What the industry is saying. And what it still cannot bring itself to do.

I am not built for this weather.

The UK heatwave has done something to my brain this week. I have been moving at approximately half speed, drinking twice the water, and questioning every life decision that led me to live in a country that cannot handle more than three consecutive days of sunshine.

But the diary does not care about the heat.

This week I have demos going out to prospective clients, investor conversations I have been working toward for a while, and meetings that could shape what Q3 looks like for Unwritten Health. The kind of week that reminds you why you started.

Oh. And England versus Ghana in the World Cup on Tuesday. That is non-negotiable. Some things are bigger than the calendar.

I thought I would do a something a little different for this Unwritten Dispatches, I didn’t get an opportunity to attend ACRP in Orlando, or HLTH 2026 in Amsterdam, but have been following the coverage from a far, and it was really interesting.

Two conferences, but the same argument running through all of them, in a different language, without ever quite being completed.

Let me tell you what I mean.

At ACRP 2026 in Orlando, the opening framing for the patient recruitment and community engagement track was described as a chicken-and-egg problem. Which comes first… thoughtful community engagement that leads to robust recruitment, or successful recruitment that leads to rewarding community engagement?

I have struggled with the chicken and egg question a bit, because I think its the wrong question, also the industry is still asking this question in 2026 tells you more than any of the answers that followed.

There is no chicken-and-egg problem. Community engagement comes first. It has always come first. The question only looks like a riddle if you have been treating engagement as a recruitment tactic rather than a precondition for trust. When you treat it as a tactic, you ask which one generates the other. When you treat it as a precondition, the sequence is obvious. You build the relationship before you need something from it. You show up before you have an agenda. You earn the right to ask.

That said, some of what came out of Orlando was genuinely useful. Jessica Propps from Merck made an argument I want to highlight, because it is one of the most practically important points I have heard from a sponsor in a long time. She said that participant payments and the decisions around them quietly determine who can participate in clinical research. When the support the sponsor provides does not match the actual burden of participation, it becomes a financial risk to the potential participant rather than an opportunity. And that risk falls disproportionately on caregivers, who are often not compensated at all.

That is a precise and honest account of something the industry rarely says so directly. The financial architecture of clinical trial participation is not neutral. It was designed around the assumption that participants have enough flexibility, resource, and slack in their lives to absorb the burden. For the communities carrying the highest disease burden, that assumption is usually wrong. The result is not just lower participation rates. It is a systematic exclusion of the people with the most at stake, built into the economics of how clinical trials are operationalised.

Pfizer's recruitment strategists, presenting from both the UK and the US, described their approach to developing patient profiles that consider the patient as a whole, for instance who they are in their day-to-day life, what barriers they might present, what challenges they might have to participate. That framing is the right framing. The question I kept asking is what happens to those profiles. Where do they go? Do they sit in a deck reviewed quarterly, or do they sit in the hands of the clinical development team who are writing the protocol as the visit schedule is being set?

The difference between those two destinations is the difference between patient engagement and patient experience data. One informs a narrative. The other changes a design.

Over in Amsterdam, HLTH Europe 2026 was having a version of the same conversation, dressed in different language. Two distinct threads ran through the sessions I want to address here.

The first was about real-world evidence and what it would take to harness it properly. Lisa Lehmann from Verily made the point that pharma companies must establish what counts as decision-grade evidence on a broad scale. Decision-grade. That is the language we use at Unwritten Health to describe what we are building. Not consultation data. Not engagement metrics. Evidence that is structured, longitudinal, and usable at the point where design decisions are actually made.

Richard Cassidy from Astellas made the companion argument: physicians do not want more data. They are overburdened. What they want is information that directly helps them make the best decision for a specific patient. The question is not how much community data you can collect. It is whether the data you have collected is structured in a way that answers the specific question being asked at the design table. Data that is not connected to a decision is not evidence. It is noise.

The second thread from Amsterdam is one that will shape the next two to three years of this industry whether we are ready for it or not. The first was that everyone is talking about data, but insights has become the new frontier. That shift in language matters. The industry spent years arguing about whether to collect patient experience data at all. It is now arguing about what to do with it. That is progress, even if the distance between collecting and acting remains enormous.

The second was the Europe versus US versus China dynamic on AI and large language models. The panel was direct about it: the US and China are significantly further along. Europe's response has been to position itself as the regulatory powerhouse, and the EU AI Act, coming into force in August, is the clearest expression of that positioning. What this means practically for clinical research is that the tools being built to process patient data at scale will face a more demanding compliance environment in Europe than anywhere else. That is not a problem for organisations that have built their data infrastructure correctly from the start. For those that have not, it is an expensive reckoning arriving on a fixed timetable.

The third was GLP-1s. The obesity and metabolic disease space is being driven by patients in a way that the industry has not experienced before. It is no longer a conversation happening in the regulatory and pharma towers. Patients are demanding access, self-managing, sharing experiences, and pushing the evidence agenda faster than the clinical development timelines were built to accommodate. The question being asked at HLTH Europe was not how do we catch up, but how do we build a world that keeps every patient safe as they continuously use these products over years and decades. The long-term safety and real-world evidence infrastructure for GLP-1s does not yet exist at the scale it needs to. Building it requires exactly the kind of lived experience data and community insight infrastructure that most organisations are still treating as a nice-to-have.

Back at ACRP, Clinical Trials Day this year carried the theme of Research Rising. The response that stayed with me came from Anita Phung at Lindus. She said: communities whose voices were once absent are beginning to shape the protocols designed around the populations they are meant to serve. These are not incremental refinements. They are a fundamental shift in whose experience gets counted as evidence.

I want to believe that. I do believe it in the specific examples she is pointing to. But I want to be honest about the gap between what is being said at conferences and what is happening in the rooms where protocols get written. The language of patient-centricity is moving faster than the practice of it. Research is rising in the vocabulary of the industry before it has risen in the design of the studies.

That gap is not a reason for cynicism. It is a reason for urgency. The frameworks are available. The evidence for what works is established. The regulatory direction is clear. What is missing, in most organisations, is the infrastructure to connect the intent to the decision. The system to ensure that when a community tells you their visit schedule does not fit their life, that information reaches the person who is still deciding whether to change it.

Two conferences. Correct diagnosis. Incomplete prescription.

The prescription is infrastructure. And the window to build it is not as wide as the conference calendars make it look.

Sources: ACRP 2026 conference report, acrpnet.org, June 2026. Clinical Trials Day 2026 thought leader responses, acrpnet.org, June 2026. HLTH Europe 2026 RWE panel coverage, clinicaltrialsarena.com, June 2026.

I have been building toward this for months. Not just the logistics of a live interview series, but the argument behind it. The clinical research industry makes decisions about communities it has never properly listened to. Now Written: Live is where that changes. One conversation at a time, on the record, with people whose stories the system was not designed to hear.

My first guest is Sahara Fleetwood-Beresford. Her journey to diagnosis took years longer than it should have. We are going to talk about what gets lost in that time, not just clinically, but personally. What it means to know something is wrong, to push for answers, and to be met with a system that keeps telling you otherwise. And what that experience should be teaching every clinical researcher who has ever wondered why certain communities do not engage with trials.

This is the conversation that usually happens off the record. I am putting it on the record.

Thursday. 2pm UK. 9am ET.

Thanks for reading. This newsletter exists because I believe the right framing, in the right hands, changes decisions. If it did that for you this week, even a little, that is enough.

Ashish.