Researchers at the Stanford University Program in Biodesign have completed a one-year study that examined how medical technology is brought to market, approved for use, and enhanced over time. First results of this work were presented at an FDA/Industry meeting organized by the Institute for Health Technology Studies (InHealth) which funded the study, and at a workshop presented at MD&M East in June. This article reports on the presentations given by the Stanford team members John Linehan and Jan Pietzsch. The full study results will published later this year in a scientific journal.

Medical-device development has become increasingly complex in recent years. The advent of new technology, stricter regulatory requirements, and the ever increasing role of reimbursement for successful commercialization require careful planning and strategy setting, coordinated decisions, and consistent, rigorous processes. The design and implementation of such processes, often captured in development models and related standard-operating procedures have become key to commercially successful devices. While various models exist in the device industry, no comprehensive development model has been published.

The team's research was to show the activities and decisions involved in the development of a device, and based on this information, differentiate device development from pharmaceutical development. The team spoke with more than 80 experts in the private and public sectors, including industry representatives and FDA officials. It became clear that most companies follow a five-phased approach to design.

Predevelopment, Phase 0

To create a new medical device, inventors and device companies must focus on the right clinical need. To accomplish this, clinical needs are first identified through direct observation, speaking with physicians, patients, other health-care providers, and personal experience. Through a funneling process, inventors narrow a large list of clinical needs based on the estimated market size and clinical impact associated with each. To further narrow the list, inventors usually examine prior art related to each need to determine if barriers to further development exist from an intellectual property perspective. A detailed market analysis is subsequently performed to ensure there is sufficient market opportunity for each need.

Inventors must also determine if they are in a position to efficiently seize the market opportunity. Once the list of clinical needs has been sufficiently narrowed, each remaining need tends to get further validated in terms of regulatory considerations, reimbursement strategies, intellectual property, and business development objectives. This leads to Phase I of the development model.

Phase I: Initiation, opportunity, and risk analysis

After identifying a need, companies perform a preliminary market analysis, financial review, and competitive product assessment. A review of the existing intellectual property within a specific market or pathology is also conducted, as well as an early stage technology risk assessment. Potential regulatory paths and their associated risks are studied within Phase I. Likewise, assessing initial reimbursement strategies is an integral aspect.

Phase II: Formulation,concept, and feasibility

If upper management accepts the project definition, development proceeds to Phase II. Concept formulation and feasibility assessment occurs during this early stage. A cross-functional, project-core team is selected, and a general project plan and timeline developed. Creation of a Design Plan is a formal requirement (per 21 CFR 820.30) and signals the startof formal design controls. The team leader is responsible for initiating and managing the Design History File, a record indicating that the device was developed as described in the approved design plan. Lack of adequate records is a frequent audit failing and can jeopardize the project timeline and ability to support regulatory filings. Early in development, the team's marketing associate and R&D engineer often meet with potential users (physicians, nurses, technicians, patients, and others) to get customer input. It also is common to generate new device ideas by considering existing product complaints, which come directly from physicians or reviewed from the FDA's Maude complaint database.

R&D in Phase II is responsible for generating early concepts. Brainstorming sessions are often held during this stage of development with members of R&D, marketing, and physician consultants. Computational analyses, such as stress and flow studies, are conducted to further understand the behavior of a proposed device. The team often develops a 3D CAD model of a proposed device, which subsequently forms the basis for the construction of physical prototypes. Throughout Phase II, R&D will hold frequent design reviews (per 21 CFR 820.30) with all cross-functional team members to systematically assess a device's design progress.

Risk management is a critical component of the analysis, prototype, and design development phase. The FDA expects companies to have a complete risk-management plan in place, which consists of the two aspects of risk analysis (identification and quantification of risks) and risk management (mitigation of the identified risks). Several years ago, the industry supported the creation of ISO 14971 which covers risk management for medical devices and outlines specific methods to identify and address risk.

Phase III: Design and development, verification and validation

After the team shapes the device concept and makes several prototypes, the development process proceeds into Phase III. In this phase, cross-functional team members generate a verification and validation test matrix. The matrix outlines the verification and validation (V&V) tests that occur before and after design freeze (Phase III and IV), as well as provide a foundation for formal validation testing in Phase IV. V&V testing is conducted primarily by research and development, test engineering, and quality engineering. Marketing often participates in validation testing, such as physician prototype evaluations. Verification and validation studies are subject to design controls including methods of documenting the studies. Without proper documentation, the studies may not be usable from a regulatory standpoint becauseall V&V studies must be reproducible.

Risk management in Phase III involves collaboration among all members of the cross-functional team, with particular emphasis on quality, manufacturing, and R&D functional areas. Design-control deliverables are updated in this phase. A process-validation plan is also created in Phase III to ensure that a manufacturing facility is in compliance with good manufacturing practices, part of the FDA's Quality Systems Regulation.

Phase III sees several additional activities conducted by members of the regulatory and clinical departments. Regulatory activities include submitting design and test data to the FDA for review and approval. If a device requires clinical trials for regulatory submission, the team's regulatory group submits an investigational device exemption (IDE) so the device can be used in a clinical study.

Phase IV: Final validation and product launch preparation

This phase is characterized by the creation of formal design drawings, final product verification and validation, sales launch preparation, and regulatory approval.

After freezing a design, the team generates formal manufacturing drawings for the new device, consisting of component and assembly-level drawings. Final prints must conform to geometric dimension-and-tolerance standards to ensure that design requirements are effectively communicated to suppliers and manufacturers. Tolerance stack-ups are also conducted on the final design to ensure that there are no mating-part interferences in a device, or between a device and another instrument with which the device interacts. Material specifications, packaging drawings, and marking and labeling specifications are also finalized in this phase.

Closure of recommended risk-mitigating-action items per the company's risk management system occur in the final verification and validation phase of device development. These accompanying design control deliverables are “living” documents which remain active throughout the life of the device.

This phase also sees the finalization of the company's reimbursement strategy for the new device. Clinical validation continues before and after FDA approval has been granted to continue monitoring device performance and possibly expand a device's indications for use.

Establishing a quality system is a big and critical task for new medical devices. If the product is a line extension to an existing device or product family, then a quality system may already be in place and working well. A quality system starts by defining, documenting, and formally approving and releasing about 90% of the business document systems, both product specific and administrative (purchasing controls). This is a resource intensive and time-consuming task on a scale similar to the development process. It is normally done in parallel with product development but cannot be done retrospectively.

Phase V: Product launch and post-launch assessment

Product launch and post-market surveillance occur in Phase V of the medical-device development cycle. “Centers of excellence” are usually hospitals, labs, or physician offices used for initial product release. Physicians in these facilities will have received pre-launch training in the use of the new device. Once it has proven successful in a limited number of medical facilities, it will be marketed and distributed for widespread clinical use. Clinical trials are frequently performed by select physicians following product launch. These trials assist in gaining reimbursement support and additional marketing literature, as well as expanding further indications for use, which requires FDA approval.

Following product launch, R&D efforts are transferred to engineers responsible for managing design changes, often referred to as process engineering groups. This functional area is responsible for continual improvements and changes made to either the product or processes that generate the product throughout the life of the device.

When a product complaint comes in from the field, one of several actions may be taken. For instance, a label change will be needed if an aspect of the device has failed that was not properly indicated on the instructions for use (IFU). A customer or feasibility change will be needed to satisfy requests of a specific physician customer. And a total product redesign may be necessary should clinical use uncover a major flaw in the design or should the FDA recall the product.

Design-control deliverables remain active throughout the life of a medical device. The Device Master Record is released as part of the design transfer process, and in it, further design changes are managed.

View Table