Applying Human Factors Engineering to Generic Combination Products
Frauke Schuurkamp, Managing Human Factors specialist, Emergo by UL
Lisanne de Vogel, Senior Human Factors Specialist, Emergo by UL
When developing a generic combination product, manufacturers can choose to compare their generic combination product to an already marketed reference product and analyse key differences between the products. In these threshold analyses, manufacturers are expected to compare the labeling, tasks, and physical features of their generic product with the reference product. The author will share FDA’s expectations and key terms in threshold analyses, and why and when threshold analysis will be helpful to perform.
Introduction
Human factors engineering (HFE) is a key element in developing an effective medical device and focuses on ensuring that a device’s design is well-matched to its intended users. In contrast to developing an entirely new combination product (where the HFE process involves a variety of activities and can be extensive), manufacturers can also compare their generic combination product to an already marketed reference product and analyse key ways the generic product differs from the reference product. In such a process, called threshold analyses, manufacturers are expected to compare the labeling, tasks, and physical features of their generic product with the reference product.
The FDA’s Center for Drug Evaluation and Research (CDER) published a draft guidance describing threshold analyses in January 2017 (Comparative Analyses and Related Comparative Use Human Factors Studies for a Drug-Device Combination Product Submitted in an ANDA: Draft Guidance for Industry). The guidance states that when developing a generic combination product, manufacturers “should carefully consider the design of the user interface (UI) of a proposed generic combination product and seek to minimise differences from the UI for the reference listed drug (RLD).” The objective of this analysis is to enable the generic product to be substituted for the RLD product without additional training or explanation from a healthcare provider (HCP).
Preparing for a threshold analysis
Assess device similarity
When developing a generic drug, manufacturers need to select a device for the drug. They can develop a new device or use one of the many available platform devices. Noting that the objective of the generic combination product is for a patient to be able to switch from using an RLD to a generic combination product without additional training or explanation, device similarity must be kept in mind when selecting a suitable (platform) device. In other words, a threshold analysis is not about innovation but about creating something that is similar to something already marketed.
After selecting a device, UI design changes might be necessary to make the generic product more similar to the RLD. For example, if there is a difference between the RLD and the new generic product in an injection device’s dose dial or injection button, consider changing the dose dial or button design for the new generic product to avoid identifying this potentially critical difference at the end of the development process when the design might be frozen.
In addition to making the generic product itself more similar to the RLD, the labeling of both products (including the Instructions for use (IFU), label and packaging) should be designed as identical as possible as well. This means that the exact text, graphics, and layout on the generic’s labeling should resemble the designs on the RLD as closely as possible.
Sourcing the right materials
To effectively prepare for a threshold analysis, all materials in scope for the analysis (i.e., the RLD and generic product) should be physically available. This does not only include the physical device but also the associated labeling, including the IFU and packaging. Importantly, the RLD materials should be the most recently marketed available materials and the generic product should have the final labeling.
Having these materials physically available is helpful to simulate using both devices to assess the differences. Similarly, being able to physically hold both devices will show any difference in device characteristics, such as the injection button’s push force or the grip.
Conducting a threshold analysis
A thorough threshold analysis evaluates all aspects of the UI. Specifically, there are three types of comparisons that manufacturers should perform to compare the proposed generic and RLD products: a labeling comparison, comparative task analysis, and physical comparison.
Labeling comparison
The labeling comparison is a side-by-side, line-by-line comparison between the proposed generic and RLD products’ labeling, including the IFU, label and packaging. The labeling comparison is an important and detailed task that should be driven by the thought that a design difference can introduce new use-related risks or use issues.
| Sample labeling aspects | |
| Alphanumeric text size and style | Organisation / layout of content |
| Format of special content | Line and paragraph spacing |
| Graphic style and quality | Paragraph versus bullet-point text |
| Graphic format | Phrasing |
Comparative task analysis
A comparative task analysis focuses on how the manufacturer intends for a user to interact with a product. The analysis serves as a systematic “deconstruction” of the user-product workflow, mapping out all individual use steps. At first glance, identified differences might not seem to introduce new risks, but they will require further analysis per the FDA’s guidance.
| Sample task analysis aspects | |
| Open carton | Confirm expiry date has not passed |
| Remove product from packaging | Dial dose |
| Attach needle | Push plunger |
| Remove inner needle cap and outer needle cap | Store product in dark and dry environment |
Physical comparison
The physical comparison covers a thorough visual and tactile examination of each product’s physical features and characteristics, and covers a variety of physical aspects.
| Sample physical aspects | |
| Actuation force | Indicator location and color |
| Audible and tactile feedback | Label positions |
| Physical dimensions | Material finishes |
| Weight | Shape |
Assessing differences
Categorising differences
Each identified difference in the threshold analysis should be categorised based on whether the difference impacts an “external critical design attribute.” The FDA defines this term as “features that directly affect how users perform a critical task that is necessary in order to use or administer the drug product.” For example, a needle cap is likely to be an external critical design attribute because it mitigates against a needle stick injury (i.e., a use error related to a critical task). After having determined the product’s external critical design attributes, the design differences between the products can be placed into one of the following three categories:
- No design difference. A given UI feature, function or characteristic is the same between the proposed generic and RLD.
- Minor design difference. A UI difference does not impact an external critical design attribute and should not affect the performance of a combination product-critical task.
- Other design differences. A UI difference that may impact an external critical design attribute involving drug administration.
Analysing design differences
After categorising each difference, it is time to justify that the generic product is still safe and effective to use, even though there are design differences. In other words, it describes why the differences do not affect product substitution in a negative way.
Presenting a strong rationale for each finding can reduce the chances of the FDA seeing the differences as negatively impacting the users’ experience. The FDA will likely judge minor differences as acceptable if the threshold analyses are comprehensive. However, if the threshold analyses identify one or more “other design differences,” the FDA will likely ask the manufacturer to modify the proposed product’s components (including labeling) or request additional human factors data through a Comparative Use Human Factors (CUHF) study.
HFE implications – keeping the end user in mind
When evaluating and categorising the differences, it is important to keep the product’s end user in mind. For example, if the RLD provides an audible “click” sound after dose delivery but the proposed product provides a “click” when dose delivery starts, think critically whether that difference might confuse the user. If the user mistakes the click made by the proposed product as the signal that the dose has been delivered, this will likely have a critical impact on whether the user receives the full dose.
When thinking about the end users of the product, also consider other performance-shaping factors and characteristics that could influence how they interact with a product, for example, a medical condition or an educational background. FDA recommends particular consideration of how the identified differences affect performance by lay users (i.e., patients, caregivers) because they are likely less accustomed than HCPs to encountering different medical UIs, posing an increased risk for (critical) use errors.
Next steps
Submission contents and format
The FDA does not describe how the threshold analysis results should be presented (i.e., the FDA only states that differences must be analysed and categorised). As such, there are different ways to document the results.
One way is to present the results from the labeling, task and physical comparisons in dedicated tables, describing and/or depicting each difference in a dedicated row, and then indicating the categorisation (no difference, minor difference, or other difference) and a brief rationale to justify that a product can be substituted without additional intervention. For example, if categorising a different injection button color as an “other difference”, the justification could be that even though pressing the injection button is a critical task, this difference is not expected to affect product substitution because users will likely be able to identify the injection button, and the IFU images represent the button’s actual appearance.
FDA feedback
After completing and submitting the threshold analyses, the FDA will carefully review the RLD and the new generic product, including the detailed categorisations and justifications of the design differences. “Minor differences” are often considered acceptable with well-written justifications, and in that case, additional HF-related data is likely not requested. That said, in many cases, a threshold analysis will have one or more “other design differences,” which will get particular attention from the FDA during their review. Notably, even when a design difference may seem to be an improvement, the FDA might still question the difference without any HF data.
For each “other difference” identified, the FDA might disagree with the justification or request more information. In those cases, providing additional or more detailed justifications might help convince the FDA; if that is not possible, one might consider changing the product’s design (e.g., the physical device or labeling) to eliminate those “other differences.” If no further design modifications are possible or implemented, the FDA might request additional HF data through a CUHF study to demonstrate that potential users of the new generic product will not commit more mistakes with the generic product than with the RLD. Notably, we recommend consulting with the FDA to discuss the path forward and avoid making major decisions without confirming the FDA’s expectations for your specific product.
Conclusion
When developing a generic combination product, manufacturers can compare their product to an already marketed RLD to justify that their proposed product is sufficiently similar that users can use the generic product as safely and effectively as the RLD.
A thorough threshold analysis compares the labeling, task analysis and physical features of both devices. Each identified difference shall be classified as “no difference,” “minor difference,” or “other difference,” depending on whether the difference impacts an “external critical design attribute.”
Finally, a key step is to justify why the design difference does not affect product substitution by documenting the rationale for the categorisation.
Frauke Schuurkamp has 12 years of experience delivering HFE services to the medical and pharmaceutical industries. Schuurkamp conducts HFE workshops, advises on HFE strategies, leads research and analysis activities and develops key deliverables to support FDA and international regulatory submissions. Schuurkamp holds a bachelor’s degree in applied psychology and a master’s degree in human factors psychology.
Lisanne de Vogel has 7+ years of experience delivering HFE services to the medical device and pharmaceutical industries. Lisanne leads and oversees human factors research activities, such as usability testing, and helps clients develop key HFE documents for their (international) regulatory submissions, as well as advising them on how to apply HFE during product development to meet regulators' expectations. She holds a bachelor’s degree in Communication Science and a master’s degree in Applied Cognitive Psychology.
