Layered Labs

AppliedNovember 2026

When education shouldn't matter: education cues change AI triage

0
patients, clinical details unchanged
0.00%
changed triage on one education cue
0.000
accuracy, the same throughout

1Abstract

Large language models are being proposed for clinical decision support, including mass-casualty triage, where who is treated first must depend only on how badly they are hurt. We tested whether two models keep their triage decisions the same when nothing changes but the patient's stated education level, a detail that should play no part in disaster triage. We took the 87 vignettes of the TRIAGE benchmark, added one sentence giving the patient's education, kept every clinical detail the same, and asked Qwen-2.5-72B and GPT-4o-mini to triage each version. Overall accuracy barely moved: Qwen-2.5-72B scored 0.644 under every condition, and GPT-4o-mini between 0.598 and 0.644. But both models changed individual decisions. Qwen-2.5-72B changed its triage for 5.75% of patients with the lowest education cue and 10.34% with the highest, and its changes leaned toward less urgent care; two of the three conditions were statistically significant. GPT-4o-mini changed fewer decisions, 4.60% to 5.75%, and leaned the other way, toward more urgent care, though not significantly. Education alone can change an AI triage decision, and aggregate accuracy does not show it.

2Introduction

Illustration. The same patient twice: the only difference is one sentence about education. The triage zone should be the same, but can change.

Mass-casualty triage ranks patients by how badly they are hurt and how soon they need care. Nothing else should matter: not who the patient is, what they do, or how much schooling they have. As language models are tried for clinical decision support, including triage, it is worth checking that their decisions follow the same rule.

Earlier work has shown that language models can change their decisions when given demographic details that have nothing to do with the task. In medicine that is a fairness problem: two patients with the same injuries should get the same decision.

We ask a narrow question. If the only thing that changes is one sentence about a patient's education, does a model change its triage?

3Methods

Every vignette is asked in all four conditions; only the education sentence differs.

Vignettes. The TRIAGE benchmark (Kirch et al., 2025): 87 short mass-casualty patient descriptions, each labeled with one of four zones: green (minor), yellow (delayed), red (immediate) or black (expectant or deceased).

Education cues. Each vignette was asked four ways. The control has no education information. The other three add one sentence at the start, such as "Patient's highest education level is some college", for less than high school, some high school and some college. Everything else is identical, and the cue says nothing about the patient's condition.

Models. Qwen-2.5-72B, a large open-weight model, and GPT-4o-mini, a closed model in wide production use, both run at temperature 0, once per vignette and condition.

Measures. Accuracy and macro F1 against the benchmark's labels, to check that the cue does not simply break the models. Then, comparing each cued answer with the same vignette's control answer: the flip rate, the share of patients whose zone changed, and the severity shift, the average change on a scale from green (0) to black (3), where a negative shift means less urgent care. Each flip pattern was tested with an exact McNemar test.

4Results

Each tile is one of the 87 patients; lime tiles got a different triage zone than with no education given.

Accuracy hides the change

Adding an education cue left overall performance almost unchanged:

Education conditionAccuracyMacro F1
Qwen-2.5-72B
No education0.6440.622
Less than high school0.6440.607
Some high school0.6440.606
Some college0.6440.607
GPT-4o-mini
No education0.6320.631
Less than high school0.5980.598
Some high school0.6440.641
Some college0.6320.630
Accuracy and macro F1 against the benchmark's labels, by education condition.

Both models changed decisions

The same patients got different triage zones when only their education changed:

Education cueFlipsFlip rateSeverity shiftp
Qwen-2.5-72B
Less than high school55.75%−2.30%0.063
Some high school66.90%−3.45%0.031
Some college910.34%−2.30%0.004
GPT-4o-mini
Less than high school44.60%+4.60%0.125
Some high school44.60%+2.30%0.125
Some college55.75%+3.45%0.063
Changes against the no-education control for the same 87 patients. A negative severity shift means less urgent triage. p is from an exact McNemar test.

Qwen-2.5-72B's flip rate rose with each step up in education, from 5.75% to 10.34%, and its decisions moved toward less urgent care in every condition. Two of its three conditions were significant. GPT-4o-mini changed fewer decisions and moved toward more urgent care, but none of its changes were significant.

5Discussion

Mean severity shift against the control. One model leans toward less urgent care, the other toward more.

Neither model's triage was invariant to education, even though no clinical detail changed and decoding was deterministic, so the differences are not random noise.

The two models erred in opposite directions. Qwen-2.5-72B moved toward undertriage, which in a mass-casualty event can delay care for someone badly hurt. GPT-4o-mini moved toward overtriage, which uses up scarce resources faster. Bias in triage is a matter of direction as well as size.

One possible explanation is that an education cue brings in learned associations about status or competence that then affect how urgent a case seems. That is a hypothesis; this study does not test the mechanism.

The broader lesson is about evaluation. Accuracy and F1 were stable while individual decisions changed. Testing a model for clinical use needs to include whether its decisions stay the same for patients who are the same in every way that matters.

6Limitations

With 87 vignettes, a handful of flips decides significance: here, 9 flips for Qwen-2.5-72B under the some-college cue.
  • One phrasing. Each education cue was a single fixed sentence. Other wordings, or placing the cue elsewhere, may give different results.
  • Deterministic decoding only. All runs used temperature 0. Behavior under sampling may differ.
  • Small sample. 87 vignettes means a few flips decide whether a result is significant.
  • Two models, three cues. Other models, and other demographic attributes, were not tested.
  • Black zone. The meaning of the black zone varies between triage protocols, which affects how severity shifts involving it should be read.
  • Policy, not practice. The study measures how a model's decisions respond to a controlled change, not how often this would happen in a deployed system.

7Availability

Every vignette comes from the public TRIAGE benchmark.

The vignettes come from the TRIAGE benchmark (Kirch et al., 2025), which is publicly available. For questions about this work, email hello@layeredlabs.ai.

Living draft, updated November 2026. This page can change before formal publication.