Originally Published MDDI February 2005

Roundtable

Medical Device Design: Developing a Winning Strategy

Design experts discuss the essential elements of setting a design strategy that results in the development of successful products.

When medical device companies create a design strategy for a product, they must consider everything from company culture to financial investment to regulatory requirements. Four design experts discuss these and other pieces of the puzzle with the aim of presenting a clear picture of what makes a winning strategy. The participants included David Warburton, director of engineering at a medical device company in Acton, MA; Stuart Karten, president and owner of Stuart Karten Design (Los Angeles); Perry Mykleby, group product manager for Gyrus ENT (Bartlett, TN); and Michael Wiklund, a design consultant and former vice president of human factors research and design of the American Institutes for Research (Concord, MA).

What are the essential elements of a design strategy?

Karten: Determining a strategy requires a holistic approach. In other words, we are not just looking at the product, but the entire user experience with the product. Our designers serve as user advocates to create the most-effective products possible, examining all the elements of the user experience including everything from user manuals to product graphics, the user interface, and servicing issues. For example, what happens when this product is at the end of its life? What do you do with it, and how do you disassemble it?

Mykleby: The objective and the team are two essential elements. But it starts with the underlying reason for the proj-ect. Too often a company designs products that do nothing more than introduce feature creep and that don't accomplish anything of significance. You've got to look for the 2x and 3x advances, not just the x + 1 and x + 2 advances. In healthcare devices, if you're not solving the customer problem or introducing delight factors—those product attributes that lend utility but that were unrecognized as needed by the user beforehand—you're going to limit your success. Product change then becomes a self-serving exercise, and the customer picks up on that. Having said that, nothing says that you can't solve both the business need and the customer need. But you've got to keep user need uppermost in your mind during the entire process.

Wiklund: Another essential element is a readiness to spend the money up front to approach design in a comprehensive manner that involves the end-users. This is something that some developers skip because the major payoff comes after the product has been brought to market. Too many product developers believe that integrating usability, user interface, engineering, human factors, or industrial design is a luxury they cannot afford when, in fact, it is a necessity. Fortunately for the medical industry, several developers have made good industrial design, good user-interface design, and excellent materials their hallmark, giving them a winning strategy.

Warburton: One essential element in a good design is a realistic approach to the feature set. It is important to identify the essential product features and to balance those against both product cost and project schedule to come up with a product that meets the users' requirements at a price that they are willing to pay. Oftentimes design teams do not come to grips with that early enough. There are going to be hard trade-offs and compromises in features against cost and schedule. Consequently, somewhere in the middle of the project, everything must be reassessed. By that time, either the cost of the product is too high, the product does not have an essential feature, or the schedule cannot be met.

How would you recommend that a team develop an approach to keep these problems from happening?

Warburton: A team must understand that at the beginning of the project you know the least about what the product requirements are. And that is when you are expected to come up with the customer requirements. It is essential to be flexible about the requirements document. There are some fundamental requirements that must be defined early in the product development process. These requirements set the lower boundaries on both product cost and project schedule. On the other hand, some requirements have less effect on fundamental design, so you can and should decide on them further into the project as your learning increases and as you bring prototypes in front of the customers. Therefore, it is important to have a controlled process for evolving the specifications as the project progresses.

What other factors are important to the design strategy?

Wiklund: It depends on whether you are developing commodity products or one-of-a-kind innovative products. It could be argued that you can be less concerned about design excellence if you are developing a commodity product for which the functionality is well established and the chief objective is to minimize cost. The counter argument would be that if you are going to succeed with a commodity product, your product needs something besides price to differentiate it from the others. Because the medical product market is price-sensitive, an innovation that reduces costs dramatically might give you a breakthrough. Alternatively, a developer can hold the line on price while elevating design quality, leading to products that simply look, feel, and work better than others.

How does a company determine its strengths and weaknesses in terms of its ability to set out a design strategy?

Mykleby: It is painful to self-assess. If the culture of an organization is one where it is okay to not be perfect, and it is okay to be wrong every once in a while, that is a company that will allow some freedom and will be able to be more flexible in setting a strategy. The culture of an organization is extremely important.

Who must absolutely be involved in setting a strategy?

Warburton: Typically the medical device industry is driven by customer need more than by technology, so at the beginning of a design cycle, it is crucial to have marketing driving that process and setting the customer specifications.

Karten: Anybody who is a stakeholder in the project—which would include marketing, engineering, manufacturing, sales, and service, as well as upper management—should be included at the onset. You want to get anybody involved who is going to have anything to do with the project.

Mykleby: The strategy must include the project manager, finance, and the core design team. The project manager is obvious. And, you need buy-in from whoever will be responsible for paying the bills. The core team members from the required functional areas of the business (R&D, marketing, sales, manufacturing engineering) should understand the project scope so they can weigh in on time frame and feasibilities.

Wiklund: You need to get people involved who are adept at forming a comprehensive vision of the final product, even though the vision is likely to evolve during detailed development. It is essential to bring in the most visionary people available. It could be someone from senior management or an empathetic and imaginative designer.

Is that visionary person easy to identify in the company? How do you determine who such people are?

Wiklund: I suggest searching for people with a good understanding of the product use environment and who also understand the design process. Find people who can effectively represent the users, such as human factors specialists. It also helps to find people who have past experience working together as a team.

Warburton: I agree. If the team has worked together for a long time, you learn who can be creative and who can look beyond an immediate linear extension of the product you have now. We often find a customer champion who can guide the process. Typically in the medical device industry, customer champions can help guide the development and choose the feature set because they are the ones who are actually going to use the product on a day-to-day basis.

How crucial is it to bring senior management in at a given stage?

Karten: Our best experiences—and those where we had the best results—have been those that come with top-down support. It is important to have a direct ear to the top from the beginning. If those people are championing the project and are fully indoctrinated, you increase the likelihood of a much more successful program.

Wiklund: In progressive companies, senior management usually insists on being involved from the start. In any case, it is essential to involve senior management at certain milestones. Otherwise, you can upset the applecart by bringing senior management in later to judge a fairly mature design. If the design does not strike them right, you may be sent back to the beginning. So it is important to get management support and feedback early. You need to get management invested early in the design process so that they can lend their support, and the development process is less subject to upheaval.

How would you determine what the milestones would be?

Wiklund: Senior management should concur with the overarching vision you have formed for the product. Then they should review alternative conceptual designs. A bit later, they can offer their feedback on a chosen aesthetic and interactive style. Finally, they should be involved near the end of the project when you are validating the design by means of a usability test, for example.

How does the size of the company affect the design strategy?

Karten: The strategy does change according to the size of the company. It is also determined by the company's concerns regarding speed and getting the product to market. Some start-ups do not necessarily have formal procedures in place. As they grow, more formal procedures become layered in.

Warburton: Typically small companies, particularly start-ups, in the medical device field are founded because somebody has an idea. Whether a surgeon or scientist, a creative individual comes up with an innovation and a company is formed around that innovation. So, that founding individual typically drives the new-product development project in a small company. It is not a process, but rather it is driven by that one person's concept of what the market is looking for.

The transition that a start-up must make to move beyond that founding idea to become a mature company is to formalize its product development process. It has to go from a company with one successful idea to a company that can generate successful ideas year after year.

Then as a company gets larger, it suffers from another problem, which is that in order for a billion-dollar company to grow at 10% a year, it needs a tremendous number of new product ideas every year. At that point, it starts to look outside either through independent consulting groups or by going out and acquiring these small, one-idea companies.

Buying one-idea companies is one way a large company can manage the risk of innovation, because the large company can afford to let the market determine the successful ideas, then buy the winners.

Wiklund: Company approaches are idiosyncratic. Some multi-billion-dollar organizations draw on consultants almost exclusively while others maintain an in-house design group. The first approach is the broker model, whereby someone internal knows where to send design teams for good design support. Taking this approach, a company can seek best-in-class design support on an as-needed basis.

Contrast that approach with organizations that set up an in-house design department that provides a constant source of support and develops institutional memory of a company's design needs and constraints. The effectiveness of these models depends on the volume of design work and the talents of the broker or design manager. I endorse a mixed approach, whereby you have a comparatively small, centralized group that establishes design standards for the company that yields a common look and feel across product lines but also reaches out as needed for design support. This way, you get the best of both worlds: steady design support from people with deep subject-matter knowledge and the occasional infusion of new ideas from outside the company. Large organizations like Philips, GE, or Siemens have large internal groups but also work with consultancies.

How much does the regulatory approval process drive design strategy? How should a company consider the regulatory issues in setting a design strategy?

Karten: It is part and parcel of the strategy. It sets up all of the gates you need to go through and to some degree drives the strategy. You cannot ignore it. You have to address it. On the other hand, you do not want it to stifle creativity and innovation.

Wiklund: When developing medical devices, it is critical to take a structured design approach. You have to link design requirements to design solutions consistent with the design control process prescribed by FDA. All companies, even small ones, are obliged to link their design solution to an original set of users' needs.

Karten: Typically we address it in the beginning to get it into the consciousness of the project and of the designers. We then let it go a little bit to loosen the reins and allow for creativity to flow. At some point we know we have to reconcile against those regulatory issues, but because our designers have experience in this realm, we can do this with relative ease.

Warburton: Regulatory issues must be brought in very early on, especially in a small company because your regulatory path—whether it is 510(k), exempt from 510(k), or PMA—has a major effect on your timeline and the funding that you need to go forward. So the regulatory path to approval must be considered when coming up with those fundamental customer requirements. Particularly in a small company, that path can tell you whether you are going to be able to introduce the product successfully with the cash you have on hand.

How important is it to have different professional disciplines on a design team?

Wiklund: It is helpful to have as much diversity as possible as long as the people can work together and are committed to achieving excellence. Beyond that you want to fulfill the goal of customer involvement. It is important to collect customer opinions at several points in the development process. Ultimately, I'd be more concerned about collecting diverse user input than input from a diversity of in-house professionals, even though the latter is undeniably useful.

Karten: I agree. Industrial designers are the reconcilers between marketing's wants and needs and engineering and manufacturing's requirements. The tools we bring to the table are what help do that: the ability to visualize, create schematics, build mock-ups, and prove out where ideas are going. These help to galvanize the team.

Mykleby: This points out the importance of team assembly. The carpenter's rule of “measure twice, cut once” applies to team construction. If you don't spend enough time identifying the right team members and developing team chemistry, you'll pay for it in the long run. We spent somewhere between 60 and 90 days pulling together a team for one product. We were successful in assembling a team with a positive, result-oriented dynamic. None of the team members—including the surgeons—had a personal dog in the fight. All we wanted to do was make a device that solved the common problems that all users were living with. And it worked.

Wiklund: You do not want your design strategy to be strictly data driven. You want some room for judgment, intuition, art, and creativity. However, once a design starts to mature, collecting user feedback in a quantitative way is useful. For example, if you are trying to decide between two aesthetic solutions, it is powerful to know that “three out of four prefer design A to design B,” and if you are trying to decide how to structure information in a software menu system, it is helpful to know how long it takes to perform various tasks associated with different structures. Those kinds of data tamp down debate and disagreement among design teams about which way to go, because it is clear from the data that one is preferred over the other.

What role do advisory panels play in a design strategy?

Wiklund: I have seen some companies draw considerable benefit from advisory panels that include people who are prominent in their field and have the imagination to look beyond the crude prototypes or initial sketches to envision completed products. They quickly grasp what you are trying to create and can offer useful guidance. Importantly, by working with the same group of individuals over time, you save the trouble of educating newcomers each time you need guidance.

Warburton: We like to bring them in after we have some mock-ups, only because feedback in the medical device industry is better when you have something that the people can actually put their hands on. We will find a consultant (a customer champion) who will guide us at the early part of the development. Then as we progress, we take the ideas developed with that champion and build mock-ups. Then we take those initial prototypes before the advisory panel to get a broader outlook. We typically will not bring the advisory panel together until we have something to show them.

Wiklund: I advocate bringing them in quite early. I often guide my clients through a process called visioning. Different people have different names for it, but it is possible for customers to picture how something would work without seeing a realistic model.

Mykleby: We wanted to create a design team that was based on a group of experts. None of us were surgeons, and we believed that surgeons should actually be integral members of the team, working alongside the designers, engineers, and marketing folks. Rather than just having team members render an opinion, we had these experts participate in the actual design of the product. Not only can you get them to be more critical, they are there to deal with the result of what they suggest.

Design teams are different from focus groups. Focus groups are horrible in my opinion. On one hand, you get groupthink, and on the other hand, you get the alpha-dog scenario. With the surgeon design team, we said, “none of you currently use our product, and we want to design the type of product that ultimately you would be proud to use. So from the very beginning we want you to focus on designing something that you would want to use more than anything else.” And that is what we got.

What input did the surgeons provide?

Mykleby: We asked them to identify the attributes of a product that made it comfortable for them to use. We said, “We want you not to think about the mechanical constraints of what we are going to try to make. We want you to design a device that you would like to do surgery with.” Out of that primordial thought process began to emerge a winning design, and the surgeon team participated in the design-refinement phase. Later, those same surgeons tested the actual production-level device in the lab, and then ultimately used it in the operating room. It's important to emphasize, though, that the surgeons did not lead the team. There was a strong product manager, but the process in sum was very democratic.

How does setting a design strategy in the medical industry differ from strategies used in other industries?

Wiklund: In the medical industry, you have to make sure you are taking a user-centered design approach. Such an approach would involve clearly defining the capabilities, knowledge, and experience of the people who are going to use your product. With medical devices, you can see migration of the products from the originally intended use environment to other environments. Take, for example, a product intended for use strictly in an intensive-care unit by highly trained nurses. It might ultimately be used on medical units where you do not have the same level of caregiver skill. Or it might be sent home with a primary-care person, such as the parent of a young child. In short, your design process has to anticipate all kinds of unplanned uses.

When do you need to consider these possibilities?

Wiklund: It is hard to back up a design process. It gets very expensive. So common sense suggests that you try to address the needs of a diverse user population early on. If a user population is likely to vary from a sophisticated clinician to a layperson, you need to find a way to design the product so that it serves the sophisticated users' needs, but also serves the needs of the layperson as well as protects them from using the product incorrectly and possibly causing harm.

Is it good practice for a product development team to draw from trends in other industries?

Karten: Absolutely. The medical industry could draw from the toy industry purely on a basis of looking at production techniques and high-volume production. We also design a lot of consumer electronics products, and there is clearly a bridge to medical when we start looking at how people respond to these products.

Wiklund: You see consumer product design characteristics crossing over into the medical industry. That is a good thing, because people become familiar with the interaction style of products like cellular phones or applications that run in Windows. If you design products in a similar manner for use in the medical environment, you give users a running start at knowing how to use them. This avoids what is called a negative transfer. You do not want to design medical products to operate so differently that it causes people to misapply their experience with consumer products. So if medical software basically works the way a lot of consumer software applications work, it means less learning and more intuitive interaction for the user.

Mykleby: Drawing from other industries depends on the project. Regardless, you can't ignore human factors and cognitive processes in the early design or the later phases of any product. We looked to Nike for benchmarking our marketing literature. Nike does a good job simplifying its complex design process and technical product attributes and communicating them in meaningful ways to a fairly nontechnical audience. In our case, nobody wanted to hear about software engineering, 3-D CAD files, and oscillating profile tuning. But we needed to find a way to communicate that a lot of thought had gone into the product we were designing.

Warburton: I would go so far as to say that the consumer market drives industrial design into the medical market. Consumer products, with their mass market, rely on industrial design, both to sell the product and to make it easy to use. A lot of attention is paid to the user interface, but the cost of that design can be spread over millions of units. Yet, experience with consumer products sets an expectation in the minds of healthcare professionals. Ten to fifteen years ago, they were using these relatively sophisticated, well-designed products at home. Then they would come to work in the hospital and use products with sheet-metal cases and with poorly thought-out user interfaces. The level of design and thought into the user interface from the consumer market drove people to look for more sophistication in medical products and to make that a selling point in the hospital markets. The problem, of course, is that the same amount of initial industrial design cost must be spread over a much smaller unit volume in the medical device market.

When do you look at other industries, and how do you take advantage of that learning curve that has already taken place?

Wiklund: Most designers pay close attention to what is going on in other industries, if only by virtue of reading design journals and visiting shopping centers. Naturally, this gives them ideas on how to enhance the products they are creating. For example, I recently advised a client to consider using a control knob similar to one in a late-model automobile. By turning the knob, you could adjust a value and by pressing it, you could lock in the value. Lots of medical devices have similar knobs, but this car's knob had a stationary center push button, reducing the chance of changing a value by rotating the knob slightly as you pushed it. It is always inspirational to see what others do well and how it might apply to your work.

Karten: We go through a process where we analyze what would be analogous to the particular product that we are designing. We try to draw out inspiration from products that are outside of the medical industry that could apply to the product we are developing.

Warburton: Our industrial designer often disassembles consumer products and uses various pieces of them in the mock-ups he builds. An ergonomic, well-designed handle is universal—it works whether it is on a stapler or a surgical instrument.

Wiklund: If a product's features have been accepted into the market, you are already ahead of the game in terms of getting your medical customers to accept that product. But a medical device also needs to be a good citizen. It has to work harmoniously with all of the other medical devices in the environment. If it is trying to be too terribly different, it creates a burden for users. If it is trying to stand out and be a real showpiece, it speaks to the narcissism of the designer and that person's desire to be noticed.

The reverse philosophy is probably a safer one and a more appropriate one when you are designing a medical device. You want to focus on how it is going to work with the other devices. It is a different process from developing a consumer product. In the medical arena, designers should be careful not to chase trends too eagerly in terms of shapes and colors and so forth, because it is not going to be a good fit. Devices have a much longer life in the medical arena, and they must be compatible both backward and forward.

Copyright ©2005 Medical Device & Diagnostic Industry