Rapport (Clinical)
Rapport is the working trust and understanding between a clinician and a patient that the rest of the encounter depends on. The Kalamazoo Consensus Statement names building a relationship, meaning establishing trust, rapport, and an ongoing partnership, as the first of seven essential elements of communication in medical encounters, and treats it as something maintained across the whole visit rather than accomplished in the opening minute. Rapport is visible in behavior a third party can observe, including whether the clinician lets the patient finish, names what the patient appears to be feeling, and returns to the concern the patient actually raised.
Source: Makoul G, Academic Medicine, 2001 (Kalamazoo Consensus Statement)
Treating rapport as a personality trait is what makes it untrainable, and the research on how encounters actually open argues against that reading. Singh Ospina and colleagues coded 112 recorded clinical encounters and found the clinician elicited the patient's agenda in 40 of them, 36%. In those 40, the clinician interrupted after a median of 11 seconds, and 67% of the time. Patients who were allowed to finish took a median of six seconds. Whatever rapport is, a good part of it is a set of decisions in the first half minute that a clinician can be taught to make differently.
The evidence that the relationship changes outcomes is real and modest. Kelley, Kraft-Todd, Schapira, Kossowsky, and Riess pooled 13 randomized trials that manipulated the patient-clinician relationship and reported a combined effect of d = 0.11 on healthcare outcomes (p = 0.02), and their own comparison for scale is aspirin's effect on myocardial infarction at d = 0.06. Separately, physician communication training raises the odds of patient adherence 1.62 times (Zolnierek and DiMatteo, Medical Care 2009). These are not transformation-sized numbers, and they are large enough to justify curriculum time.
Rapport is trainable in hours, not years. Riess and colleagues randomized 99 residents and fellows across six specialties to standard postgraduate education or the same education plus three 60-minute empathy training modules, and the trained group improved more on patient-rated CARE scores (difference 2.2, p = 0.04). What that trial does not settle is durability, and three hours of training is a floor rather than a program.
For assessment, rapport has to be written as behavior or it cannot be scored or coached. "Builds rapport" is not an item a rater can defend and not something a learner can practice. "Lets the patient complete their opening statement without interruption", "names the emotion before adding information", and "returns to the concern the patient raised first" are all three, and all three can be found in a transcript. That is the translation the Kalamazoo essential elements and the Calgary-Cambridge guide were built to make, and it is what turns a complaint about a learner's manner into a plan.
What this looks like in a program
- Rehearsing only the opening 30 seconds of an encounter is a cheap, high-yield drill, because that is where the agenda is either elicited or lost.
- A rubric item that names an observable behavior can be coached. A rubric item that says builds rapport produces a rating nobody can explain to the learner.
- Rapport gets harder, not easier, in the encounters that matter most, so it is worth practicing inside a goals of care or bad news case rather than only in a routine history.
Read next
Last updated August 2026