Cockcroft–Gault vs CKD-EPI

Use Cockcroft–Gault for drug dosing and CKD-EPI 2021 for CKD staging. They estimate different quantities, in different units, from different inputs — so the two numbers can differ for the same patient and are not interchangeable.

Output and Units

Cockcroft–Gault returns an absolute creatinine clearance in mL/min. CKD-EPI returns an eGFR indexed to 1.73m² of body surface area. For an average adult the two are close; for an unusually large or small patient they diverge, and the eGFR must be de-indexed to absolute mL/min before any direct comparison. Use the GFR unit converter for that step.

Inputs

The decisive difference is weight: Cockcroft–Gault uses it, CKD-EPI does not. Because of that, a patient's body size shifts the Cockcroft–Gault result but never the indexed eGFR. In obesity the weight chosen matters — use adjusted body weight rather than total body weight to avoid overestimating clearance.

Validation and Accuracy

Cockcroft–Gault was derived in 1976 from 249 hospitalized adults and predicts a timed creatinine clearance, which runs slightly above true GFR because of tubular secretion. CKD-EPI 2021 was developed against measured GFR in large modern cohorts and is the more accurate estimate of true filtration. Neither measures GFR directly; both estimate it from serum creatinine.

Primary Use

Most FDA drug labels and renal-dosing studies were validated against Cockcroft–Gault, so it is the equation whose values map to published dose thresholds in renal drug dosing. CKD-EPI 2021 is the equation guidelines recommend for assigning a CKD stage (G1–G5) and reporting kidney function. Matching the equation to the task keeps dosing aligned with the evidence behind each drug label.

A Worked Comparison

For a 78-year-old man, 60 kg, serum creatinine 1.4 mg/dL: Cockcroft–Gault gives about 38 mL/min of creatinine clearance, while his CKD-EPI eGFR is roughly 50 mL/min/1.73m². Both describe reduced function, but only the Cockcroft–Gault value should be used to check a drug's dose threshold; only the eGFR should be used to assign his CKD stage.

The Two Equations Side by Side

Written out, the contrast is concrete. Cockcroft–Gault is [(140 − age) × weight in kg × (0.85 if female)] ÷ (72 × serum creatinine in mg/dL), returning mL/min. CKD-EPI 2021 is 142 × min(Scr/κ, 1)^α × max(Scr/κ, 1)^−1.200 × 0.9938^age × (1.012 if female), with κ = 0.7 female / 0.9 male and α = −0.241 female / −0.302 male, returning mL/min/1.73m². The first multiplies an age term by body weight; the second runs serum creatinine through a two-part spline and an age decay. They share only two inputs in common form — age and serum creatinine — and differ on everything else.

Why Weight Changes Everything

Weight is the single input that splits these equations. Cockcroft–Gault scales clearance to body weight, so two patients with identical age, sex, and creatinine but different weights get different clearances — which is appropriate for dosing, since larger patients clear more. CKD-EPI deliberately omits weight because it reports a rate already indexed to a standard 1.73m² body, so the same creatinine yields the same eGFR regardless of size. This is why the two values diverge most at the extremes of body habitus, and why the weight chosen for Cockcroft–Gault matters: total body weight overstates clearance in obesity, so adjusted body weight is used instead. The ideal body weight calculator supplies the starting figure.

How Tubular Secretion Separates Them

The two equations also target subtly different physiology. Cockcroft–Gault predicts creatinine clearance, which includes the small amount of creatinine the tubules secrete on top of what the glomerulus filters, so it runs above true GFR. CKD-EPI is calibrated against measured GFR, so it targets filtration itself. The practical consequence is that even a perfectly computed Cockcroft–Gault value will tend to read higher than a perfectly computed eGFR for the same patient — not an error, but a reflection of what each is built to estimate. See creatinine clearance vs GFR for the underlying distinction.

Common Mistakes to Avoid

The recurring error is using the wrong equation for the task. Dosing a renally-cleared drug from an eGFR can move a small or elderly patient across a label threshold and cause overdosing, because the indexed eGFR reads higher than the patient's true absolute clearance. The reverse — staging a patient from a Cockcroft–Gault value — ignores body-surface indexing and the guideline cutoffs that were defined on eGFR. A third mistake is comparing the raw numbers without de-indexing the eGFR to absolute mL/min first; for an average adult they are close, but for an unusually large or small patient they are not, and the GFR unit converter handles that step.

A Second Worked Example

Compare a young, muscular patient with an older, frail one. A 30-year-old man, 90 kg, serum creatinine 1.0 mg/dL, has a Cockcroft–Gault clearance of about 138 mL/min and a CKD-EPI eGFR near 100 mL/min/1.73m² — both report robust function, with the weight-scaled absolute clearance running higher. Now an 85-year-old woman, 45 kg, serum creatinine 1.0 mg/dL, has a Cockcroft–Gault clearance of only about 24 mL/min while her eGFR is around 55 mL/min/1.73m². Here the gap is large and clinically decisive: dosing must follow the low Cockcroft–Gault value, not the more comfortable-looking eGFR.

Validation Cohorts in Detail

The cohorts behind each equation explain much of their behaviour. Cockcroft–Gault was fit to 249 hospitalized adults in 1976, a small and relatively narrow sample, and it was regressed against timed creatinine clearance rather than a measured filtration marker. CKD-EPI 2021 was developed from large, pooled, contemporary datasets spanning the full range of kidney function and was anchored to measured GFR. The newer, broader, GFR-anchored derivation is why CKD-EPI is the more accurate estimate of true filtration, while the older clearance-anchored derivation is why Cockcroft–Gault maps cleanly onto the dosing literature that grew up alongside it.

Why Drug Labels Still Specify Cockcroft–Gault

It is reasonable to ask why dosing has not simply moved to the more accurate equation. The answer is consistency of evidence: the dose-finding and safety studies behind most renally-cleared drugs binned patients by Cockcroft–Gault creatinine clearance, and FDA labels codified those bins. A threshold such as “reduce the dose below 30 mL/min” therefore refers to a Cockcroft–Gault value, and applying a different equation breaks the link between the number and the evidence that justified the dose. Until labels are re-derived against eGFR, Cockcroft–Gault remains the equation that keeps dosing aligned with the studies. See renal drug dosing for how the thresholds are applied in practice.

Which to Use When: The Verdict

Cockcroft–Gault for dosing, CKD-EPI 2021 for staging. When the two numbers disagree for one patient, that is expected — read each in its own context rather than forcing them to match, and de-index the eGFR before any direct comparison. For the broader picture, read CrCl vs eGFR, or calculate with the CrCl calculator and eGFR calculator.