In a chronic pain clinic the patient’s report is the evidence. Justin Coffeen spent years learning that what someone says and what is happening are not always the same thing, and that the gap is usually the clinician’s to close.
What this episode covers
- Why patients minimise, amplify, or report the number they think is wanted
- Open questions and silence as techniques rather than courtesies
- Functional Capacity Evaluations and the discipline of separating observation from inference
- How the quality of a conversation limits the quality of a participant-reported outcome
The question shapes the answer
A question can be phrased so that it supplies its own answer. Asking a participant whether they have had any problems since the last visit invites a no. Asking what has been different invites a description.
Justin Coffeen treats this as a technique with a learnable form rather than a matter of bedside manner. Open phrasing, no leading, and a willingness to leave a pause unfilled while the participant decides how much to say.
The pause is the part people skip. A coordinator working through a schedule of assessments has every incentive to move, and moving is exactly what closes the window in which someone volunteers the thing they were not sure counted.
Observation and inference are different records
Conducting Functional Capacity Evaluations required Coffeen to keep two things apart that blur together easily. What a person did under standardised conditions is an observation. What that performance means for their function is an inference drawn from it.
Documentation that mixes them produces a record more subjective than it looks and harder to defend under review. Documentation that keeps them separate can be read by someone who was not in the room.
Participant-reported outcomes carry the same requirement. The record should show what the participant said, in terms close to their own, before it shows what anyone concluded from it.
Where the data actually comes from
Most of a trial’s data is generated in a room with one participant and one coordinator. Protocol design, statistical planning, and monitoring all sit downstream of that conversation and none of them can repair it.
Justin Coffeen’s argument is that treating participant conversation as a soft skill misplaces it. In a study where the primary endpoint is something a person reports about their own body, it is the measurement instrument.
Transcript
So I spent several years working in a medical chronic pain clinic, alongside my chiropractic practice, and I want to talk about something that felt very much like a clinical skill at the time but turned out to be directly applicable to clinical research. Which is the ability to listen to a patient report and actually understand what they’re telling you.
In chronic pain, the patient’s report is, in many cases, the primary evidence you have. There’s no lab value that tells you whether someone’s back pain is better or worse. They tell you. And what I learned, pretty quickly actually, is that what they say and what’s actually happening are not always the same thing. Not because patients are being dishonest, but because the clinical encounter has its own dynamics. Some patients minimize because they’ve been dismissed before. Some amplify because they feel like that’s the only way to be taken seriously. Some give you the number they think you want to hear.
So I got very interested in how to ask questions that got around that. Open questions. Silence. Not rushing to the next item on the assessment. And what I found is that the quality of the information I got back changed significantly.
And then I started doing Functional Capacity Evaluations, which are structured assessments of what someone can actually do physically, and those are interesting because the whole point is to separate observation from interpretation. You’re not inferring. You’re documenting what happened under standardized conditions. That discipline, being really precise about what you observed versus what you concluded, that transferred directly to research.
In clinical trials, participant-reported outcomes are the primary endpoint in a lot of studies. The data is only as good as the collection. And the collection is only as good as the conversation that produced it. If the participant is minimizing, if they don’t understand what they’re being asked, if the interaction is rushed, the data reflects that.
Honestly, the chronic pain clinic prepared me for research data collection in ways I didn’t expect at all. The disease areas are completely different. But the skill of listening to what someone is actually reporting, rather than what’s convenient to hear, that’s the same.