The KFRE: does our clinical practice limit its theoretical benefits?

The kidney failure risk equation (KFRE) estimates the risk of progression from chronic kidney disease (CKD) stage 3a or greater to “kidney failure” (defined as receipt of either dialysis or a kidney transplant).¹ The model is remarkably simple, relying on the four variables of age, sex, estimated glomerular filtration rate (eGFR) and urine albumin to creatinine ratio (uACR). First published by Navdeep Tangri in 2011, the equation has been recalibrated for use in numerous populations around the world, including that of the UK.² In 2021, NICE recommended inclusion of the KFRE within guidelines for referral of patients with CKD to specialist nephrology services. A KFRE-predicted risk of kidney failure >5% at 5 years now replaces the previous “threshold” of an eGFR <30 mL/min/1.73 m². The remainder of the previous 2014 NICE referral criteria remain unchanged.

The KFRE was added to the 2021 NICE guidelines to improve early identification and referral of patients who would most benefit from specialist care. By removing the prior eGFR threshold, it was hoped that we would simultaneously reduce referral of those likely to die of non-renal disease prior to reaching kidney failure. In a retrospective analysis of data from 10,874 adults with incident CKD (stages 3a–5), Walker et al confirmed that the KFRE effectively identifies individuals at high-risk of developing kidney failure, and does so earlier than the previous threshold criterion.³ Not only did the KFRE appear effective, butthe clinical burden of referrals appeared unchanged (at least numerically). Interestingly, and of relevance to the expansion of conservative kidney management, the authors defined kidney failure as the earliest of receipt of dialysis, a kidney transplant or an eGFR <15 mL/min/1.73 m².

This all sounds optimistic. However, Walker and colleagues found a uACR result was available for only 44.5% of individuals with CKD stages 3a–5.³ The KFRE could not be performed on more than 50% of individuals, and thus more than 50% of the population were immediately ineligible for referral to Nephrology under this criterion. It hardly comes as a surprise that women, the elderly and those living in socioeconomically deprived areas were the least likely to have recorded uACR values. We can assume that similar inequities would have been identified in other underserved groups (those of non-White ethnicity, with mental health disease or learning difficulties, for example) had the data been sufficiently granular to explore these variables. Of course, undertesting of albuminuria impacted eligibility for referral under the 2014 NICE guidelines also. Sadly, however, it appears that introduction of the KFRE may have exacerbated, rather than diminished, these dimensions of healthcare inequity.

Walker’s analysis, therefore, demonstrates that, with perfect clinical use, inclusion of the KFRE into UK referral guidelines should be effective, without significantly increasing the burden on the NHS. Yet again, however, these data also highlight the vast crater between guideline and clinical practice. Questions remain as to how large this crater is, and what we can do to overcome it. For example, what is the lag between an individual meeting the criteria for specialist referral (as in Walker’s data) and being referred? What inequities exist in this pathway between meeting criteria and realisation of referral for our underserved populations? 

Last, but not least, why is uACR testing proving so challenging? Remarkably, despite the number of papers reporting inadequacy of proteinuria assessments^4,5 and the range of interventions proposed to address these concerns,^6,7 there exists a single, survey-based study from the US investigating why measurement uACR is so poor.^8,9 I worry that, without high-quality qualitative research to truly understand barriers experienced by patients and clinicians, we are settling for a “best-guess” approach in addressing these issues. By doing so, we may not only limit the potential efficacy of our interventions, but simultaneously fail to address clear and inequitable disparities in healthcare access for our underserved populations.

A digest of the study can be read here.

References

1. Tangri N, Stevens LA, Griffith J et al (2011) A predictive model for progression of chronic kidney disease to kidney failure. JAMA 305: 1553–9
2. Major RW, Shepherd D, Medcalf JF et al (2019) The Kidney Failure Risk Equation for prediction of end stage renal disease in UK primary care: An external validation and clinical impact projection cohort study. PLoS Med 16: e1002955
3. Walker H, Khan S, Padmanabhan S et al (2025) Analysis of the kidney failure risk equation implementation in routine clinical practice and health inequalities in chronic kidney disease care: a retrospective cohort study. BMC Nephrol 26: 113
4. Sullivan MK, Jani BD, Rutherford E et al (2023) Potential impact of NICE guidelines on referrals from primary care to nephrology: a primary care database and prospective research study. Br J Gen Pract 73: e141–7
5. Kim LG, Cleary F, Wheeler DC et al (2018) How do primary care doctors in England and Wales code and manage people with chronic kidney disease? Results from the National Chronic Kidney Disease Audit. Nephrol Dial Transplant 33: 1373–9
6. Kam S, Angaramo S, Antoun J et al (2022) Improving annual albuminuria testing for individuals with diabetes. BMJ Open Qual 11: e001591
7. Keong F, Gander J, Wilson D et al (2023) Albuminuria screening in people with type 2 diabetes in a managed care organization. AJPM Focus 2: 100133
8. Groehl F, Garreta-Rufas A, Meredith K et al (2023) The drivers of non-adherence to albuminuria testing guidelines and the clinical and economic impact of not identifying chronic kidney disease. Clin Nephrol 100: 145–56
9. Abdel-Kader K, Greer RC, Boulware LE, Unruh ML (2014) Primary care physicians' familiarity, beliefs, and perceived barriers to practice guidelines in non-diabetic CKD: a survey study. BMC Nephrol 15: 64