← Blog · MedLegal AI

Cross-Examining ER Physicians on Chest Pain: The HEART Score Trap

Verify it yourself — free, no login

See how AI medical-record review links every fact to the exact Bates page that proves it — click any citation and jump straight to the record.

See the 60-second demo →
May 23, 2026 · 7-minute read · By John Mahoney

Missed acute myocardial infarction is the most expensive missed-diagnosis category in U.S. medmal litigation. Average paid claim sits around $470,000 per The Doctors Company data, with the largest verdicts north of $5 million for the under-40 patient with permanent heart-failure complications.

Yet defense counsel routinely walks these cases out with a "clinical judgment" framing. The ED doctor testifies they used their experience, the patient didn't look acutely ill, the troponin came back negative — done. Standard of care met.

The 2021 ACC/AHA Chest Pain Guideline made that argument significantly weaker. This post is the cross-examination script that exploits the gap.

What changed in 2021

The 2021 ACC/AHA/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain formally adopted risk-stratification scoring as the chest-pain standard. The HEART score (History, ECG, Age, Risk factors, Troponin) became the most-cited tool, with the HEART Pathway protocol providing the operational standard for ED chest-pain disposition.

What this means for litigation: an ED physician who discharges a chest-pain patient without documenting risk-stratification scoring no longer has the "clinical gestalt" defense that worked pre-2021. The guideline is now the standard, and the chart either documents the standard or it doesn't.

The cross-examination script

This is the core 10-question sequence. The goal is to lock the witness into admitting the validated decision rule is the standard, then walk through their chart to demonstrate they didn't use it.

Q1: Doctor, you're board-certified in emergency medicine, correct? — Yes. Q2: You're familiar with the 2021 ACC/AHA Chest Pain Guideline? — Yes. Q3: That guideline was developed jointly by the American College of Cardiology, the American Heart Association, and SAEM — the Society for Academic Emergency Medicine — correct? — Yes. Q4: It is currently the most widely-cited chest-pain guideline in the U.S. emergency medicine literature, correct? — Yes. Q5: The HEART score is the risk-stratification tool that guideline endorses for ED chest-pain evaluation, correct? — Yes. Q6: The HEART score has been validated in multiple studies involving tens of thousands of chest-pain patients? — Yes. Q7: A HEART score of 0 to 3 places a patient in the low-risk category — appropriate for discharge with outpatient follow-up? — Yes. Q8: A HEART score of 4 to 6 is intermediate risk — appropriate for observation and serial troponin, not discharge? — Yes. Q9: Looking at your medical record for [patient name], your MDM does not contain a documented HEART score, does it? — [chart speaks for itself] Q10: So when you decided to discharge [patient] on [date], you did not apply the most widely-cited risk-stratification tool that your specialty's joint guideline endorses for this exact clinical scenario, correct?

The witness has two losing options at Q10. They can confirm — which converts the case theory from "the doctor made a bad judgment call" to "the doctor failed to apply the documented standard of care." Or they can argue — which forces them to defend a position the guideline has already taken away from them.

The "I used my clinical judgment" defense, and why it fails

Defense counsel will try to rehabilitate with: "Doctor, you're not required to mechanically apply every published score, are you? You're trained to exercise clinical judgment?"

This is a strong move pre-2021. After the 2021 guideline, the witness's truthful answer makes things worse:

The Bates anchors you need

Before the deposition, identify the specific record pages that support the cross:

  1. The ED MDM (medical decision-making) page — the absence of a HEART score is the central exhibit
  2. The ED triage vitals page — confirms abnormal-vital trends not addressed
  3. The discharge summary — confirms the disposition decision was made on a single troponin or absent serial draws
  4. Hospital chest-pain protocol — request via discovery; nearly all U.S. AHA-accredited centers have one that mirrors ACC/AHA
  5. The defendant's prior CME records — confirms they were trained on the 2021 guideline (most are)

With those five Bates anchors, the cross becomes self-executing. The witness can confirm or fight every step, and every step still ends in the same place: the chart doesn't document the standard.

The Daubert side

The same canon works on the defense expert. If their expert reports try to justify the disposition without engaging the 2021 guideline, that's a methodology vulnerability under FRE 702(c). Motion in limine to limit testimony to the published canon, then on cross at trial: "Doctor, the ACC/AHA 2021 guideline is the current chest-pain standard in U.S. emergency medicine. Your report does not engage with it. Why?"

Practice the cross before you give it

The hardest part of cross-examining an ED physician is pacing. The script above is logically clean, but in a deposition the witness will resist Q5 and Q6 with hedges ("the HEART score is a tool, not a mandate"). You need to be drilled on the right closing-question follow-ups.

This is the kind of preparation that benefits from rehearsal against an AI that knows the same canon. Our deposition trainer's cardiology and emergency-medicine specialty packs include the full HEART/TIMI/GRACE drilldown, the ACC/AHA citation chain, and the most-common defense-witness deflections.

Try the 2-minute demo — no signup

Sample case is a missed SAH on thunderclap headache (neuro pack). Each question reveals the ideal witness response and the trap.

Try the 2-minute demo →

Bottom line

Pre-2021, missed-MI plaintiff cases turned on the "clinical judgment" defense. The 2021 ACC/AHA guideline took that defense away from cases where the chart doesn't document the standard. Most ED chest-pain charts still don't.

If you're working a chest-pain plaintiff case, the cross is mostly built. The work is the prep — and the work that decides the case is whether your expert can hold the line through Q10 without giving back the ground the guideline gave you.

— John Mahoney, Founder, MedLegal AI

See the AI cite its source — no login
Most legal AI is wrong 17–33% of the time. Watch MedLegal AI pin every finding to the exact record page — click any citation and it jumps to the line that proves it.
Watch the 30-second demo →