Calibration and appropriate abstention

Confidence vs correctness; rate of deferral to a clinician when uncertain.

Who judgesAutomatable and clinician judgement
EvidenceGrade E2
Product typeCross-cuttingID X5

What calibration and appropriate abstention means

Confidence vs correctness; rate of deferral to a clinician when uncertain.

Why it matters. Safe systems know when they don't know. Calibration measures whether confidence matches correctness, and abstention measures whether it defers.

Example, illustrative

A tool that is 95% sure and right half the time is miscalibrated, and should hand those cases to a clinician.

How to measure it

01

Define the unit

Decide whether you count per sentence, per note, per patient or per call.

02

Sample on purpose

Pick cases that cover specialties, languages and rare events, not only easy ones.

03

Judge with clinicians

Clinicians label a sample. Graders are then checked against those labels.

04

Report the margin

Show an interval, and the grader's own error rate beside the score.

Watch out for

A single average hides rare, serious errors and weaker subgroups.

What a good report shows

A number with a clear unit, a stated sample, a margin and the grader's agreement with clinicians.

Report the unit (per sentence, per note, per patient), the sampling method, the severity scale, and how often the automated grader agreed with clinicians compared with how often clinicians agreed with each other.

Reviewed by the Precision research team.

How strong is the evidence

E2 · Preprint, benchmark page or vendor-run result. Backed by numbers from a preprint, a benchmark page or a vendor-run test. Treat as strong but not settled.

What the sources say

HealthBench "responding under uncertainty" theme (E2); conformal methods in healthcare unverified (E4).

  • Automatable. Can be scored by software, including an AI grader that is spot-checked against clinicians.
  • Clinician judgement. Needs a clinician to judge or label, because the answer depends on clinical meaning.

Found something out of date? Write to hello@precision.example.

Quick answers

What is calibration and appropriate abstention in healthcare AI?

Confidence vs correctness; rate of deferral to a clinician when uncertain. It sits in the every product type group.

Why does calibration and appropriate abstention matter?

Safe systems know when they don't know. Calibration measures whether confidence matches correctness, and abstention measures whether it defers. A tool that is 95% sure and right half the time is miscalibrated, and should hand those cases to a clinician.

How is calibration and appropriate abstention measured?

Define the unit, then Sample on purpose, then Judge with clinicians, then Report the margin. Report the unit (per sentence, per note, per patient), the sampling method, the severity scale, and how often the automated grader agreed with clinicians compared with how often clinicians agreed with each other.

What should I watch out for with calibration and appropriate abstention?

A single average hides rare, serious errors and weaker subgroups. A good report gives: a number with a clear unit, a stated sample, a margin and the grader's agreement with clinicians.

Can calibration and appropriate abstention be automated?

Only in part. It needs automatable and clinician judgement. Software can screen, but a clinician must judge the cases that matter.

Measure calibration and appropriate abstention on your own data