Design controls are required for most medical devices marketed in the United States, Europe, and Canada. The “controls” are actually tasks such as human-factors studies, risk analyses, and descriptions of how different disciplines in your company communicate as a design evolves. Design controls have been incorporated into the ISO 13485:2003 and the 21 CFR 820 Quality System regulations in response to a growing number of recalls related to design issues. In today's complex and computer-driven world of medical devices, design controls are critical to a safe and effective device.

Data published by the FDA for 2003 (www.fda.gov/ora/about/enf_story/default.htm) shows a significant rise in the number of medical-device recalls classified as “having a reasonable probability that use of, or exposure to, the product will cause serious adverse health consequences or death” (Class I recall). Many of these recalls are the result of design flaws that could have been minimized with closer adherence to what was defined in its design controls.

Implementing effective design-control procedures also improves design efficiency. They are not just academic exercises. Planning, documenting, and reviewing the design process as required by the regulations can cut development time, improve communications, and solve problems that arise in design work.

The FDA wants device manufacturers to maintain a Design History File (DHF) for all Class II and Class III devices along with those that contain software. An FDA Inspector will review a product's DHF during an inspection. For a PMA (premarket approval) device, companies are required to provide a Design Control Procedure as part of the submission and the Design History File will be reviewed during the PMA preapproval inspection.

Here are ten ways to improve compliance with design-control requirements, and help ensure the safety of your device. Most are required for a product's DHF and the others are good ideas that show a conscientious design.

1. Define Intended Use and Target Market

   The first step to a well-designed device is to record how and where it will be used. A device will have different considerations depending upon whether it will be used by a trained professional, a patient, in a surgical suite, or a field-emergency setting. It is also important to understand and define user needs — not just device capabilities. If its capabilities do not address an identified need, it may not be useful to an intended user. Intended use also determines what's called a regulatory strategy.

2. Regulatory Strategy

   The classification of your device will in large part determine its regulatory path, so establish a regulatory strategy as early as possible. This control area shows how federal regulations, business conditions, and the design influence each other.

   For example, if your device falls in Class I, it's exempt from Premarket Notification and requires only registration and listing under 21 CFR 807.20. It's still necessary to implement a Quality System (21 CFR 820). A device falling into Class II requires Premarket Notification (often called a 510(k)), (21 CFR 807.81) and may need clinical trials. And Class III devices might require Premarket Approval (PMA, 21 CFR 814) and an Investigational Device Exemption (IDE) to perform clinical trials (21 CFR 812). Information on how the device will be classified, device registration and listing 510(k), and PMA submissions can be found on the FDA Web site under Device Advice.

   Regulatory considerations can also influence decisions on where to introduce the first product. The United States, Europe, and Canada can have different requirements for the same device. Understanding these regulations can be important in time-to-market.

   Fees dictated by the Medical Device User Fee and Modernization Act (MDUFMA) may also impact a regulatory strategy. MDUFMA authorizes the FDA to charge a fee for reviewing PMAs and 510(k)s. The difference between submitting a Premarket Notification and a Premarket Approval can mean user fees of several thousand dollars or several hundreds thousand of dollars. In addition, a 510(k) (PMN) may take three to four months, but a PMA can stretch to a year or more. Regulatory strategy also determines the requirements for costly clinical trials.

   Understanding the use of the proposed device and its implications for the cost of development in time, user fees, and clinical trials allow reasoned judgment on the feasibility of the success of a new device.

3. Design Input

   Design input starts by defining user needs and translating that data into engineering terms. Defining design inputs is not as simple a task as it sounds. Base the requirements on user needs and not technological capability. Extraneous bells and whistles may add nothing to the essential function of a device. Defining required tolerances and ensuring that adequate safety is built into the device will take time and often several iterations of the requirements. And plan on updating design inputs as new information crops up during design.

   If a design is being outsourced to another firm or the project involves a variety of groups, it is critical to provide detailed and accurate design requirements before embarking on actual design activities to avoid lost time and effort.

4. Risk Analysis

   The FDA recognizes ISO 14971 as a standard for risk management. Agency inspectors routinely ask for a risk analysis during the review of a Design History File.

   Risk management should also start early in the planning stages of a device's design with an assessment of its hazards. Risk should be reassessed as the design proceeds and culminates in a final risk report. This should describe the total residual risk, if any, after taking all actions to reduce risk.

5. Human Factors

   Designing for human factors is more important as companies design more complex devices. For example, a recent presentation by the FDA indicated up to one-third of the about 80,000 medical reports per year involve human error. (“IEC Develops Standard for Medical Device Human Factors Design”, (www.fda.gov/cdrh/humfac/sandiego4.ppt)

   Designing for human factors considers the end user, his or her physical condition, education, and circumstances of use. The FDA expects to see a human-factors analysis for devices intended for use by patients as well as those used by professionals. And think beyond ergonomics. Human-factors design is also critical for a computer's user interface.

6. Internal Communication

   Many new medical devices involve diverse groups in the design process. While engineers are developing hardware and electrical diagrams, programmers are writing software to drive the hardware, and others are developing materials that may have to be biocompatible, autoclavable, or meet other requirements. Communication between these various groups is critical to an efficient design. The FDA wants to know how and when you plan design reviews. One frequent communication problem is the failure to work with the most recent design revision. The error can lead to wasted months and even an ultimately flawed device.

7. Verification and Validation

   Understanding the requirements of design verification and validation early in the process trims time and effort from development cycles. In general, verification and validation should be as independent of the design developer as possible. This ensures that habits of the designer will not mask problems that arise in the hands of different users.

   Other verification and validation issues include biocompatibility, sterilizability, reliability, and operating conditions. And some devices will require formal clinical trials.

   Software validation gets particular FDA attention. The Agency's requirements are detailed in the document, “General Principals of Software Validation: Guidance for Staff and Industry” (www.fda.gov/ohrms/dockets/98fr/97d0282gdl01.pdf). Software validation should include a traceability matrix that links specification to source code and test cases to code and requirements.

8. Transfer to Manufacturing

   Outputs from a design program must include information to manufacture a device. Of course, designing for manufacturability is often easier said than done. But no matter how beneficial the device, if it cannot be sold at a profit, it is not useful.

   If the design is to be transferred to a contract manufacturer, consider its quality system as well as its technical capability. The company that markets the device is responsible for ensuring that it's manufactured in compliance with the requirements of 21 CFR 820.

9. Review of Changes to a Marketed Device

   Design controls never really finish as long as a device is in the market place. Changes to the design or software, whether triggered by corrective actions or improvements in the product, are subject to design control. No matter how small a change, it must be evaluated for its impact on product performance, safety, or use. Changes require a reevaluation of risk and verification, or validation, or both. And if the changes are to software, they will require validations.

10. Reimbursement

    While not strictly related to Design Control, reimbursement has become a hot topic as more novel devices receive approval. You have to know whether or not insurance covers the device and at what rate. This makes it critical to understand similar devices on the market. A device costing significantly more than an expected reimbursement may be in financial trouble. Reimbursement concerns can also impact the design of clinical trials.

Other links of interest

21 CFR 820.30 Design Controls, www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?CFRPart=820&showFR=1

User Fees for PMA Supplements, www.fda.gov/cdrh/mdufma/guidance/1201.html

Guidance on Disputes Concerning MDUFMA User Fees, www.fda.gov/cdrh/mdufma/guidance/1303.pdf