IDEO & Lilly Auto Injector

Lilly’s product pipeline was expanding to protein-based therapeutics for a dozen disease states, all of which had to be injected to bypass the digestive system. Each drug had different requirements, such as user context, dose, viscosity, stability, and storage.

My small team researched patient experiences across five serious diseases — rheumatoid arthritis, HIV, lupus, cancer, and Alzheimer’s disease — and proposed the development of a design platform: two injector mechanisms that would support the pipeline, with industrial design changing per drug category to address specific user needs for interaction. We interviewed 22 patients in their homes across the U.S., and did extensive research with families of children with ADHD. As a result, IDEO & Lilly created a long-term strategy with patient-friendly devices that provided a massive differentiation in the market while supporting a consistent brand expression.

After years of iterations, user testing, and FDA approvals, Trulicity (for Type 2 Diabetes) was the first therapeutic in the new auto-injector, released in 2015. Taltz (for psoriasis) was marketed in 2016 and won gold in the Medical Design Excellence Awards in 2018. Mounjaro (for Type 2 Diabetes) was added to Lilly’s injectable drug portfolio in 2022. Zepbound (for weight management) was released in 2023. The platform now delivers some of Lilly’s most widely used medicines. It started with design research in patients’ homes, and I am proud that this work is having impact at huge scale.

The research behind the platform

Lilly’s pipeline of monoclonal antibodies, injected proteins for a range of diseases, could not be packaged in any of its existing devices. In Project Reach, IDEO and Lilly designed a delivery platform for these new biologic therapies. We started with rheumatoid arthritis (RA), the first likely therapy, because stiff, painful and deformed hands made it the most demanding test of an injector. Across four phases of human factors research, we tested more than 20 prototype systems with more than 50 patients, caregivers and medical professionals, from early concept shapes to fully functional injection-molded devices.

1. SubQExplored self-injection needs across several disease states
2. Reach 1Introduced RA; tested simple prototypes
3. Reach 2ARefined prototypes with RA patients
4. Reach 3CAdvanced prototypes with RA, type 2 diabetes and Alzheimer’s

Phase 3C: testing a platform across diseases

In 2006, Lilly asked whether one device design could work beyond RA. Two design engineers and I ran in-home evaluations in the San Francisco Bay Area and Chicago, comparing five prototypes with patients and caregivers in three disease groups.

Each visit followed the same structure. Patients and caregivers told us the history of their illness and how they felt about injections. We then introduced the prototypes as if the decision to self-inject were imminent, and watched each person act out a real injection on a practice pad before comparing the devices.

What we learned

Patients	What mattered	In their words
Rheumatoid arthritis	A missed dose means pain within days, and many already linked syringes with pain. They wanted visible and audible confirmation that the full dose was delivered.	“If I lose the shot, there’s no way to replace it.”
Type 2 diabetes	Frequent insulin shots felt daunting to people who had never injected. A once-weekly injection was far more acceptable and could improve adherence.	“I do have site reactions sometimes. This gives me more places to inject.”
Alzheimer’s caregivers	Patients may pull away or resist. Caregivers wanted a quick, simple injection and assurance that the dose was taken.	“If he feels it, he will pull away. That’s why the syringe won’t work.”

Watching people use the prototypes revealed which features made the difference:

• A wide, stable base let patients hold the injector with both hands, and most skipped pinching up their skin without even noticing.
• A pull-back needle lock was so intuitive that most patients used it without instruction, and parents valued a needle that could never be exposed to children.
• Transparency, a larger body and grippy surfaces helped hands affected by arthritis. The competitor’s autoinjector needed so much force against the body that RA patients struggled to activate it.
• Seeing the needle was optional: some never wanted to see it, but most appreciated having the choice.

Both Reach devices were clearly preferred over a bare syringe, with the automatic rated highest overall. We recommended a single design language across Lilly’s whole class of biologic therapies, with each version tuned to the needs of its patient group: the value of the platform lay in its mechanisms, intellectual property and design language, not in any one device. That thinking became the two-mechanism injector platform described above.

Bringing patients to Lilly’s leadership

The core of my work with Lilly was to represent patient experiences to executive management, so that patients’ realities shaped the pharmaceutical and device design of pipeline products across disease states. I presented our research directly to Lilly’s CEO, John C. Lechleiter, and the VP of device design, translating what we learned in patients’ homes into decisions about how each medicine would be delivered and experienced.

Presentations could only go so far. To let leadership see what we saw, I took John Lechleiter into the field, together with IDEO’s CEO, Tim Brown, to interview patients in their own homes. Speaking with people who manage serious illness and medications every day, and observing them handle an injection, made the case for patient-centered device design more powerfully than any report could.

The visits also underscored the evidence that patients with chronic illness are often depressed, a barrier to adhering to medication protocols. Lilly’s CEO directed his team to link depression treatments to chronic illness, which had been in product silos to that point.

Building the research infrastructure

As Design Research Lead, I also created a standard review & approval process for human factors research proposals, and trained IDEO researchers on HIPAA compliance and patient-drug adverse event reporting. Establishing this process opened Lilly’s access to patient-centered research, enabling IDEO to do high-risk, patient-facing research for Lilly going forward.