A 13-week pilot at Morristown Emergency Medicine Residency shows that repeated case-based practice builds meaningful clinical confidence — and that the programs benefiting most from DDx are the ones catching gaps before they compound.
What is the unsolved resident remediation problem?
Every residency program has residents who struggle. That's not a failure of selection; it's the reality of training physicians across a wide spectrum of clinical experience. The challenge isn't identifying that a resident is struggling — it's figuring out exactly why, and targeting the right intervention in a timely manner.
Today, that process is almost always resource-intensive. Individualized improvement plans require faculty to diagnose the root cause of every deficiency in medical knowledge, professionalism, and clinical skills, and to tailor feedback for each resident. ACGME standards require detailed, legally defensible documentation of every concern, intervention, and outcome. And all of it — the check-ins, the progress monitoring, the plan construction — sits on top of normal clinical and administrative duties.
Programs need a tool that can identify gaps earlier, deliver consistent case-based practice, and generate longitudinal evidence of improvement without adding hours to every faculty member's week. That's exactly what DDx was built for.
The Morristown pilot: 13 weeks, 30 residents, real results
At Morristown Medical Center's Emergency Medicine Residency Program, faculty piloted DDx as a supplemental tool for targeted skills development and oral board preparation, running from March through June 2025.
- The study enrolled 30 EM residents (PGY-1 through PGY-3).
- Participants were assigned two DDx cases per week over 13 weeks (26 total), working alongside three traditional in-person oral board sessions over the same period.
- Residents were surveyed before and after on confidence, comfort, engagement, cognitive load, usability, and study habits.
What did the Morristown results show?
The results suggest DDx adds meaningful value alongside traditional clinical education and board prep workflows:
+0.6 improvement in confidence managing undifferentiated cases (5-pt scale; moderate, educationally meaningful effect)
87% of residents said AI-based educational tools would be a useful supplement to their training
80% of residents rated DDx engaging or very engaging
Stress levels were also lower with DDx compared to traditional oral board review — an important finding for a remediation context where performance anxiety often compounds the underlying clinical gap. PGY-1s showed the largest gains, consistent with building foundational clinical reasoning skills where there's more room to grow. Senior residents also showed meaningful improvements from a higher baseline, suggesting DDx has utility across all training levels.
Why DDx works: scalable practice without faculty overhead
Traditional remediation is high-touch by necessity. Faculty have to create cases, observe performance, diagnose reasoning gaps, and design individualized improvement plans — work that is essential but difficult to scale.
DDx doesn't replace that clinical mentorship relationship. It handles the practice and assessment side of remediation so faculty can focus on what only they can do. Residents complete cases asynchronously on their own schedule, without dedicated simulation lab time or faculty presence. Longitudinal performance is tracked across every case, automatically. Rubric-based assessment surfaces not just whether a learner got the answer right, but how they approached the problem. And a documented, continuous record of reasoning over time creates objective evidence of both the gap and the improvement, without periodic check-ins or manual review.
When those patterns are visible, the individualized remediation plan becomes much easier to construct. Faculty are no longer starting from zero trying to diagnose the deficiency. The data has already identified where reasoning diverges: whether a resident's differential is breaking down, whether they're ordering the right studies for the right reasons, whether they're managing undifferentiated presentations effectively.
How early identification changes remediation outcomes
The Morristown pilot focused on residents already in training. But the programs getting the most from DDx are also deploying it at the very start of residency — and the early identification benefit is perhaps the most valuable of all.
At Indiana Ascension St. Vincent, where DDx was deployed during intern orientation, faculty described the platform as a gap analysis tool at the program level, surfacing broad patterns across the cohort, not just struggling individuals.
"Without a tool like this, we'd only discover those deficits later in the year. This lets us capture them early and tailor our teaching from day one."
When you identify a reasoning gap in Week 1 of residency instead of month 6, the intervention is more targeted, less intensive, and more likely to succeed before the stakes get higher and the gap has had time to compound. DDx's longitudinal analytics and centralized dashboards give faculty visibility into individual and cohort performance over time, so remediation doesn't have to wait for a struggling resident to become visible through patient care.
Frequently asked questions
How do residency programs typically identify struggling residents?
Most programs rely on episodic tools — milestone evaluations, end-of-rotation feedback, in-training exams, and supervisor reports — that capture performance at fixed points in time. These tools miss how residents reason under real clinical pressure and often surface gaps only after they've compounded, triggering intensive remediation that is more disruptive and resource-intensive than early intervention would have been.
How much faculty time does resident remediation require on average?
Research shows remediation averages approximately 45 hours of faculty and administrative time per learner once a concern is formally identified, typically unfolding over 6 to 12 months alongside ongoing clinical duties. Identifying gaps earlier — before they reach formal remediation thresholds — significantly reduces this burden by enabling lower-intensity, targeted support.
Can AI simulation tools reduce the faculty workload of resident remediation?
Yes, when used as a supplement to clinical teaching rather than a replacement. Platforms like DDx allow residents to complete targeted case-based practice asynchronously, generating longitudinal performance data that faculty can act on without direct observation for every encounter. This shifts faculty time toward coaching and decision-making rather than monitoring and data collection.
