This simulation wasn't testing whether you practiced medicine correctly. Every patient you saw was part of a set designed so that some had clinically similar presentations but different demographics: different ethnicity, gender, age, or insurance.
The goal was to let you see, in your own decisions, whether those differences quietly changed how you treated people.
Most bias training tells you what to watch for. This shows you what you actually did. Of the many implicit-bias interventions studied, only a handful produced measurable behavior change, and the ones that worked tended to involve feedback on real decisions, not lectures.
Differential is built around that idea: give clinicians a safe place to make calls, then hand back the data. No score is final; the point is to notice patterns before they become habits.
Physicians underestimate African American patients' pain 47% of the time, versus 33.5% for non-Black patients. [1]
Women are 50% more likely than men to be misdiagnosed after a heart attack. [2]
In one study, 55% of Hispanic patients with fractures received no pain medication at all. [3]
Drwecki BB et al. (2011). Reducing racial disparities in pain treatment. Pain, 152(5).
Pope JH et al. (2000). Missed diagnoses of acute cardiac ischemia in the ED. NEJM, 342(16).
Todd KH, Samaroo N, Hoffman JR (1993). Ethnicity as a risk factor for inadequate emergency department analgesia. JAMA, 269(12).
The point isn't to guilt you. Implicit bias is automatic and nearly universal, and the value is in noticing it. Awareness of your own patterns, seen in your own decisions, is the first step most training never reaches.
Patient photos are real, model-released portraits from Pexels.