A. Introduction

There is a rich tradition of innovation in the life sciences, with technological advances that have improved and prolonged the lives of many millions of patients around the world. Our standard approach to thinking about these innovations is in a “technology-push” framework: that is, academic or industry scientists create a new invention, and then search for a clinical application where the innovation will ultimately be incorporated into practice. What is less well recognized is a mirror image of this process of innovation, which is also highly productive – what can be called “need-pull” innovation.

In this framework, the first step of the process is to identify an important area of clinical need. The innovator then researches that need area fully in order to gain a detailed understanding of the potential for a new solution – and then uses this understanding as the basis for inventing a new approach.

The Stanford Biodesign innovation process was formally created fifteen years ago with a goal of developing a stepwise, teachable approach to need-pull innovation of health technologies. At a high level, the process has three phases: identifying needs, inventing concepts, and implementing those concepts in order to deliver them into health care (the “3 I’s,” see Figure 1)1. Each of these phases is divided into a series of steps that incrementally lead to the creation of a knowledge base about the need and the emerging solution. In reality, the process is not linear, but cyclical – with a frequent need to loop back on one or more steps to revise and expand, as more information is gathered and new insights are gleaned.

A primary distinction and main pillar of the biodesign innovation process is a focus on needs identification. The central dogma of the process is that “a well-characterized need is the DNA of a great invention.”2 This approach is derived from design thinking methodology3, with an emphasis on user-centered design (that is, beginning the innovation process by developing a careful understanding of the user). A primary distinction of health technology innovation, however, is that no single or straightforward “user” exists. Inventors of health technologies must take into account a complicated matrix of stakeholders – patients, providers, payers, regulators, and many others – all of whom have influence on the nature of the need, the requirements of a potential solution, and the extent to which that solution is ultimately adopted into practice (or not).

This web of stakeholders complicates the process of need characterization, requiring significantly more time than consumer product design. However, it also means that an innovator who is able to fully characterize a need is at a great advantage in inventing a technology that will actually succeed through to patient care. The effectiveness of the biodesign innovation process, which students and fellows learn in an intensive, hands-on manner, is demonstrated by the many technologies and companies created by first-time trainees in Stanford Biodesign (and other similar programs).^2,4

While the biodesign innovation process is specialty- and technology-agnostic, cardiovascular medicine is especially well-suited for its application, as a rich history of innovation in medical devices has played a major role in the treatment of patients with a wide range of cardiovascular diseases.

B. Phase 1: Need Identification

As noted, the identification of unmet medical needs is the cornerstone of the biodesign innovation process. Innovators typically initiate this work via direct observation in a wide variety of clinical environments, within and outside the hospital. In the Stanford training process, observations are performed by multidisciplinary teams of physicians and engineers, which has the advantage of bringing different perspectives to bear. However, the process can be performed successfully by anyone willing to adopt a discovery mindset – that is, by looking at a clinical situation from a fresh perspective, and putting aside any preconceived ideas about problems or solutions.

The goal is to identify opportunities linked to suboptimal patient outcomes, procedural complications, frustrations voiced by providers, inefficiencies in the delivery of care, outliers in terms of the cost of care, or any other issues encountered by stakeholders in the system. This part of the process is an accumulating or collecting stage (or, in the language of design thinking, “flaring”), and has the open-minded feel of brainstorming in that the innovators are looking for as many interesting needs as possible. In the Stanford Biodesign Innovation Fellowship program, the teams are asked identify at least 200 needs before proceeding to the next stage.

Although needs identification centers on direct observation, it also relies on secondary research to build out an understanding of the need landscape. If the need is related to a particular disease, it is vital to understand as much as possible about the pathophysiology of that condition. System or delivery needs require understanding and learning about the background of the current approach, and how it might vary in different settings. In the current environment where health care affordability is a central focus, it is also essential to develop a clear understanding of the economics associated with a given need. Important questions include: what does this problem currently cost the healthcare system? Who is paying for diagnosis and treatment? What are the general types and levels of reimbursement available in the space? And where might there be potential cost savings?

Needs finding culminates with the creation of a series of need statements based on the observations that have been made. The need statement is a single sentence that captures the essence of the problem, the population it affects, and the improved outcome most desired by affected stakeholders. Careful thought goes into creating need statements, as they serve as a fundamental template for any project that emerges. Well-crafted need statements take the following structure (see Figure 2):

For example:

“A way to reduce the incidence of urinary tract infections in ICU patients that reduces hospital length of stay.”

Here, the true nature of the problem is clearly defined, the target population is narrowed to a specific group, and the outcome is a defined in a measureable way. One challenge in creating needs statements is that it can be tempting to (inadvertently) embed a solution in the need statement. For example, the statement, “A way to improve urinary catheters in ICU patients to reduce the length of stay” embeds a type of solution – that is, better catheters – into the need statement. This has the consequence of limiting the potential range of solutions the innovators consider. In this case, a novel approach may exist to treat urinary tract infections that does not involve modification of a catheter.

Once a relatively large list of potential needs with corresponding need statements has been created, the next stage is need screening (see Figure 3). Preliminary research into each of the needs provides enough information to filter out less-promising options (relative to the other identified needs). While needs filtering requires considerable effort, the process can proceed relatively swiftly by comparing needs to one another, with the most promising needs surviving to the next round of screening.

Typical filters for eliminating needs include clinical impact, the level of understanding of the pathophysiology underlying the need, assessment of the efficacy of existing therapies (and the nature of the competitive landscape), and a preliminary assessment of market potential in this need area. Additionally, each relevant stakeholder must be carefully considered, as these groups have strong influence regarding whether or not new solutions have a reasonable likelihood of being adopted and eventually reaching patients. The needs that survive the screening process are then ranked and prioritized based on further research and discussions with stakeholders.

In the Stanford Biodesign Innovation fellowship, innovators take their top 16 needs forward and create need specifications for them. A need specification acts as an “executive summary” of all the relevant research performed to date. Importantly, it also contains the “need criteria” that the innovators define based on everything they have learned about the need.

Need criteria are the specific characteristics that any solution to the need should address to have a reasonable chance of being adopted by the affected stakeholders. In effect, they are a “wish list” for the ideal solution. These criteria are succinct, bullet-point descriptors of the characteristics, such as impact on clinical care and/or cost-effectiveness, required by decision-makers and key influencers from any new solution. Accordingly, the output of the next steps in the process will be evaluated against these criteria. For instance, these criteria provide a roadmap for brainstorming, while also helping to ensure that potential solutions are focused on truly addressing the clinical need that was identified. It is useful to separate need criteria into two categories: 1) “must-have” characteristics, those required for the right solution, and b) “nice-to-have” criteria, those that would provide benefit, but are not absolutely essential. A useful set of criteria includes 3-5 “must-haves” and a similar number of “nice-to-have” characteristics.

Defining the need criteria forces the innovators to develop an even deeper understanding of the needs, which in turn enables them to further screen and prioritize their list of needs. In the Stanford Biodesign Innovation Fellowship, the creation of the need specification and need criteria enables the trainees to reduce their list to the top 3-4 needs that they will take forward into the next phase of invention.

Although it is clearest to describe the Biodesign innovation process as a set of sequential steps, it is important to re-emphasize that the process is in fact quite iterative. Many steps occur in loops and in parallel (note the cyclical arrows in Figure 1). For example, needs finding may well continue while screening is occurring on the needs already identified. With each successive iteration the understanding of each need deepens, as does the innovators’ confidence in advancing forward.

C. Phase 2: Invention

Once a few top needs have survived the rigorous filtering process, it is time to brainstorm solutions to these needs (see Figure 4). The diligence spent in investigating needs during the Identify phase guides innovators through the Invent phase, as they use the need criteria to initiate and fuel focused, creative ideation. The standard brainstorming approach involves a small team of participants who generate as many ideas as possible in a free-form, rapid-fire session that lasts approximately one hour.

A productive brainstorming session may result in 50-60 concepts in a variety of categories (e.g., solutions utilizing electrical, chemical, biological, or mechanical means, etc.). Capturing the results of a brainstorming session is important and nontrivial. In addition to photographing or otherwise retaining all of the ideas generated, it is useful to specifically identify a few of the most promising solutions. These may be “winners” in different categories – for example, the idea that seems to be most practical, an idea that represents the biggest reach or stretch, the most economical option, and/or the idea that the team finds most novel or interesting. Typically a team will conduct a least a few brainstorming sessions around a given need, and may even perform up to 5-10 sessions.

Once the innovators have developed a robust set of concepts for each needs, they can proceed to an activity referred to as initial concept selection. This involves evaluating each solution idea against the need criteria to assess its potential to effectively address the must-have and nice-to-have requirements. Usually only a select few of the brainstormed concepts truly show promise in satisfying the most important need criteria. This is the power of the need criteria: they provide what is, in effect, an internal contract for the team to help ensure they stay true to the actual need, and avoid being distracted by a clever idea that does not, in fact, meet the characteristics that stakeholders deem to be most important. For this reason, initial concept selection often leads innovators to conduct more brainstorming, using what has been learned to stimulate additional ideas for addressing the need.

It is worth noting that early prototyping may be helpful during concept generation. Drawings or rough sketches, or foam-core mock-ups can help clarify initial ideas and lead to additional concepts. Again, this part of the process is often iterative or cyclical. For example, team members may learn from prototyping they lack a deep understanding of some aspect of a need, thus requiring further research, which in turn can prompt the revision of one or more need criteria.

For the concepts that survive initial concept selection, a second, much more elaborate set of rigorous filters is brought into play during the concept screening stage. These include (see Figure 5):

• Intellectual property (IP): Is the new idea sufficiently novel to be patentable, and does it have “freedom to operate”?
• Regulatory: How challenging is the pathway for approval by the FDA (and/or regulatory agencies in other countries)?
• Reimbursement: Are there existing methods of reimbursement through Medicare or private insurers? Is there an option for self-pay?
• Business model: What type of business would be required to take this idea forward into patient care? How viable is that approach given what is known about the need area and the stakeholders within it?
• Technical feasibility: How difficult will it be to build this new technology?

Each filter requires extensive background research and consultation with experts in these respective areas to gain a sufficient detailed understanding of the potential impact for the solutions under consideration. However, it is important to note that even more in-depth investigation will be required for the final concept that is chosen to take forward into implementation.

A diligent screening process will reduce the number of concepts to 2-3 of the most promising potential solutions. One useful mechanism for screening is to rate the degree of difficulty (or risk) to the project that each filter represents. For example, if the intellectual property landscape is clear, a concept may be given a “green light.” However, the same concept may have a concerning outlook with regard to reimbursement (a significant amount of time and money may be required), meriting caution (a “yellow light”). Concepts are then compared to each other, and those with the most promising overall risk matrix are taken forward (see Figure 6).

In parallel, more serious prototyping begins. The most effective way to proceed is to identify the most important questions that building a prototype can answer, and then construct the fastest, most simple prototype to provide this answer (see Figure 7). In some cases, this will be a “looks like” prototype (where, for example, the way that the user interacts with the technology is critically important). In others, it may be a “works like” prototype (where some mechanism of action of the technology is important to explore – giving new insight into technical feasibility, for example, or suggesting a way around an important intellectual property barrier).

It is important to understand that the needs screening stage will inevitably reveal significant risks in at least some of the categories for every need. However, these risks can be prioritized, and this evaluation can be used as a way of selecting the concept that is most likely to succeed by minimizing these risks. It may be the case that after prototyping and further investigation, none of the concepts for a particular need are sufficiently risk-free to proceed. This is a clear signal to pivot to an alternative need (from the Identify phase) and/or return to brainstorming once again.

While the task of screening a long list of needs and their related solutions can be daunting, it is important to focus on obtaining just enough information at each point in the process to make a decision without investing too much time on any given need. Needs with major flaws identified should be dropped, and those that survive will have a progressively greater amount of information attached so that the path forward becomes much more apparent. One or two final concepts will become clear front-runners, and these move on to the Implement phase.

D. Phase 3: Implementation

In the Implement phase, in-depth analysis and planning for each concept is initiated (see Figure 1, under “Implement”) with the goal of creating a multi-year plan to bring the concept to patients. Note that many of the core topics in the implementation phase are dealt with in some depth in other chapters in this text. In order to move forward, the innovation must be safe and effective for patients, while also being attractive to providers and payers – ideally producing better results than existing therapies at a cost the system can support.

A few key concepts are worth highlighting with respect to the Implement phase. In order to successfully deliver a technology into patient care, key decision-makers must be convinced that the innovation is worth the cost of adopting the new approach (both financially, and in any other way). The argument presented to these stakeholders is referred to as a “value proposition,” and it is central to the business strategy of the Implement phase. In addition to the cost of the technology (which is usually central in the value proposition), other factors that influence the assessment of value include associated economic variables (e.g., Will physician reimbursement be affected?), barriers to changing from the existing approach (e.g., Is there substantial education or training required?), convenience (e.g., Is the new solution substantially easier to use? Does it easily integrate into existing workflows?) and any elegance or “cool” factor (e.g., Can the hospital advertise that it will be the first to feature this exciting new breakthrough?).

The most fundamental concept in the Implement phase is the prioritization of risk. Every project has a host of unknown factors across the entire development spectrum. Often the most critical factors belong in one or more the screening categories mentioned above, viz., IP, regulatory, reimbursement, business model, and technical feasibility. The job of the innovator in planning for implementation is to understand which risks are most important, and then determine how to mitigate those risks in the shortest time and with the least amount of money spent.

It may be the case that the risk is focused on technical feasibility; in this case, a plan for efficient prototyping and early testing can be devised. However, the greatest risk may actually be in a different category (IP, for example), in which case initial time and resources should be concentrated in this area. The most challenging risk situation occurs when the biological or clinical impact of the innovation is difficult to predict (for example, when there is no adequate animal model), and only the results of a definitive clinical trial will answer the key question.

Once the risk matrix is more deeply understood, efforts can shift to developing a detailed plan for device development and testing. This will include both pre-clinical and clinical testing, and identifying potential models for use prior to first-in-human testing. A quality management protocol must be devised after a working model of the concept is developed. Simultaneously, attention must be centered on developing a viable business plan (sometimes the truly novel component of a new concept lies here). This involves an understanding of sales and distribution, financial modeling, funding strategies, a marketing plan, and a strategy to articulate a value proposition for all stakeholders. Creating a clear message of competitive advantage is central to a successful product or process launch.

At the end of the Implement phase, core strategies for the launch of a new business (or for a new program within an existing business) will be developed, and a roadmap for product development and market initiation will be created. It is important to underscore that there may be alternative pathways to successfully bring a new concept to the market beyond creating a new business. Licensing or selling a solution to an existing company can be the most viable way of introducing a new technology into patient care, especially when there is an established sales and marketing organization in that area of clinical practice. Regardless of the business pathway, the critical steps in the early-stage implementation planning for the technology remain the same: all innovators need to demonstrate the value brought by the innovation, and ensure that the path to product realization is clear.

A final point deserves emphasis. The set of knowledge and skills required to successfully execute the biodesign innovation process is so broad that no single innovator can master all facets. In fact, even a multidisciplinary team will find areas where outside guidance is required. Utilizing professional consultants will frequently save a significant amount of time and money in the process of bringing an idea to market. Hiring experts, bringing in strategic talent onto the team, and utilizing advisors are all effective means of minimizing risk and ensuring a good outcome. A successful innovator is able to realize when help is needed – and is able to obtain it quickly and effectively.

E. Conclusion

The Stanford Biodesign innovation process is a needs-based, structured approach to healthcare innovation that provides a roadmap to facilitate the development of new health technologies. The three-phase, iterative process – identify, invent, and implement – has been tested by students, first-time inventors, and seasoned innovators over many years, and has been shown to provide a practical and effective framework for technology development and commercialization. In the new era of cost-effectiveness in health care, emphasis on the opportunity to create value in the up-front needs assessment stage of the process is critically important.

References

1. Yock P.G., Brinton T.J., Zenios S.A. Teaching biomedical innovation as a discipline. Sci Transl Med. 2011;3(92):cm18.
2. Yock P.G., Zenios S., Makower J., et al. Biodesign: The process of innovating medical technologies. 2nd edition. Cambridge University Press, 2015.
3. Kelley, T., Littman, J. The Art of Innovation: Lessons in Creativity from IDEO, America’s Leading Design Firm. Crown Business, 2001.
4. Brinton T.J., Kurihara C.Q., Camarillo D.B., et al. Outcomes from a postgraduate biomedical technology innovation training program: the first 12 years of Stanford Biodesign. Ann Biomed Eng. 2013;41(9):1803-10.