Kidney Failure Risk Calculator (KFRE)
The Kidney Failure Risk Equation (KFRE) estimates the probability that someone with chronic kidney disease (CKD) will progress to kidney failure requiring dialysis or a transplant within 2 years and within 5 years. Enter your age, sex, eGFR, and urine albumin-to-creatinine ratio (ACR) for the standard 4-variable result, or switch to the 8-variable model and add four serum lab values for a more precise estimate. Validated in more than 700,000 patients across 30 countries.
What is the Kidney Failure Risk Equation?
The Kidney Failure Risk Equation (KFRE) was developed by Navdeep Tangri and colleagues and published in JAMA in 2011, then validated in a landmark multicohort study across 31 cohorts and 721,357 patients in 2016. It uses a Cox proportional hazards model to estimate the absolute probability that a person with chronic kidney disease (CKD) will progress to kidney failure requiring dialysis or a kidney transplant within two years and within five years. Two versions exist: a 4-variable model using age, sex, estimated glomerular filtration rate (eGFR), and the urine albumin-to-creatinine ratio (ACR); and an 8-variable model that adds serum albumin, phosphate, bicarbonate, and corrected calcium for a modest improvement in precision. The equation has been externally validated across diverse populations in North America, Europe, Asia, Australia, and the Middle East, and is endorsed by KDIGO (Kidney Disease: Improving Global Outcomes) as the preferred risk stratification tool for CKD stages G3-G5.
How the KFRE formula works
The KFRE uses a survival analysis framework. For the 4-variable model, a linear predictor (LP) is calculated as: LP = -0.2201 x (age/10 - 7.036) + 0.2467 x (sex - 0.5642) - 0.5567 x (eGFR/5 - 7.222) + 0.4510 x (ln(ACR) - 5.137), where sex is 1 for male and 0 for female, and ln(ACR) is the natural logarithm of ACR in mg/g. The 2-year risk is then 1 - 0.9832^exp(LP) and the 5-year risk is 1 - 0.9365^exp(LP). The baseline survival values (0.9832 and 0.9365) represent the fraction of North American CKD patients who had not yet reached kidney failure at 2 and 5 years respectively in the derivation cohort. The 8-variable model extends this LP with additional terms for serum albumin, phosphate, bicarbonate, and calcium, each centred on their mean values from the derivation cohort. Higher eGFR reduces risk (negative coefficient), while higher ACR and older age increase risk. Male sex is associated with slightly higher risk than female sex in the derivation data.
KFRE in clinical decision-making
Three KFRE thresholds have been widely adopted. A 5-year risk above 5% is the threshold most guidelines (including NICE in the UK) use to support a decision to refer to a nephrologist, alongside other clinical criteria such as rapidly declining eGFR, resistant hypertension, or suspected rare kidney disease. A 2-year risk above 10% is the point at which multidisciplinary team-based kidney care, covering dietitians, social workers, and patient education, is typically initiated. A 2-year risk above 40% signals the period when vascular access creation (for haemodialysis), peritoneal dialysis catheter planning, or kidney transplant evaluation should be actively pursued, because lead times for these interventions can be several months. The KFRE does not replace clinical judgment: some patients with low KFRE scores still warrant prompt referral because of the underlying cause of their CKD or rate of decline, and some patients with high scores choose conservative management.
Inputs, units, and common questions
eGFR is most commonly calculated from serum creatinine using the CKD-EPI 2021 equation (or the older 2009 equation), and should be from a steady-state sample not taken during an acute illness. The KFRE was validated for CKD stages G3a through G5 (eGFR 1-60 mL/min/1.73m²); it is not intended for patients with acute kidney injury or eGFR above 60. ACR can be measured from a spot urine sample (preferred over a 24-hour collection for practicality); US labs report in mg/g, while UK and Canadian labs typically use mg/mmol - this calculator converts automatically using the factor 1 mg/mmol = 8.84 mg/g. The 8-variable inputs (serum albumin, phosphate, bicarbonate, calcium) are standard chemistry panel values. Serum calcium should be corrected for albumin if using a colorimetric assay: corrected calcium (mg/dL) = measured calcium + 0.8 x (4.0 - albumin in g/dL).
KFRE risk categories and clinical thresholds
| 5-year KFRE risk | Risk category | Clinical guidance |
|---|---|---|
| < 5% | Low | Routine primary care monitoring |
| 5 - 9% | Low-moderate | Consider nephrology referral |
| 10 - 19% | Moderate | Nephrology referral recommended |
| 20 - 39% | High | Urgent nephrology referral; start KRT education |
| >= 40% | Very high | KRT planning and vascular access preparation |
Based on 5-year kidney failure risk and published guideline recommendations (KDIGO 2022, kidneyfailurerisk.com).
Frequently asked questions
What does the KFRE result actually mean?
The result is an absolute probability expressed as a percentage. A 2-year risk of 8% means that, among 100 patients with the same values as you, approximately 8 would reach kidney failure (need dialysis or a transplant) within 2 years. It is not a certainty for any individual - some people with high KFRE scores progress slowly, and others with low scores progress quickly, especially if a treatable cause is missed.
What is the difference between the 4-variable and 8-variable KFRE?
Both models use age, sex, eGFR, and urine ACR. The 8-variable model adds serum albumin, phosphate, bicarbonate, and corrected calcium. The 8-variable version has a slightly higher C-statistic (discrimination) in the original derivation cohort, but the improvement over the 4-variable model is modest and the 4-variable model performs comparably in most external validations. The 4-variable model is preferred when serum chemistry results are unavailable or when simplicity is a priority.
Can I use this calculator if my eGFR is above 60?
No. The KFRE was developed and validated specifically in patients with CKD stages G3 through G5, meaning an eGFR below 60 mL/min/1.73m². Using the equation outside this range produces unreliable results. Patients with eGFR above 60 but with significant albuminuria (CKD stages G1-G2) should be evaluated with other risk stratification methods.
Why does the KFRE show male sex as higher risk?
In the derivation cohorts, men with the same age, eGFR, and ACR values had a slightly higher observed rate of kidney failure progression than women. The coefficient for sex reflects this epidemiological pattern. Biological differences in creatinine metabolism, muscle mass, and hormonal factors may contribute, but the mechanism is not fully established. The 2021 CKD-EPI creatinine equation for eGFR no longer includes a race variable and uses sex to help account for differences in creatinine production.
Is a high KFRE score reversible?
The KFRE gives a snapshot probability based on current values. If eGFR and ACR improve, for example through better blood pressure control, use of RAAS inhibitors (ACE inhibitors or ARBs), SGLT2 inhibitors, or finerenone, the KFRE score will improve too. The equation is most useful as a dynamic tool: recalculating it at each clinic visit allows clinicians and patients to see whether interventions are reducing the projected risk trajectory.
What is the difference between 2-year risk and 5-year risk in the KFRE?
Both timeframes estimate the cumulative probability of reaching kidney failure (requiring dialysis or transplant). The 2-year risk is used primarily to guide timing of kidney replacement therapy preparation, because creating vascular access for haemodialysis or listing for a transplant takes months. The 5-year risk is used more often for referral decisions, because nephrology involvement is most valuable when there is still time to slow progression and plan ahead. Both numbers come from the same linear predictor: only the baseline survival value (0.9832 vs 0.9365 for the 4-variable North American model) differs.
Does the KFRE apply to people with diabetes or other specific kidney diseases?
Yes, in external validation studies the KFRE has performed well across a range of underlying kidney diseases including diabetic nephropathy, IgA nephropathy, hypertensive nephrosclerosis, and polycystic kidney disease. However, discrimination and calibration vary somewhat by underlying diagnosis. The KFRE was also validated in kidney transplant recipients. Clinicians should use local recalibration data if available, since the absolute risk level can differ between populations even when the equation ranks patients correctly.
Sources
- Tangri N et al. Multinational Assessment of Accuracy of Equations for Predicting Risk of Kidney Failure: A Meta-Analysis. JAMA. 2016;315(2):164-174.
- Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int. 2024.
- Kidney Failure Risk Equation - Official Calculator and Clinical Practice Points (kidneyfailurerisk.com).