the “workaround” of driving in nails with a screwdriver, the work would be slower and the results of lower quality. Tools fit the task are critical to good performance.

Small design flaws can have a large impact. For example, a pair of data-entry fields on a medication order form could be presented in the opposite order than they are typically written on a paper prescription. However, this would require the clinician to perform an extra step, a mental “switch,” on each order. Eventually, a clinician in a great hurry is likely to omit the “switching” step because the tool does not support how they think, and an error will occur. A central conclusion of the 2009 NRC report, Computational Technology for Effective Health Care, was: “…the nationwide deployment of health IT will not be sufficient to achieve the vision of 21st century healthcare, and may even set back the cause…. [Success] will require greater emphasis on providing cognitive support for healthcare providers.”2

What is User-Centered Design?
Usable software doesn’t happen by chance – it is created with a user-centered design process, a set of methods to address user needs throughout the product life cycle.
The first step is user and task research. This starts very early in a project, but can continue in parallel with other work. Designers must know about the work the tool will perform – who performs it, how and in what environment. The research process need not be lengthy but is critical to solving the right problems.

Research informs artifacts, such as personas and task analyses, which guide the design process. Personas are fictional characters with typical traits of real users; they help designers to concretely understand their users. Task analyses show the order of steps and decisions, helping to optimize the workflow.

As designers learn more, they propose design concepts. When the team has sketched several feasible designs, they can evaluate the designs with potential users. This is not a formal presentation to a CIO or focus group; rather, it is a form of early usability test in which individual users envision working directly with the designs. This process identifies problem areas and often generates improved designs, which are evaluated again in subsequent iterations.

Throughout the project cycle, a human factors professional can help ensure the product consistently follows best design practices. They may perform design reviews periodically, but have the greatest impact when more regularly involved in development.

Just as software undergoes quality assurance (QA) testing to ensure it functions well, software should also undergo usability testing to ensure it supports users’ work. A usability test session features a real user attempting to independently perform a typical task. Usability testing can use early sketches, prototypes or initial versions of the product itself. Usability tests can uncover problems with layout, controls, wording, navigation, etc. For example, one might find out from a usability test the screen layout doesn’t match the users’ workflow.

What can clinicians do?
If there is an area where product design could introduce problems, speak up. Poor design is often a factor in what we accept as “user error.”

Healthcare organizations can do their own usability testing as they make purchasing decisions. (e.g., Seto et al, 20063). Vendors will increasingly commit to user-centered design if their customers insist candidate products meet measurable usability standards. It is also perfectly acceptable to ask a vendor to see their usability test results. If none can be produced, the product’s usability might need to be examined more closely. As Nemeth et al found in 2009, a real purchasing instance, extraneous factors, can affect the usability of clinicians’ tools.4

Usability Myths

Usability Facts

Automating a process inherently makes the job easier and less error-prone.

Automation that is designed without attention to the needs of the user may make the job more difficult. It may not give enough information for the user to know the true situation or it could give the user too much information or too many alerts, masking what is essential.

The most important measure of usability is the number of keystrokes or mouse clicks to do a task.

While counting clicks can point to inefficiency in some processes, there is much more to usability than counting clicks. A process with fewer clicks may have too many modes of operation that change how the system behaves or may require too.

Developers needn’t think about usability until the product is nearly complete.

There is no spray-on usability. Many fundamental design decisions, made early in the design process, are difficult to change later and deeply affect the usability of a product.

Usability is entirely a matter of subjective opinion and cannot be measured.

User-centered design teams set measurable, usability criteria early in the process, and then perform usability testing to measure how the system performs. Some sample criteria:

  • Users will consistently enter new medication orders in under 90 seconds.
  • Users will consistently find a patient’s latest lab results upon first using the new system, without errors or requests for help.

If the product designers are good enough, there is no need for usability testing.

Even the most skilled designers are not mind readers. Who would buy software from a company that boasted, “Our programmers are so good we don’t use a QA process?”

References:

  1. Kantowitz, BH, Sorkin, RD. Human factors: Understanding people-system relationships. New York: John Wiley & Sons; 1983.
  2. Stead, WW, and Lin, HS, eds. National Research Council Committee on Engaging the Computer Science Research Community in Health Care Informatics. Computational Technology for Effective Health Care: Immediate Steps and Strategic Directions. Washington, DC: National Academy Press; 2009.
  3. Seto, E, Roach, C, Chagpar, A, MacDonald, A. Usability evaluations as part of the procurement process: Case study of hospital point of care carts. Proceedings of the Human Factors and Ergonomics Society 50th Annual Meeting (1095-1098). Human Factors & Ergonomics Society: Santa Monica, CA: 2006.
  4. Nemeth, C, Nunnally, M, Bitan, Y, Nunnally, S., Cook, RI. Between choice and chance: The role of human factors in acute care equipment decisions. Journal of Patient Safety. 2009; 5: 114-121.