Field notes.
Notes from the set and the edit, written by Alan Schaaf.
Start here.
- Trial recruitment
- MLR on the first pass
- Consent
- Mechanism of action
- The edit
- Rare disease
- Fair balance
- Versioning
Written between shoots.
- Patient video and trial enrollmentEnrolment stalls when a trial stays abstract. What turns a listing into a phone call is recognition.Read the note.
- How patient films clear medical reviewA clean first pass is designed. It starts at the script, long before a camera is switched on.Read the note.
- Where AI helps a mechanism filmAI made the mechanism films faster to iterate. Here is where we let it in, and where we hold the line.Read the note.
Every note comes from the work itself.
More than
Eighty percent.
Clinical trials routinely miss the enrolment deadline they set.
Perspectives in Clinical Research, 2020
Recruitment and medical review fail in the same place: when the person it is for cannot see themselves in it. That is what these notes are about.
What the notes keep circling back to.
Trials stall because the people who could enrol never see themselves in the study. When someone with the same diagnosis describes the exact symptom that sent them looking, people stop and watch.
Bringing medical and regulatory reviewers in before we film is the one change that fixed our first-pass rate. It also means we only ask a patient to say things that will clear review, so their most vulnerable moment stays in the film.
Vague, one-time consent surfaces at the worst possible moment, usually days before a launch. Precise consent protects the person first and the programme second, and it is the reason a story stays up once it is out.
AI is good at accelerating a pathway we already understand and can specify. It turns unreliable the moment judgment is needed: how far we can simplify before we are wrong, and where a visual quietly implies a causation the data does not support.
Send us the question these notes left open.
If the answer runs long, it becomes the next note.