Lead time

Hours of warning before event.

Who judgesReal-world outcomes
EvidenceGrade E4
Product typePredictive modelsID PM4

What lead time means

Hours of warning before event.

Why it matters. Warnings must come early enough to act.

Example, illustrative

A sepsis alert fires six hours before the event.

How to measure it

01

Use local data

Validate on your own recent patients.

02

Check discrimination and calibration

Both, not just AUROC.

03

Measure alert burden

Positive predictive value and alerts per day.

04

Monitor over time

Track subgroups and drift after go-live.

Watch out for

A strong AUROC can coexist with poor calibration and heavy alert fatigue.

What a good report shows

Local discrimination, calibration, PPV at your prevalence, and drift tracked over time.

Report discrimination and calibration on local data, the alert rate, positive predictive value, and drift over time.

Reviewed by the Precision research team.

How strong is the evidence

E4 · Recalled, not verified. Recalled from background knowledge and not verified. Treat it as a lead for investigation, not a finding.

What the sources say

E4.

  • Real-world outcomes. Needs outcome data from live use, such as time saved, burnout or patient results.

Found something out of date?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

Quick answers

What is lead time in healthcare AI?

Hours of warning before event. It sits in the predictive models group.

Why does lead time matter?

Warnings must come early enough to act. A sepsis alert fires six hours before the event.

How is lead time measured?

Use local data, then Check discrimination and calibration, then Measure alert burden, then Monitor over time. Report discrimination and calibration on local data, the alert rate, positive predictive value, and drift over time.

What should I watch out for with lead time?

A strong AUROC can coexist with poor calibration and heavy alert fatigue. A good report gives: local discrimination, calibration, PPV at your prevalence, and drift tracked over time.

Can lead time be automated?

Only in part. It needs real-world outcomes. It depends on data from real use, not a test set.

Measure lead time on your own data