01
Contextual Fit
Evidence works when it is shaped for the setting where decisions happen.
Make evidence travel.
Identify evidence gaps before your next decision.
Strong evidence can fail to change decisions, not because it is wrong, but because it lacks context, traceability or clarity for the people who need to act on it. We review your draft evidence plan or health content early, identify those gaps while it can still be changed, and give you a written diagnosis with priorities for what to change or investigate.
For Medical Affairs, evidence-generation and health technology teams preparing evidence for regulators, payers, clinicians and patients.
Clinical and pharmaceutical experienceFounded by Dr. Steve Watt, a physician and former Pfizer evidence leader. Meet Steve
Implementation science in practice: Steve co-led the MAPS Americas 2026 workshop on applying it in Medical Affairs.
View the official workshop listingResearch you can inspectOpen, registered studies of large language models in health: how consistent they are, and where they can be used in care and R&D. Explore the work
Publications on patient preferences, evidence communication and AI methods.
See selected publicationsClear limitsWe do not guarantee adoption, regulatory acceptance or stakeholder agreement.
The shared problem
Evidence can be missing, or it can leave the next decision unanswered.
Strong evidence and good content can pass the next review and still leave the people who need to use them unable to act. We call this the evidence-to-impact gap.
Bring us a study protocol, integrated evidence plan or piece of patient or clinician content before it is locked, and leave with a written list of what to change first. We look at what the next review needs, what later decision-makers will ask, and whether the evidence or content can be used in routine care.
02
Traceability
See where evidence came from, how it was built and how far to trust it.
03
Clarity
The people who need to act must be able to understand the evidence.
Start with the draft in front of you
What are you working on?
Evidence design
For R&D and evidence-generation plans
Will your evidence still work after the next milestone?
You are drafting a study or pilot protocol, integrated evidence plan, patient or clinician survey, or evidence generation strategy and want to identify gaps before it is locked.
Case: shaping a hospital study around the decision it must inform.
Explore evidence designEvidence communication
For patient and clinician communication
Will your health content be understood and used?
You are drafting patient or clinician materials, such as plain language summaries or infographics, or running a content workflow that produces them, and want to identify gaps before MLR review.
Case: advising an infographic study to measure patient understanding, not only preference.
Explore evidence communicationThe research behind the practice
A powerful tool. A new trust problem.
Large language models now help produce and explain health evidence. They can be powerful tools, but the same question can get different answers, and people need to know how far to rely on them. We study LLM reliability in registered work anyone can inspect.
How much does expressed empathy vary across language model responses?
A review of how empathy is measured and reported, with attention to variation and consistency.
Read the registrationSame prompt. How consistent is the response?
A registered study of expressed empathy across repeated large language model responses. Registration is a study plan, not a result.
Read the registrationDiscuss your next evidence decision.
A 30-minute call to understand the decision, who makes it and what they need to see. If PEL is not the right fit, we will say so.