In the last post, I alluded to the possibility of using software prototyping tools* as aids to EHR usability testing and product selection. While there are many types of prototyping tools, I have been most intrigued by those that allow the creation of interactive user interfaces. Using these tools, I have been able to experiment with web and mobile screen layouts, colors, components, and interactions to my heart’s content. Changing layouts and components is easy, which makes interface adjustments painless. Consider how applications that allow one to mimic EHR interfaces might come in handy as usability, research, and educational tools.
User interface issues are one of the most common types of EHR-related complaints because poor user interfaces affect both productivity and safety. This is common knowledge; the question is what to do about it. Below are a few examples of interface issues that frequently draw complaints.
|Size||Screen element too small or large|
|Color||Poor use of color for critical information
Inconsistent use of color
|Layout||Too much information on screen at one time
Finding needed information requires opening additional screens
|Workflow||Too many clicks to accomplish common tasks|
All of these issues can be identified through structured interface testing. This is not as forbidding as it sounds. Using test scripts as an adjunct to EHR selection counts as structured testing for buyers. Vendors usually make use of structured testing during beta-testing before releasing products. Given the importance of the user interface, prototyping software seems like the perfect what-if tool for ensuring that users have a mechanism by which to express their preferences and needs.
As someone engaged in software development, interface prototyping tools are a no-brainer for gathering requirements. Obviously, the best time to gather requirements is before any code is written. However, life is fluid, and requirements change. In my experience, even features that users are certain they want end up being under (or never) used. By giving potential users a working interface to play with, prototyping tools could help to determine the initial set of interactions and components, so that time is not wasted on questionable features. There is a lot to be said for having discussions with users based on their having interacted with an interface prototype, as opposed to discussions based solely on their having viewed a slide set and heard a description of how the interface will behave.
Once a product has progressed to the point where end users can test it, prototypes can still be useful. In response to interface issues uncovered during usability testing, prototyping tools would make it possible to evaluate additional designs prior to initiating any code changes, thereby making testing more efficient and productive. EHR interface prototyping tools could contribute to the success of HIT projects in the following ways:
- Make requirements gathering more efficient and productive by providing objective information regarding interface preferences
- Make it easier to create and test multiple designs
- Act as configuration aids by making user preferences easier to capture and compare
- Test specific work sequences and interaction patterns for comparison across users
- Act as training tools by emulating specific product interfaces
- Act as user feedback gathering tools for determining updates and prioritizing new features
Sophisticated EHR prototyping tools could offer benefits beyond those seen when developing and implementing products. Such tools could offer informatics researchers the ability to perform simulations in order to study information system properties that are difficult or impossible to examine in live systems. Alert fatigue is a good example.
Researchers studying alert fatigue on live systems cannot conduct what-if scenarios in order to see how alternative designs might affect users. Interactive prototypes could help with this. Moreover, prototyping tools, by giving informatics researchers the capability to design and test prototypes without having to rely on professional programmers or having to gain access to a live system, could spur workflow, decision support, and other types of information system-oriented research.
Prototyping software could also be used in HIT workforce training. These tools could be used to acquaint new HIT developers with clinical interface issues while allowing them to experiment with interfaces designs. Those being trained as implementation assistants might benefit as well. Finally, prototype models for a range of clinical systems, if made widely available (think library or repository), could be used by anyone to get a feel for clinical interfaces without having to access a live product.
Unfortunately, I am not aware of any prototyping tools designed specifically for clinical systems. However, creating some should be doable. Since high-quality prototyping software already exists, the concept has been proven–we are not starting from scratch. The main issue then is creating clinically-oriented products. Given the potential value of prototyping tools and the role HIT is increasingly playing in national healthcare policy, the ONC might want to consider encouraging the development of such tools. The ONC could jump-start the effort in a number of ways, such as by sponsoring a competition or providing funds for companies to adapt non-clinical products. All things considered, I believe clinical interface prototyping tools would be a boon for informatics education and research as well as HIT development and implementation. What do you think?
* Here are two links to applications (Prototyper, Prototype Composer) that are good examples of available prototyping tools. (I am not endorsing these products.) Free versions of both are available. You may also wish to read this Wikipedia article on software prototyping practices.