Dana Lewis is an independent researcher, a patient, and a member of the Executive Committee of the Clinical Trials Transformation Initiative (CTTI). With deep experience navigating clinical research both as a participant and as a researcher, Lewis brings a rare systems‑level perspective to how trials are designed, how data are collected, and how emerging technologies (particularly AI) are changing what is possible. In advance of CTTI’s Patient Summit, Lewis spoke with Morgan Hanger about data capture, consent, researcher assumptions, and why many trials remain anchored in outdated models.
Note: This interview has been slightly edited for brevity.
Hanger: Such a thrill to speak with you today. You’ve emphasized that trials should capture all data that participants are willing to share, particularly as AI capabilities accelerate. From your perspective, why is this so important now?
Dana Lewis: We don’t know what future technologies will enable, but what we do know is that we’re constrained by the data we collected in the past. Those constraints are often artifacts of the effort and cost involved, for both participants and researchers. Most trials are designed around a narrow endpoint and so they collect a very narrow set of data. That limits what we can answer later. Questions about titration, implementation, or real‑world use often matter deeply to patients, but the data simply aren’t there.
More data gives us more opportunities for future analysis. That’s especially critical in small populations, where patients may have very few chances to participate in trials at all. If we don’t collect these data now, we lose that opportunity.
Hanger: At the same time, the field is under pressure to simplify trials — reduce complexity, manage ballooning costs, and limit participant burden. Many stakeholders see that as directly at odds with collecting more data. How do you see that tension?
Lewis: I think the call for more data is really about recognizing that data capture has changed. We have tools for passive data collection that didn’t exist or weren’t widely available five or ten years ago.
Participants already carry phones or wearables that collect movement or other wearable data. There are apps that make meal tracking easier through photos, text, or audio. It may not be the perfect gold‑standard measurement every time, but we can get useful estimates with significantly less effort, and in many cases having less accurate data is still better than no data.
So the question becomes: are we not collecting data because it’s truly burdensome, or because we’re still thinking with an outdated understanding of what participant burden looks like? Or because the burden is for the research team? That’s why patient co‑design is essential, to honestly assess where that burden still exists and where it no longer does.
Hanger: You also introduce a concept that’s less frequently discussed: researcher burden. Can you expand on that?
Lewis: A lot of decisions about what data not to collect come from the researcher perspective, including the perceived burden of collecting, cleaning, managing, and analyzing additional data. But that burden has also changed. We often default to saying, “It’s too hard to do anything with that data,” even when patients are explicitly asking us to use it. That mindset is frequently based on technology limitations from years ago. Today, many tools make data ingestion, cleaning, and analysis significantly easier, faster, and lower cost, plus more accessible to researchers with different backgrounds and expertise areas. We absolutely should think about participant burden, but we also need to be honest about whether researcher burden is being over‑weighted in design decisions for clinical trials, especially when there is clear potential value.
Hanger: You wear so many hats, Dana. You’ve also spoken from personal experience as a trial participant about how results are communicated. What’s missing today?
Lewis: As a participant, when results are published years later, they’re almost always population‑level findings. It’s not clear whether I was a responder, a non‑responder, or how, or if, those findings apply to me at all.
There are many opportunities to return data to participants in meaningful ways. When participants receive their own data, they can better contextualize the population‑level findings and have more informed conversations with their clinicians about how the results may or may not apply to their individual situation. This is especially relevant for people with multiple conditions, where there may never be a study that perfectly matches their profile.
Hanger: We know how clinical research often relies on altruism as a motivating factor for participation. Does returning data shift that paradigm?
Lewis: It can. When participants can see and understand their own data, it adds individual value alongside collective value. That doesn’t eliminate altruism, but it strengthens the overall value proposition of participation.
Hanger: When discussions turn to maximizing data use, issues of privacy, consent, and stewardship quickly emerge. What are you hearing from patient communities?
Lewis: Patient communities are not monolithic. Even within a large category like diabetes, perspectives vary significantly. Someone with type 1 diabetes plus additional autoimmune conditions may worry about identifiability and have higher privacy concerns. Or not: someone else in that exact situation may want their data reused as broadly as possible because no one is ever going to design a study that exactly reflects them. Historically, trials tend to adopt the most conservative approach in order to protect participants and respect preferences. The intent is good, but that approach doesn’t reflect the range of patient preferences.
Hanger: You’ve proposed a layered consent model as a way to address this. What would that look like operationally?
Lewis: The first layer is consenting to participate in the trial, with clear explanation of how privacy and data are managed within that study. The second layer (separate question, one that does not affect trial participation) is whether a subset of the data can be shared or reused for future research. Participants can say yes or no. Both responses are valid.
Right now, most trials don’t explicitly ask that question about data re-use. As a result, even participants who want their data reused never have that opportunity. We could support both perspectives within the same study design if consent is structured intentionally from the start.
Hanger: Do preferences change based on the type of data being collected?
Lewis: Absolutely. Continuous glucose monitor data with timestamps may feel very different to someone than genetic data. Even for the same individual, comfort levels vary depending on the type of data and the context in which it’s collected. That nuance is important and it’s something we are capable of addressing if we design for it.
Hanger: I love the intersection of design and hope. You’ve repeatedly said that these challenges are solvable. What’s the biggest obstacle to change?
Lewis: One of the biggest obstacles to change is copy‑and‑paste study design. Too many protocols follow patterns established five or ten years ago without stopping to ask what’s possible now. We should be asking: What technology exists for passive data capture? What tools exist for cleaning and analyzing these data during the study? What infrastructure supports layered consent and clean sub‑datasets aligned with participant preferences? All of this is possible, but it requires intentionally rethinking design decisions instead of defaulting to precedent.
Hanger: As we wrap up, what’s your message to stakeholders who may feel this perspective doesn’t represent them?
Lewis: That’s it’s an invitation to participate in this discussion at the CTTI Patient Summit on April 28, 2026, and beyond. My understanding of the range of perspectives is necessarily limited by my experiences, for example. If someone is listening and thinking, “That doesn’t reflect my experience,” that’s exactly the voices we need to hear from, so that they are represented, too.
We can only move research design forward by hearing from people with different use cases, constraints, and concerns, and working through where current approaches do and don’t apply.
The CTTI Patient Summit on April 28, 2026, brings together patients, caregivers, and patient advocates to continue these discussions. Register now to attend.
