Determining External Critical Design Attributes when Conducting Threshold Analyses for Combination Product

By Rebecca Glasner

Overview of Threshold Analyses for Combination Products

Threshold analyses help determine if differences related to external critical design attributes lead to additional use-related risks to the end user when comparing a proposed generic product to a reference product. The methodology for threshold analyses is described by the U.S. Food and Drug Administration (FDA) draft guidance document released in 2017.

In a previous article, my colleague and I discussed how to conduct Threshold Analyses to meet FDA expectations. This article will explain how to best determine external critical design attributes when conducting Threshold Analyses for combination products.

Importance of External Critical Design Attributes

In the 2017 draft guidance, the FDA outlines three types of threshold analyses to conduct in order to determine design differences between a proposed and reference product: labeling comparison, comparative task analysis and physical comparison. Once these comparisons have been completed, a manufacturer should begin categorizing the design differences.

Threshold analysis findings should be categorized based on whether or not a design difference impacts an external critical design attribute. The FDA defines a critical design attribute as a “feature that directly affects how users perform a critical task necessary to use or administer the drug product.” Design differences can be categorized as having no difference, being a minor difference or being an “other” difference. It can sometimes be challenging to differentiate minor attributes from critical attributes when conducting threshold analyses. External critical attributes are defined most easily by first determining a proposed product’s critical tasks.

Defining Critical Tasks

In their 2023 Q&A of Application of Human Factors Engineering Principles for Combination Products, the FDA defines a critical task of a combination product as “a user task which, if performed incorrectly or not performed at all, would or could cause harm to the patient or user, where harm is defined to include compromised medical care.”

When conducting threshold analyses, it is typically considered best practice to reference the proposed product’s use-related risk analysis (URRA) when determining how to classify attribute differences. The FDA has highlighted their particular focus on tasks that can impact dosing, the administration of a product or have the potential to result in harm. Additionally, manufacturers should consider their proposed product’s time sensitivity and/or urgency. Products that deliver life-saving medication, such as an epinephrine injector, are often considered time-sensitive or even urgent, and as such, most, if not all, tasks are likely to be considered critical tasks per FDA’s definition for combination products.

Categorizing Design Attributes

Once critical tasks have been defined in the proposed product URRA, a manufacturer can begin assessing which attributes of the product design impact critical tasks and which do not. Manufacturers must then categorize differences resulting from the threshold analyses into one of three user interface design attribute categories:

No difference

A feature, characteristic or function that is found to be the same between an FDA-approved product (i.e., reference listed drug or reference product) and a proposed product in development can be classified as “no difference.” For example, if the instructions for use (IFU) of both a reference and a proposed product both state to “clean the injection site with an alcohol wipe before the injection.”

Minor difference

A minor difference is a feature, characteristic or function that does not impact an external critical design attribute. Categorization of minor differences is context-dependent and will be discussed further in the section below.

“Other” differences

An “other” difference is a feature, characteristic or function of the user interface that has the potential to influence an external critical attribute. The categorization of “other” differences is context-dependent and will be discussed further in the section below.

Determining Minor versus “Other” Differences

While it may be tempting to categorize minor and “other” differences by the scope of the difference, it’s important to remember that an “other” difference is related to a critical task, whereas a minor difference is not. In other words, a minor design attribute finding does not impact the performance of a critical task, whereas “other” differences are features that directly impact how a user would perform a critical task.

For example, imagine an auto-injector (AI) as a reference product. At first glance, the only noticeable difference between this AI and the proposed product is the color of the cap—the reference AI has a blue needle cap, and the proposed product has a green needle cap. The difference in needle cap color between the reference and proposed products is unlikely to impact how, when, or if a user would remove it. Additionally, the color difference in needle caps is not expected to impact dosing, or the product administration or have the potential to result in harm. Therefore, such a difference could be (pending the context of use) considered a minor difference. Another example of a minor difference might be the text size of the IFU. If the reference product uses 18pt font, while the proposed product uses 20pt font, the difference would be categorized as minor as it ultimately does not impact an external critical design attribute.

Conversely, imagine that the same reference AI produces an audible click when pressed down against the skin to indicate the start of an injection. The reference AI then produces a second audible click after 15 seconds indicating the end of an injection. Meanwhile, the proposed AI provides haptic feedback that an injection has started once actuated but only produces a single audible click to indicate a complete injection. In this example, the difference in auditory clicks is directly related to product administration, because the difference in clicks could lead to a premature removal from the skin if the user believes that product administration has been completed, resulting in an underdose. This example represents an external critical design attribute. As such, this difference would be defined as an “other” difference.

Justify your Categorizations

After the categorization of differences, it is time to justify the rationale for each categorization. Justifications should ultimately provide strong explanations of the categorization of differences. We typically suggest a few sentences with commentary reflecting the analyses. In the IFU font size example above, where the reference product uses 18pt font and the proposed product uses 20pt font for their respective IFUs, a manufacturer may justify their evaluation as a minor difference by explaining “the slight difference in text size is unlikely to impact the substitution of the proposed AI for the reference AI.”

The FDA will carefully review threshold analyses. When submitting for a proposed product, the goal is to closely align the user interface with the reference product so that patients can switch from the reference product without further training. If the threshold analyses results in many “other” differences, a manufacturer should consider modifying the product to more closely align with the approved reference product. We recommend asking for FDA’s input throughout the submission process to avoid the possibility of postponement or delays in approval. Based on the categorization and the justification of the threshold analyses, the FDA may recommend conducting different human factors activities to move forward with approval. These activities may include a human factors (HF) validation test or comparative use human factors (CUHF) study.

For additional reading on human factors considerations for generic inhalers, including threshold analyses considerations please check out my colleague Yvonne’s recent article in OINDP News.

Contact our team to learn more about threshold analyses. Or, sign up for a complimentary account with OPUS, our team’s software platform that provides HFE training, tools and templates.

Rebecca Glasner is a Human Factors Specialist at Emergo by UL.