Clinical Scores: 29 Bedside Formulas with Sources

Clinical Scores is a six-panel reference bench for 29 published scores, equations and scales: Renal & Dosing, Cardiovascular, Acute Care, Hepatic, Screening and Metabolic. Every card carries its band, item arithmetic, derivation population, exclusions and citation. Versions run from 1916 to 2024; superseded and current forms coexist — VASc beside VA, MELD-Na beside MELD 3.0 — because guidelines and labels still cite the older ones.

Updated 2026-10-01Reviewed by CalcX Editorial Team6 sources5 min read

How to use the panels

Pick a panelfrom the tab strip or open its address (linked above).
Set the shared fields first.Sex and age feed several cards at once — CHA₂DS₂-VASc, eGFR, ASCVD, APACHE II.
Enter values or tick items.Fields take clinical units (mg/dL, µmol/L, mmol/L); scales use 0–3 buttons per item; invalid entries fail.
Read the whole card:score and maximum, band, extra cards (stroke rate, absolute GFR, mortality) and the item table.
Check warnings and the source card before quoting.Amber notes flag extrapolation and conventions; the purple card gives population, exclusions and caveat.
Compare versions deliberately:their bands and thresholds differ, so match the year your protocol cites.

Worked readouts

Fixed inputs.

Panel	Fixed input	Readouts
Renal	male 65 y, creatinine 1.2, 80 kg, 175 cm	eGFR 67.1 mL/min/1.73 m² (G2); C-G 69.4/61.2/64.5 mL/min (total/ideal/adjusted)
Cardio	male 72 y, no risk flags	CHA₂DS₂-VASc 1/9 (≈0.6%/year); CHA₂DS₂-VA 1/8
Cardio	QT 400 ms, HR 100, male	Bazett 516.4, Fridericia 474.3, Framingham 461.6, Hodges 470 ms; spread 54.8 ms
Cardio	White male 55 y, TC 213, HDL 50, SBP 120 untreated	ASCVD 5.4% (Borderline); optimal peer 3.6%
Acute	70 y, RR 24, 95/60 mmHg, E4V5M6, urea 9	qSOFA 2/3; CURB-65 3/5 (high, 14.0% 30-day); GCS 15/15
Hepatic	bilirubin 3.0, creatinine 1.5, INR 1.8, sodium 130, albumin 2.8	MELD(i) 21 → MELD-Na 25 (≈19.6% 3-month); Child-Pugh 8 (class B)
Hepatic	55 y, AST 60, ALT 40, platelets 150 ×10⁹/L	FIB-4 3.48 (> 3.25)
Screen	PHQ-9 3,3,3,2,2,2,1,1,0; AUDIT-C 1,1,1	PHQ-9 17/27; AUDIT-C 3/12 (cut-offs 4/3)
Metabolic	insulin 10 µU/mL, glucose 5.5 mmol/L; TC 200, HDL 50, TG 150	HOMA-IR 2.44; LDL 120, VLDL 30, non-HDL 150 mg/dL

The QTc card disagrees: at HR 60 all four corrections return the entered QT, at 100 bpm they span 54.8 ms, and at QT 390/HR 95 they straddle the 450 ms male threshold.

Where the arithmetic bends

Validity travels with the source population.: CKD-EPI 2021 assumes stable adult kidney function; FIB-4 warns below 35 and lifts its low cut-off from 1.45 to 2.0 at ≥65; a negative qSOFA does not exclude sepsis; HOMA-IR has no universal cut-off (insulin assays are not standardised).
Conventions are part of the arithmetic.: MELD floors bilirubin, INR and creatinine at 1.0, caps creatinine at 4.0 (3.0 in MELD 3.0), corrects for sodium only above MELD(i) 11 and clamps it to 125–137; GCS records V1 when intubated; Wells PE's half points put 4.5 at "PE likely" while 6.0 stays moderate; APACHE II's mortality logit omits diagnostic weights; Epworth is paraphrased and copyrighted.
What it does not do.: Educational clinical-reference arithmetic: it cannot diagnose, cannot replace examination or testing, and must not drive a treatment decision. It omits full SOFA, PERC, GRACE and NIHSS, stores no records and gives no individual advice. For a wider well-being battery use Assessment; for decision support follow the sources below.

Every score is a dated specification

A score is a dated specification, not a constant. CKD-EPI 2021 is race-free: creatinine 1.3 mg/dL in a 50-year-old man reads 66.9 mL/min/1.73 m² with no race input. The 2013 Pooled Cohort Equations one panel over still carry four coefficient sets by sex and race — the same 55-year profile returns 2.1% (White female), 3.0% (Black female), 5.4% (White male) and 6.1% (Black male) — and name PREVENT (2023) as the race-free successor. CHA₂DS₂-VA removes the female point: a 70-year-old woman with hypertension and diabetes scores 4 on the 2010 scale and 3 on the 2024 one. MELD 3.0 adds sex and albumin, caps creatinine at 3.0 instead of 4.0, and returns 27 (female) or 26 (male) on labs that give MELD-Na 25.

Where it helps

Reproducing a published example

The 2013 ACC/AHA appendix prints the 55-year example above; the four QTc corrections share a fixed point at HR 60. Enter the paper's inputs and compare.

Teaching what counts

Wells PE gives 1.5 points to three items and 3 to two; APACHE II scores a GCS deficit (15 − GCS), not the GCS; a DOAC patient scores the HAS-BLED labile-INR item as 0.

Checking a quoted number

Units drift: creatinine ÷ 88.4, glucose × 18.016, TG ÷ 5 (mg/dL) or ÷ 2.2 (mmol/L). Friedewald gives LDL 120 at TG 150 but 170 and an invalid flag at TG 450.

References

Inker LA et al., New creatinine- and cystatin C-based equations to estimate GFR without race, N Engl J Med 385:1737–1749 (2021), pubmed.ncbi.nlm.nih.gov （访问日期：2026-10-01）
Cockcroft DW & Gault MH, Prediction of creatinine clearance from serum creatinine, Nephron 16:31–41 (1976), pubmed.ncbi.nlm.nih.gov （访问日期：2026-10-01）
Lip GYH et al., Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation, Chest 137:263–272 (2010), pubmed.ncbi.nlm.nih.gov （访问日期：2026-10-01）
Van Gelder IC et al., 2024 ESC Guidelines for the management of atrial fibrillation, Eur Heart J 45:3314–3414 (2024), pubmed.ncbi.nlm.nih.gov （访问日期：2026-10-01）
Goff DC Jr et al., 2013 ACC/AHA guideline on the assessment of cardiovascular risk, Circulation 129(25 Suppl 2):S49–S73 (2014), pubmed.ncbi.nlm.nih.gov （访问日期：2026-10-01）
Kim WR et al., MELD 3.0: the model for end-stage liver disease updated for the modern era, Gastroenterology 161:1887–1895 (2021), pubmed.ncbi.nlm.nih.gov （访问日期：2026-10-01）