By Andy Schaudt
I hope by now the overwhelming majority of medical device manufacturers have at least heard of human factors. If you are reading this and are unfamiliar with human factors or usability, I fear that you may learn about it the hard way — a "wakeup call" from the Food and Drug Administration (FDA) in response to your 510(k) submission requiring human factors and usability testing to be performed. This type of testing, when executed correctly, will usually identify areas for design improvements to reduce use errors. Human factors should not be perceived as a hindrance to the successful launch of a medical device; rather, if incorporated early and often, it gives you the opportunity to design a better and more lucrative product.
I have an interesting job. I lead a division within a large healthcare system that provides in-house usability support for internal healthcare products, medical device evaluations, and procurement decisions. Our team also provides usability services to those in the medical device industry who require assistance getting access to users and environments for testing and navigating their way to FDA approval. In a sense, our usability team sits on both sides of the equation — we reside within a healthcare system so we are a customer (buyer), and we contract our services to device manufacturers (sellers). This gives our team incredible insight into the value that a medical device can have when designed through human factors methodologies. This value exists both in terms of patient outcomes and in cost to the buyer and seller.
What is Human Factors?
Human factors is a scientific discipline that studies how people interact with systems. Human factors engineers seek to optimize human performance by designing systems to match the cognitive and physical capabilities and limitations of users. They take an overall system perspective and, within healthcare, consider all aspects of the care delivery system, including people factors (clinical personnel, non-clinical personnel, patients, patients’ families), technical factors (tools, equipment, hardware, software, processes), physical environment factors (layout, environmental conditions), and organizational factors (practices such as staffing levels and workload, organizational values).
The FDA has developed guidance on applying human factors and usability engineering to optimize medical device design. This guidance provides a nice roadmap on how human factors and usability engineering should be utilized during the development process. The guidance classifies two methods for identifying, evaluating, and understanding use-related hazards: analytical approaches and formative evaluation. After applying these methods, there is one final step, which is performing human factors validation testing (summative testing). The purpose of this article is to present key tools you can incorporate — many at low cost — that will result in a safer product and an overall better design.
One of the earliest and most critical steps early in the design process is to differentiate between the "customers" and "intended users." Often the end user of a product is not the customer, and therefore he or she may not know the requirements for the device. Clearly, balancing the needs and desires of both can be challenging, but it is a critical delineation to encourage the safest, most usable design.
Interviewing intended users can be a very quick and powerful tool as early as the concept stage, and should also be used throughout the entire design and development process. In order to collect objective information it is important to employ interview techniques that do not lead interviewees to the concept currently being investigated. First, let them identify the pros and cons of current methods and devices they are using. Next, design the interviews so that the users can propose their own potential solutions to the problems. This can provide insight into environmental influences of potential importance for design considerations.
Ethnographic observation is one of the most valuable techniques to use in human factors design. The ethnographic method is direct observation of the device of interest (or similar devices) being used by intended users in the natural environment. It is recommended that this observation not interfere with normal operating procedures or bias the performance and behavior of the individuals. Ethnographic observations generally take place over long periods of time, are systematic in nature, and are carefully documented. The result is a gold mine of data to inform the design of a user-centered device.
Using information obtained from interviews and observations, you can then draft one or more use cases. A use case is a written description, from a user's point of view, of a device as it is put through specific tasks. These are, in a way, small task analyses beginning with a user's goal and ending when the goal is fulfilled. Working with users to build use cases is a great way to understand the use workflow of the device and potentially identify failure modes of the legacy systems in place (if any) that will influence the new design.
Traditionally, engineers will perform either a design or process failure mode and effects analysis (FMEA) as a way to analyze the risk factors associated with a device. To understand use-related risks, human factors engineers perform a use FMEA. The backbone of this FMEA is a detailed task analysis of everything a user can do when interacting with a device. Failure modes can then be identified and engineers, using different scales, can then rate the severity, probability of occurrence, and likelihood of detection of each. The results of the risk analysis identify areas of weakness in the device design and/or training program and materials. Furthermore, the tasks that resulted in these failure modes can then be included in formative and summative user testing protocols. A use FMEA is critical to designing and executing successful formative and summative evaluations.
Another standard technique unique to usability testing is the heuristic evaluation. The goal of a heuristic evaluation is to assess compliance of a device or system with regard to a predetermined set of human factors design principles, or heuristics. Heuristic evaluation is a “discount usability evaluation” technique used to identify major usability problems in a timely manner within a reasonable cost. This technique requires evaluators to apply a set of usability heuristics to a design, identify violations of the heuristics, and assess the severity of each violation. The heuristics for medical devices are available in the published literature by Zhang et al (2003).
Formative User Testing
After your team has deployed the appropriate analytical approaches to early-stage design, and you now have one or more functional prototypes, formative user testing can be performed. The goal of this testing is get users to interact with the device and run through realistic tasks and scenarios identified as critical during the earlier task analysis and use FMEA. It is recommended that at least four users per user group be used, with a maximum of eight. In addition, it is sometimes useful to review data received after just a few users and stop testing if significant changes are needed. (This technique can be useful to avoid burning through participants, who can be difficult to recruit for some studies). You will perform formative user studies for each new prototype (iterative testing) until the product has been finalized.
Another technique that is quite common in the medical device industry is utilizing usability studies to evaluate labels and comprehension of instructions for use (IFU). Label and IFU comprehension studies can be performed using a variety of methods, but an efficient method is interviews and questionnaires before, during, and after formative user testing, when participants are seeing the materials for the first time.
The summative usability evaluation is the final validation intended to demonstrate that all risks have been identified and adequately addressed, and that the impact of residual risks is minimized or acceptable. The summative phase can begin after any final changes are made to the design and a production ready model of the device has been produced. It is expected that the study protocol will closely follow the format of those conducted during the final formative evaluation.
Summative testing for the FDA requires a minimum of 15 participants for each targeted user group and takes place in an actual use environment or simulation lab. Task scenarios are similar to those used for the formative evaluations and reflect high fidelity situations in which the device will be used. Potential error types or device malfunctions may be simulated in order to validate that the user is provided with enough fail-safes against safety hazards or device misuses. If needed, the study protocol can be submitted to the FDA for review in order to address any concerns or questions that may arise with regards to regulatory compliance. Based on experience, review times may take anywhere from two weeks to two months.
Many of the tools and methods described above can be performed by medical device manufacturers, provided they have guidance from internal or external human factors engineers. Incorporating human factors early and often throughout the design can result in a better, safer, and more usable design. In addition, you can proactively avoid the high costs associated with device recalls and making changes to designs late in the development process. Do not fear the cost and complexity of incorporating human factors and usability engineering into your product development cycle. Embrace it and reap the benefits.
About The Author
Andy Schaudt is the director of usability services for the National Center for Human Factors in Healthcare at the MedStar Institute for Innovation. In this role he plans, coordinates, and manages the projects, programs, and daily operations for the usability division, which is chartered to conduct medical device and health IT usability evaluations, both for the industry and for MedStar Health (a 10-hospital healthcare system in the Washington, DC / Maryland region). The author can be contacted through www.medicalhumanfactors.net.