eGFR and the urine albumin test
Last reviewed November 17, 2026. OwieClear editorial team. Waiting for clinician review. Not medical advice. Talk with your doctor about your own care.
Two tests define chronic kidney disease. A blood test estimating how well the kidneys filter, and a urine test showing whether the filter is leaking.
They answer different questions, and the urine one, which carries most of the information about what happens next, is the one most often left undone.
Related pages: Chronic kidney disease
The bottom line
- eGFR is calculated from blood creatinine with age and sex.
- The urine albumin-to-creatinine ratio detects damage before eGFR falls.
- CKD requires the abnormality to persist for three months.
- Muscle mass, illness, hydration and recent meals all affect creatinine and so affect eGFR.
- Cystatin C provides a second estimate when creatinine is unreliable.
- Whether general population screening is justified, which guidelines still dispute.
- How widely cystatin C should be used.
- Whether eGFR thresholds should be age-adjusted.
The story, step by step
- 1. The blood test ()
- 2. The urine test ()
- 3. When the numbers mislead ()
- 4. The race-free equation ()
- 5. Who should be tested ()
What the evidence labels mean
What eGFR is telling you
What happens
An estimate of filtering capacity.
What the research shows
- Creatinine is a waste product of muscle, produced at a fairly steady rate and cleared by the kidneys. When filtering falls, it accumulates.
- eGFR converts creatinine into an estimated filtration rate using age and sex, reported in mL/min/1.73m2.
- Above 90 is normal filtering. 60 to 89 is mildly reduced and is common with age. Below 60 for three months defines CKD.
- Below 30 is severely reduced and usually prompts nephrology referral. Below 15 is kidney failure.
- Small fluctuations of a few points between tests are normal and are not progression.
- A single low result is not a diagnosis. It needs repeating after at least three months to distinguish CKD from acute kidney injury.
Why it matters
People frequently see an eGFR in a results portal before anyone explains it. One number, uncontextualised and unrepeated, is not a diagnosis of anything.
What's still uncertain
- Whether reporting eGFR without context does more harm than good.
- The best repeat interval for a borderline first result.
Read more
- NIDDK: Tests and diagnosis of chronic kidney diseaseGovernment
- MedlinePlus: Kidney testsGovernment
- KDIGO: CKD evaluation and management guidelineNonprofit
Next: The urine test
The one that gets skipped
What happens
Albumin-to-creatinine ratio.
What the research shows
- A spot urine sample, ideally first thing in the morning, measured for albumin and creatinine, and reported as a ratio in mg/g.
- Under 30 is A1, normal to mildly increased. 30 to 300 is A2. Over 300 is A3, severely increased.
- A standard urine dipstick for protein is much less sensitive and misses the early range entirely, so a negative dipstick does not rule this out.
- An abnormal result should be confirmed on a repeat sample, because exercise, fever, infection, heart failure and menstruation can all raise it temporarily.
- Albuminuria predicts both progression to kidney failure and cardiovascular events, independently of eGFR.
- Testing rates remain low. In people with diabetes, where it is recommended annually, a substantial share never receive it.
Why it matters
If only one test is done, the eGFR gets ordered and the urine test does not. Asking for the albumin-to-creatinine ratio by name is the single most useful thing a patient can do.
What's still uncertain
- How to raise testing rates in primary care.
- Whether first-morning samples should be required.
Read more
- NIDDK: Tests and diagnosis of chronic kidney diseaseGovernment
- KDIGO: CKD evaluation and management guidelineNonprofit
- NIDDK: Diabetic kidney diseaseGovernment
Next: When the numbers mislead
Known failure modes
What happens
Mostly about muscle and timing.
What the research shows
- eGFR reads falsely high in people with low muscle mass, including many older, frail or amputee patients, and in advanced liver disease.
- It reads falsely low in very muscular people, in those taking creatine supplements, and after intense exercise.
- A large cooked-meat meal before the blood test raises creatinine for hours.
- Some drugs, including trimethoprim and cimetidine, raise creatinine by blocking its secretion without changing actual filtering.
- Dehydration, vomiting, diarrhoea and acute illness reduce kidney function temporarily; a test taken then shows acute injury, not chronic disease.
- Cystatin C, produced by all nucleated cells, is less affected by muscle and is recommended to confirm when the creatinine-based result does not fit.
Why it matters
A surprising number in a frail older person or a bodybuilder should prompt a second look rather than a new diagnosis.
What's still uncertain
- How often creatinine-based eGFR misclassifies people in practice.
- The cost-effectiveness of routine cystatin C confirmation.
Read more
- KDIGO: CKD evaluation and management guidelineNonprofit
- NIDDK: Tests and diagnosis of chronic kidney diseaseGovernment
- MedlinePlus: Kidney testsGovernment
Next: The race-free equation
What changed in 2021
What happens
An adjustment that was removed.
What the research shows
- Earlier eGFR equations applied an upward adjustment for Black patients, based on average differences in measured creatinine in the study populations.
- The adjustment used race, a social category, as a biological input, and it produced higher reported eGFR for the same creatinine.
- A higher reported eGFR means a later diagnosis, later referral, and later placement on a transplant waiting list.
- United States laboratories adopted a race-free equation in 2021, and transplant programs subsequently recalculated waiting time for affected patients.
- The change reclassified some people into a lower eGFR category, which was the point: those people had kidney disease that the previous equation had understated.
- Black Americans experience kidney failure at several times the rate of white Americans, driven by differences in diabetes and blood pressure prevalence, access to care, and in some people the APOL1 gene variants.
Why it matters
This is a case where a measurement convention, rather than any treatment, was delaying care for a specific group. Changing the arithmetic changed access.
What's still uncertain
- The downstream effect of the change on transplant access.
- How APOL1 testing should be used in practice.
Read more
- KDIGO: CKD evaluation and management guidelineNonprofit
- NIDDK: Tests and diagnosis of chronic kidney diseaseGovernment
- National Kidney FoundationNonprofit
Next: Who should be tested
And how often
What happens
Risk-based, not universal.
What the research shows
- Annual testing is recommended for people with diabetes, high blood pressure, cardiovascular disease, or a family history of kidney failure.
- Also for those with a previous episode of acute kidney injury, long-term NSAID or lithium use, HIV, or a history of kidney stones or recurrent urinary infections.
- Both tests should be done together. An eGFR alone misses stage 1 and 2 disease entirely.
- The United States Preventive Services Task Force has concluded the evidence is insufficient to recommend screening the general adult population without risk factors, which is a statement about evidence rather than about the tests themselves.
- Established CKD is monitored at an interval set by the risk category, from annually at low risk to several times a year at high risk.
- Kidney function should also be rechecked after any serious illness, hospital admission, or new medicine likely to affect the kidneys.
Why it matters
The people most likely to benefit are easy to identify, and most of them are already attending for something else.
What's still uncertain
- Whether broader screening would be cost-effective now that effective treatments exist.
- The optimal monitoring interval by risk category.