The Center for Devices and Radiological Health (CDRH) has issued a revision to their Human Factors guidance, Medical Device Use-Safety: Incorporating Human Factors Engineering into Risk Management (Issued July 18, 2000). The revised guidance is focused on the process recommended for pre-market approval of certain devices and presents a process of applying human factors methods during device design and development leading to a validation of use. There are specific suggestions for organizing a complete human factors submission including a report format that details all human factors work in development. The elements in this draft guidance reflect the Office of Device Evaluation's (ODE) current thinking and to make the work of reviewing and understanding a sponsor's efforts much easier, and thus, quicker to approve.

When will the guidance be approved by FDA? Current estimates for final issuance of the guidance are now year-end, 2012. There were over 600 comments from industry that needed consideration from industry and within the FDA.

How is this guidance different from the 2000 guidance? One difference is the emphasis in this guidance on preliminary analyses (user profiles, environment analysis, use error risk analysis), and formative testing as expected aspects of pre-market submissions. Also, this guidance provides a set of criteria that reviewers look for when reviewing the validation of use data you provide.

Risk not in the title? The 2000 guidance was monumental, because it tied human factors to risk management . Although it was necessary then to explicitly connect risk to HF since this was not done previously in the medical device industry, the field has evolved and the connection between HF and risk is generally understood (for instance, see risk emphasis in IEC 62366). The new guidance draft talks of "optimization" of the user interface implying focus on eliminating use related hazards and also improving the overall efficacy of interaction. Removing risk for safety is still the main interest of the agency, however.

Expectations made clear. This version of the guidance is more direct about what the human factors team at FDA expects. This is a huge advantage in outreach to industry because the new expectations for validation studies had not been communicated in this manner to industry in the 2000 guidance. It was not the authors' intention to provide a structure for human factors in the device approval process. We look forward to industry comments now that they are finally in black and white.

Use the Draft Guidance Today! We recommend using the templates in this draft now, even before official approval, because it is the best window on the group's current thinking about human factors for pre-market approval. Our hope is that it cuts the number of iterations with your reviewer down significantly!

The public meeting on infusion pumps held by CDRH in May 2010 highlighted the need for device manufacurers to present their case for safety and efficacy in the form of the assurance case model developed by Carnegie Mellon University and others. The elements of the assurance case are threefold--(1) the "claim, (2) the "evidence", and (3) the "argument." What does this mean for human factors work related to validation testing of the user interface?

Using Safety Claims to communicate testing strategies to the FDA reviewer. The Assurance Case Model is not required for devices other than infusion pumps, but the basic premise of stating your safety assertions regarding use related hazards is still very useful when developing your final conclusions regarding human factors validation. We recommend presenting these safety assertions (or, claims) early in the final human factors report, or as part of Section 4 (Task Selection, Prioritization, Characterization) which deals with use risk assessment. By providing these claims, you more easily communicate what you intend to "prove" to the reviewer. These claims can be stated in terms of their relationship to known use problems or to particular critical task outcomes, i.e. "the injector allows for accurate and complete self-dosing of a patient's prescribed amounts of insulin." Each of these claims can have sub-claims, as well, such as "users can accurately draw and administer dosages requiring multiple injections when required." Tasks related to these claims are then identified as part of your validation test plan.

In a recent paper developed by the FDA's Center for Devices and Radiological Health (CDRH) and the Center for Drug Evaluation and Research (CDER) scientific review staff, a set of "known use errors" for various kinds of combination products was presented. CDER's Division of Medication Prevention and Analysis (DMEPA) is actively inolved in shaping the nature and scope of simulated use testing for these products by routinely requesting that sponsors test for specific problems that have been found in adverse event reporting pertaining to their combination product type. We suggest sponsors understand what these problems are and provide analysis asserting that these use errors have been eliminated through design or how they will be tested in their final simulated use study.

As CDRH works more closely with CDER, their review teams will include increased human factors submission scrutiny. We suggest asking your CDER submission project manager and review team who will be (or has been) identified as the CDRH HF representative and to call them about specific questions regarding study plans and submissions.

Avoiding critical omissions and lack of clarity limits time-consuming questions and additional information requests

Human Centered Strategies has helped a number of recent clients resolve requests through letters from ODE regarding inadequacies in their Human Factors pre-market submissions. These requests for additional information or clarifications can mean valuable time lost in the process of clearing the device for market. The following list of inadequcies in sponsor submissions has been summarized based on feedback from ODE human factors evaluators over the past two years.

No Use Error History Provided: Sponsor has omitted a historical analysis of use error risks with the predicate device or similar devices.

Insufficient Detail of Device User Interface: Screen shots, sequences of operation, task analyses, and other data describing the user interface should be provided to allow the evaluator to become familiar with what the user is required to do with your device.

Lack of User-Device Interaction Risk Analysis: A comprehensive risk analysis of the user's tasks has not been provided. This includes lack of prioritization of the risks based on potential harm to the user or patient.

No Description of Formative Testing Work: If formative usability tests and evaluations have been performed by the sponsor, they are not summarized in the submission. In particular, the agency needs to know what risks the sponsor removed from each iteration of testing.

Omission of a Simulated Use Validation Study: Sponsor has submitted a pre-market document with no identified simulated use validation study. This study is necessary for reviewers to determine safe and efficient use of the device.

Insufficient Rationale for Validation Study Tasks: The sponsor has not provided reasoning for the choice of the study tasks, with respect to anticipated risks. The agency needs to understand that the extent of the tasks represent the full range of possible use error events as delineated in the sponsor's hazard analysis and risk assessments.

Validation Study Tasking Unclear or Missing: Sponsor reports did not discuss how the device system was used during the evaluations to simulate use conditions.  Please describe how the device was used by study participants and particularly the use scenarios involving critically important tasks.