Mapping Communication Coaching to ACGME Milestones

Clinical Competency Committees carry significant responsibility. Twice a year, they convene to review each resident's progress and assign milestone ratings across six core competency domains. For patient care and medical knowledge, the evidence base is often reasonably dense: procedure logs, in-training exam scores, observed clinical encounters, attending evaluations tied to specific rotations. For Interpersonal and Communication Skills (ICS) and Professionalism, the picture is frequently thinner. Ratings get assigned, but the behavioral evidence behind them is often sparse, secondhand, and difficult to defend if a resident challenges a rating or requires a structured remediation plan.

This gap is not a failure of CCC members. It is a structural problem—one that stems from how communication competency is actually assessed in most residency programs. Addressing it requires connecting coaching on real patient encounters to the specific behavioral language the milestones already define. Done well, that connection turns routine coaching into longitudinal, defensible formative evidence.

A quick orientation to ICS and Professionalism milestones

The ACGME's six core competencies—patient care, medical knowledge, practice-based learning and improvement, interpersonal and communication skills, professionalism, and systems-based practice—have been part of the accreditation framework since 2003. Milestones, introduced in 2013 as part of the Next Accreditation System, provide the developmental language for tracking trainee progression within each competency domain.¹

In Milestones 2.0, four of the six competencies—ICS, Professionalism, Practice-Based Learning and Improvement, and Systems-Based Practice—were harmonized: multidisciplinary workgroups developed common subcompetencies applicable across all specialties, replacing the earlier specialty-by-specialty variation.²

The harmonized ICS subcompetencies are:³

• ICS-1: Patient- and Family-Centered Communication — building the therapeutic relationship, identifying and addressing communication barriers, eliciting patient values and goals
• ICS-2: Interprofessional and Team Communication — consulting effectively, integrating recommendations from consultants, adjusting communication in team settings, and using feedback
• ICS-3: Communication Within Healthcare Systems — documenting clearly in the health record, selecting appropriate modes for sensitive communications, and using system channels to improve care

The harmonized Professionalism subcompetencies are:⁴

• PROF-1: Professional Behavior and Ethical Principles — demonstrating respect, integrity, and ethical conduct across clinical contexts
• PROF-2: Accountability/Conscientiousness — fulfilling responsibilities to patients, teams, and the profession
• PROF-3: Self-Awareness and Help-Seeking — recognizing limits, responding to fatigue and stress, seeking assistance appropriately

Each subcompetency follows a five-level developmental progression. Level 1 reflects what is expected of a novice entering the specialty. Level 4 describes the skills and behaviors required for unsupervised practice—the effective graduation target for most trainees. Level 5 is aspirational and is not required for completion.⁵ CCCs are responsible for making a defensible judgment about where each resident falls on this continuum at each reporting period.

The assessment evidence gap

The problem that most CCCs recognize but rarely name directly is that the evidence base for ICS and Professionalism milestone ratings is often insufficient for the judgment being asked.

Qualitative research on CCC decision-making has documented the pattern clearly. Rather than drawing on behavioral records tied to specific clinical encounters, committee members typically work from aggregated numerical ratings in electronic systems—mean scores from end-of-rotation evaluations. When those scores are available and consistent, members compare a resident to peers and to expectations for their year of training. When data are absent, the default is to assign the expected average rating rather than mark the resident as "not yet assessable."⁶

As one study of CCC process framing it: committee members act as "passive recipients of information from frontline assessments, viewing these averaged, aggregated data as the current representation of the resident's sub-competency achievement, without further interpretation."⁶ The assessment system is designed to yield ratings; it is not designed to generate the behavioral evidence that would make those ratings robust.

Several factors compound the thinness of the evidence. Faculty evaluation of communication is vulnerable to the halo effect: research has documented that physician ratings of overall resident competency can account for nearly 97% of the variance across all competency domains, suggesting that evaluators form global impressions that wash out discrimination between subcompetencies.⁷ Distinguishing professionalism from interpersonal and communication skills in the mind of an evaluator is particularly difficult, given the conceptual overlap between the two domains.⁷

Direct observation—the method most likely to produce behavioral evidence—is sparse. Studies of faculty-resident shared clinical time have found that direct observation accounts for less than 5% of that time even when faculty and trainees are in the same setting.⁸ Without observed encounters, feedback becomes impressionistic and general, which is exactly what makes it difficult to map to specific milestone language.

A CCC facing a borderline resident in ICS-1 needs more than a set of 3.2-average rotation evaluations. It needs evidence of what that resident actually did in patient conversations—what they said, where the communication succeeded, and where it did not.

Turning real encounters into milestone evidence

The solution is not to schedule more formal observation events. It is to structure the coaching that programs already provide so that it generates a behavioral record that maps to specific subcompetency language.

ICS-1, ICS-2, and ICS-3 are defined in behavioral terms precisely because the ACGME wanted assessors to move away from trait-level impressions toward observable conduct. The three milestone threads within ICS-1—relationship building, addressing communication barriers, and eliciting patient values—are not abstract constructs. They describe things a resident either did or did not do in a patient encounter. A resident who consistently struggles to elicit patient preferences before discussing treatment options is displaying a specific ICS-1 deficit. A resident who fails to structure consultation requests in ways that integrate the consultant's perspective is demonstrating a gap in ICS-2.

When coaching is anchored to real encounters and structured around rubrics that use the milestone vocabulary, two things happen. First, the resident receives feedback specific enough to act on—tied to a concrete moment in a conversation, not a trait attributed to them. Second, the coaching session produces a record that is genuinely useful to a CCC: a pattern of behavioral evidence across multiple encounters, longitudinally collected, rather than a single observation or a rotation average.

Building a longitudinal picture requires consistency of structure. A resident who reflects on ten patient encounters over a six-month period using a framework keyed to ICS-1 and ICS-2 is generating something a CCC can actually use: a pattern of evidence that shows where development occurred and where it did not. A single well-observed mini-CEX is valuable, but it captures one moment. Structured coaching across many encounters captures a trajectory.

The ACGME's own supplemental guides for ICS milestones provide behavioral examples for each level—specific descriptions of what a resident at Level 2 versus Level 3 in patient-centered communication actually does.⁹ Those examples are there precisely to make milestone ratings more defensible. Programs that connect coaching to those examples are using the milestones as they were designed to be used.

Keep it formative, not punitive

There is a risk that any systematic connection between coaching and milestone data becomes, in residents' experience, a surveillance mechanism dressed up as development support. That risk is worth taking seriously, because if residents approach coaching defensively, the data it produces will reflect what they are willing to say rather than what they actually experience in clinical encounters.

Formative coaching requires genuine separation from summative evaluation. Residents need to know that what they disclose in a coaching session about a difficult patient conversation—the moment they felt lost, the exchange that did not land—will not appear in their next CCC review as evidence of deficiency. Formative data should be learner-facing first, with residents having access to their own longitudinal record before anyone else does. Where coaching generates patterns that are meaningful for program evaluation, that analysis should be aggregate and de-identified.

The coaching stance matters as well. Communication is one of the few domains in residency training where a learner might otherwise never receive feedback on something they are genuinely struggling with—because the struggle is invisible unless someone is specifically looking. A skilled coaching conversation opens that space without making vulnerability feel consequential. Residents who trust the process develop faster; the formative record produced is also more accurate.

This is not naive about the reality of residency. Program directors and CCCs need evidence. Residents need to advance. But the evidence most useful for defensible milestone ratings is evidence produced by a system that residents participate in honestly—and that requires protecting the formative space.

How Inflect supports this

Inflect provides structured, rubric-aligned coaching tied to real clinical encounters, with a framework keyed to the ACGME ICS and Professionalism milestone language. Individual coaching is learner-owned and explicitly formative. Program-level analytics draw on de-identified aggregate data, so GME leaders can see where communication development is and is not occurring across a cohort without the system functioning as individual resident surveillance.

If your program is working on the evidence side of ICS and Professionalism milestone assessment, the Inflect GME solutions page describes how structured encounter coaching translates to milestone-aligned evidence. You can also schedule a demo to see how the resident-facing coaching tools and program-level analytics work together.