Home/Critical Appraisal
Critical Appraisal

Calibration & Discrimination

EM FINAL EXAMS Critical Appraisal · Prediction Calibration & Discrimination The two distinct facets of a prediction model’s accuracy — ranking patients, and getting the actual risks right. Definition Discrimination = can the model separate those who will from those who won’t have the event? (ROC / AUC, the c-statistic). Calibration = are the predicted […]

EM FINAL EXAMS Critical Appraisal · Prediction

Calibration & Discrimination

The two distinct facets of a prediction model’s accuracy — ranking patients, and getting the actual risks right.

Definition

Discrimination = can the model separate those who will from those who won’t have the event? (ROC / AUC, the c-statistic). Calibration = are the predicted risks themselves correct? (calibration plot; Brier score). A model can discriminate well yet be badly calibrated — systematically over- or under-predicting risk.

The picture
perfect calibration 0 0.5 1 0 0.5 1 over-predicts ↓ below line ranking can still be good (AUC fine) Predicted probability Observed frequency on the line = well-calibrated off it = systematic over / under-prediction

Good ranking, wrong numbers: this model over-predicts high risks even if its AUC is fine

What it shows

Whether the model’s risks are right: predicted probability (x) against the frequency actually observed (y), point by point. The dashed diagonal is perfect agreement. Here the curve sags below the line at the high end — when the model says “high risk” the true event rate is lower, so it over-predicts. A curve above the line would mean under-prediction. Crucially, this can happen while the model still ranks patients well (good AUC).

How to read it

Discrimination and calibration are read from different things. The AUC (ROC) asks only “are higher-risk patients ranked above lower-risk ones?” — it is unchanged if you shift every predicted risk by the same factor. The calibration plot asks “is a predicted 20% really a 20%?” Always ask for both; a model can ace one and fail the other.

Why it matters

Clinical thresholds run on absolute risk: “treat if 10-year risk >10%” only works if a predicted 10% is true. A miscalibrated model with a great AUC will still tip patients onto the wrong side of a threshold — over-treating, over-admitting, or falsely reassuring — even though it ranks them correctly.

Key
  • Discrimination = ranking · ROC / AUC (c-statistic)
  • Calibration = risks correct · calibration plot, Brier
  • AUC is blind to miscalibration — report both
Pitfall
Pitfall Judging a model by AUC alone and ignoring calibration. Miscalibrated risks mislead decisions even when discrimination is good — and calibration commonly drifts across settings and over time, so a model needs recalibrating in each new population.
emfinalexams.com · FRCEM / MRCEM revision
EM trial in the wild

Risk scores that travel by ranking but need re-calibrating by numbers. Cardiovascular risk equations are the classic case: a model can keep good discrimination (it still ranks high-risk above low-risk patients) yet over- or under-estimate absolute risk when moved to a population with a different baseline event rate — so guidelines recommend re-calibrating or re-fitting the intercept locally before the predicted percentages drive treatment thresholds. A good AUC does not guarantee the numbers are right: check the calibration plot — especially around the risk threshold your decision actually uses.

Examiner traps
  • Appraising a model on AUC only and never looking at calibration.
  • Ignoring calibration drift across settings or over time.
  • Forgetting that over/under-prediction can coexist with excellent discrimination.
Quick check

A model has a great AUC but over-predicts risk two-fold — fine to use?
Answer: No — poor calibration misleads thresholds and decisions despite good discrimination. A predicted 20% that’s really 10% will push patients across treat/no-treat cut-offs the wrong way; it needs recalibrating first.

Related cards

Ready to build your plan? EMF Premium gives you all 40,000+ questions, 20 mocks and 1,215 OSCE stations from £29/month — or a one-off 3- or 6-month pass.

Share
0
    0
    Your Cart
    Your cart is emptyReturn to Shop