28sungmin/m4-add-skills/meld-score
Claude Code and agent skill by 28sungmin/m4-add-skills
- community
Checking that the SKILL.md file is still in the repository.
Calculate MELD (Model for End-Stage Liver Disease) score for ICU patients. Use for liver disease severity assessment, transplant prioritization, or hepatology research.
Install
npx vskill@latest install 28sungmin/m4-add-skills/meld-scoreSKILL.md
MELD Score Calculation
The Model for End-Stage Liver Disease (MELD) score predicts 3-month mortality in patients with liver disease. Originally developed for TIPS procedure prognostication, it is now the primary organ allocation criterion for liver transplantation in the US. The MELD-Na variant incorporates serum sodium for improved prediction.
M4Bench Use
In M4Bench, target concept tables listed in the task configuration are removed or unavailable in the agent database. Use this skill as procedural guidance and derive the requested output from available source or intermediate tables; do not rely on a precomputed target table or bundled SQL script.
When to Use This Skill
- Liver disease severity assessment
- Transplant prioritization research
- Hepatology outcome studies
- Cirrhosis prognosis evaluation
Score Components
MELD uses 4 laboratory values with logarithmic transformations:
| Component | Variable | Formula Weight | Floor | Ceiling | |
|---|---|---|---|---|---|
| Creatinine | Serum creatinine (mg/dL) | 0.957 × ln(Cr) | 1.0 | 4.0 (or if RRT) | |
| Bilirubin | Total bilirubin (mg/dL) | 0.378 × ln(Bili) | 1.0 | — | |
| INR | International Normalized Ratio | 1.120 × ln(INR) + 0.643 | 1.0 | — | |
| Sodium | Serum sodium (mEq/L) | 137 - Na | 125 | 137 |
Calculation Steps
- Component scores (all values floored at 1.0 for logarithm):
- Creatinine:
0.957 × ln(max(Cr, 1))— capped at Cr=4 if RRT or Cr > 4 - Bilirubin:
0.378 × ln(max(Bili, 1)) - INR:
1.120 × ln(max(INR, 1)) + 0.643
- MELD Initial:
round(Cr_score + Bili_score + INR_score, 1) × 10
- Capped at 40 if sum > 4
- Sodium adjustment (MELD-Na, only if MELD Initial > 11):
- Sodium score:
137 - max(min(Na, 137), 125) MELD = MELD_Initial + 1.32 × Na_score - 0.033 × MELD_Initial × Na_score- If MELD Initial ≤ 11: MELD = MELD Initial (no sodium adjustment)
Critical Implementation Notes
- Creatinine Cap: Patients on renal replacement therapy (RRT/dialysis) OR with creatinine > 4.0 are scored as creatinine = 4.0. This prevents extreme renal failure from dominating the score.
- INR Constant Term: Unlike other components, INR scoring adds a constant 0.643 even when INR = 1 (ln(1) = 0). This means the minimum contribution from INR is always 0.643.
- Sodium Conditional: The sodium adjustment is ONLY applied when MELD Initial > 11. Below this threshold, mild liver disease is not affected by hyponatremia.
- Score Range: Theoretical minimum ~6 (all normal values), maximum 40 (hard cap).
- RRT Detection: In raw M4Bench tasks, detect dialysis/renal replacement therapy from available procedure or charted treatment records in the first 24 hours.
- Missing Lab Values: When a lab value is not available for a stay, the component defaults to its minimum contribution: ln(1) = 0. This means creatinine scores as 1.0, bilirubin scores as 1.0, INR scores as 1.0 (but still contributes the +0.643 constant), and sodium defaults to 137 (no adjustment). Include ALL ICU stays in the output, even those missing some or all lab values.
Required Tables for Custom Calculation
mimiciv_icu.icustays— ICU stay identifiersmimiciv_hosp.labevents— creatinine, total bilirubin, INR, sodium- ICU procedure or charting tables — dialysis/renal replacement therapy flags
References
- Kamath PS et al. "A model to predict survival in patients with end-stage liver disease." Hepatology. 2001;33(2):464-470.
- Kim WR et al. "Hyponatremia and mortality among patients on the liver-transplant waiting list." N Engl J Med. 2008;359(10):1018-1026.
- Biggins SW et al. "Evidence-based incorporation of serum sodium concentration into MELD." Gastroenterology. 2006;130(6):1652-1660.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingbenchmark/tasks/meld/mimic-meld-24h-raw/skills/meld-score/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/28sungmin/m4-add-skills/meld-score)