Advancing the Connected and Smart Drug Delivery Ecosystem: A Path Forward

Photo by TARIK KIZILKAYA/iStock / Getty Images

Earlier this month, I attended the Drug Delivery West Conference to learn how industry leaders in biotech, pharma, and drug delivery companies address scientific and business solutions for the best route of delivery across different therapeutic areas with an emphasis on emerging technologies.

The intersection of consumer technology and medical devices has been a trend for the last few years, but conference leaders believe the time is finally here for drug delivery devices. One panel on connected drug delivery that I particularly liked focused on next-generation emerging technologies, design needs and challenges, and regulatory hurdles to advancing drug delivery in a smart and connected world.

Moderated by Justin M. Wright, Ph.D., Global Head of Innovation for Novartis, the panel included Kimberly Ringenberger, Ph.D., Associate Engineering Advisor for Eli Lilly, Rajan Patel, CEO and Founder of iO Life Science, and Dirk Schapeler VP, Digital Health for Bayer.

During this session, the experts discussed the trend of adding smart technology to drug delivery devices and how to make this shift within a pharmaceutical company. Below is a summary of the main discussion points and key takeaways.

Question 1: What are some great examples of digital health success in the drug delivery ecosystem?

The panel agreed there hadn’t been a “home run” yet as far as an example of great success in bringing smart technology to the drug delivery space. One panelist cited Peloton as a successful example in the digital health space for their success in creating a community and changing user behavior.

The key to success in drug delivery will be to apply human-centered design to the product development process. For example, it is essential to go into the home and observe patients contextually using your product (versus in a clinical setting).

Kimberly Ringenberger agreed that following a user-centered approach is essential - she added the tip to perform user studies early and often. Doing so will allow you the time to adjust your thinking and strategy based on your findings.

Takeaway - The lack of significant success should not discourage you but rather is an excellent opportunity to bring new technology to a market that is ready to embrace smart drug delivery devices. The key to success: shift your thinking to understand the user and their environment of use.

Question 2: How do your companies make Buy vs. Partner decisions for drug delivery technology?

Dirk Schapeler shared Bayer’s strategy of developing its core technology in-house; which is comprised of data and algorithms. Bayer hired the machine learning experts and data scientists necessary to control its core technology then partners with technology companies to develop the device.

Kimberly Ringenberger said Lilly looks at their internal expertise and identifies where the holes are. If the new product has cutting edge technology that needs to be developed, Lilly will either look to buy the technology or partner with a firm to bring the required expertise into the team.

Rajan Patel said iOS Life Science has decided to keep the core of the product within the company strategically. He identified the core of the product as being the diagnostic technology (sensor and algorithm). The rest of development (i.e., instrument and non-diagnostic sensors), he said, they outsource to outside partners.

Takeaway - Determine your strategy for working with outside partners early. If you can’t buy an appropriate delivery technology, identify your core technology and keep that internal, then look for a partner to develop the rest of your delivery device.

Question 3 - People talk about the importance of failing early and often during product development. Can you give an example of a successful failure during the development of a drug delivery device?

Kim shared that at Lilly, they discovered late in the development of a product that there was no freedom-to-operate for the technology. To keep things moving, they did a rapid scrub for existing technology to use and still managed to beat their competitors to market.

Dirk likes to use a lessons learned process that he picked up from his background in the automotive industry. This process includes a project recap in which the team talks about what went well and what didn’t. They then share their learnings so everyone can learn from the collective project’s mistakes. Dirk also noted that he doesn’t like the term “fail fast” but instead thinks it is essential to be honest when something isn’t working, then be prepared to shut it down quickly and then learn from it and pivot (if you can). This approach gives you more time to find a successful path.

Similarly, Rajan shared that he prefers “build to discover” instead of build to “fail fast.” At iO, Rajan’s team designed a product for a clinical study that patients were expected to wear for 17 hours. Before the study, they had performed some user research and asked patients how long they would wear it. The patients’ response was 3 hours. They discovered the big gap between the 17 and the 3 hours within enough time to adjust their thinking. They found the device needed more empathy and inspiration to encourage patients to use it to reach the 17-hour requirement. If they hadn’t been able to adjust quickly, their clinical study results would have been much different.

Takeaway-Be open to learning throughout the product development process. When it doesn’t go as planned, don’t let that discourage you. Set yourself up for success by using this discovery process to learn, evolve, and launch an even better product.

Question 4: Where do you draw the line between asking patients what they want and relying on your expertise to figure out what they need?

Human-centered design is not about asking patients about specific features they may want, Dirk offered. It is about finding a good solution to solve their problems.

Justin Wright, the Global Head of Innovation for Novartis, noted that when he was at Lilly, they evaluated “awesomeness.” They found that patients would know it when they saw it but couldn’t necessarily verbalize what they needed.

Takeaway - Human-centered design is not about asking patients what they want - the key is to listen to your users, identify their pain points, and then to use your expertise to find solutions that alleviate them.

Question 5 - Developing devices in a pharmaceutical company is not always easy as most groups focus on developing and marketing the drug; they are not technology focused. How do you recommend accomplishing the digital health agenda within a drug company?

It is critical to check in with the marketing and commercial groups before pushing a digital health agenda, Rajan noted. If they aren’t ready, it won’t work. They need to understand how to market and commercialize the technology, which is often geared more towards the patient than health professionals.

Justin said, “begin with the end in mind.” Understand what you are going to do with the data before you start gathering it.

Dirk added that there needs to be a shift in thinking about creating IP in areas other than the molecule. He suggested you bring in people from tech to help produce IP around the delivery device.

Takeaway - In developing smart technology for your device bring in or partner with experts from consumer companies.

Question 6 - Agile product development is often used for technology and software. This process is new to drug companies and may not fit well within their existing processes. How do you balance being agile while also following the appropriate regulations?

Make sure to identify and address the most significant risks first, Kim shared. Focus on reducing the top 5 product risks; determine pinch points and solutions before heading into formal design control.

Once you’re under design control, she added, be prepared to quickly identify when risk appears and branch down another path, if necessary.

Also, it is essential to know your regulatory strategy upfront so you can make sure the right features are there (or not there) for your device’s classification level, Kim continued. For example, a class one device typically can not make diagnostic decisions. So you will need to be careful about how you present data to the patient.

Rajan shared that at iO, he views the paperwork for design control as a risk. So, using a risk-based approach, he requires a rough draft of paperwork during the kickoff, research, and proof-of-concept phase before entering formal design control where paperwork is required.

Takeaway - Use a risk-based approach to reduce risk early and often.


The connected drug delivery device market is expected to grow tremendously with emerging technologies that connect patients, physicians, researchers, health systems, and more while reducing the burden to all. To bring these products to market successfully, it is best to use a human-centered design approach, reduce risk early and often, and look to partner with a firm that has the technical expertise and established processes required to do so.