When Numbers Start Making Decisions · Season Three, “The People the Average Does Not See” · Article 5
1. One digit can move a result across a diagnostic category
An estimated glomerular filtration rate of 59 mL/min/1.73 m² sits below a widely used boundary in chronic kidney disease classification. A result of 60 sits above it. On a pathology report, the two values can look like different states: one abnormal and one acceptable.
Kidney function did not change discontinuously between them. eGFR is calculated from serum creatinine and patient characteristics through an equation. It estimates filtration normalised to a standard body surface area. The result carries biological, laboratory and model uncertainty.
The KDIGO 2024 guideline defines chronic kidney disease as abnormalities of kidney structure or function present for at least three months, with implications for health. A GFR below 60 is one criterion, but chronicity is essential; other markers of kidney damage also matter. People in G1 or G2 categories are not classified as having CKD without evidence of kidney damage. KDIGO 2024 CKD Guideline
The line at 60 therefore organises risk and follow-up. It is not a diagnosis delivered by one digit on one day.
2. eGFR is produced by a model
Directly measuring GFR is complex and usually unnecessary in routine care. Laboratories measure creatinine, a waste product influenced by kidney filtration, then use an equation to estimate GFR. Age and sex are among the variables used in current equations. Results are normalised to 1.73 square metres of body surface area so people can be compared and disease categories applied.
Creatinine is also affected by muscle mass, diet, supplements, recent exercise, amputation, frailty and some medicines. A person with unusually low muscle mass can have a reassuring creatinine despite impaired filtration; a highly muscular person may have a higher creatinine without equivalent kidney disease. Acute illness and rapid change make a steady-state equation less reliable.
The Royal College of Pathologists of Australasia advises that eGFR is an estimate and identifies settings in which interpretation requires caution. RCPA Manual: GFR estimated
The equation is not defective because it cannot contain every body. Models make routine kidney assessment feasible. Their output becomes misleading only when the estimated label is forgotten and the result is treated as a direct, individually exact measurement.
3. Why 60 is useful
Clinical categories allow primary care, pathology systems and public health guidance to coordinate. Values below 60 over time are associated with higher risks of kidney failure, cardiovascular events and other complications. The threshold prompts clinicians to check chronicity, albuminuria, causes, medicines and progression.
It also supports a two-dimensional risk structure. GFR categories—from G1 through G5—are considered alongside urine albumin-to-creatinine ratio categories. Someone with an eGFR above 60 but substantial albuminuria may have clinically important CKD. Someone with a single eGFR of 59 during dehydration may not have chronic disease.
Kidney Health Australia’s handbook combines eGFR, albuminuria and clinical assessment to guide detection and management. Chronic Kidney Disease Management in Primary Care
The boundary is therefore a coordination point, not the whole map. It allows systems to say “look more closely here”. Trouble begins when a flag becomes a fixed identity, or when a result just above the line stops appropriate investigation.
4. Chronicity prevents a snapshot from becoming a permanent label
Kidney function changes with hydration, infection, medicines and acute illness. Laboratory imprecision and ordinary biological variation can shift results near 60. Repeating a test is not bureaucratic delay; it tests whether the finding persists.
KDIGO’s minimum three-month duration separates chronic kidney disease from temporary reduction or acute kidney injury. Historical results, repeat creatinine and other evidence of kidney damage establish the time dimension. A single eGFR can still require urgent action if the clinical context suggests acute deterioration, but urgency and chronic diagnosis are different judgements.
Electronic records often make categories durable. Once “CKD” enters a problem list, it can influence insurance, prescribing, referrals and the patient’s understanding long after the original result. That makes careful confirmation especially important. Conversely, repeatedly hovering between 59 and 61 should not produce alternating concern and reassurance. The trend and uncertainty are more meaningful than categorical flicker.
The system needs to preserve both: an actionable threshold and the evidence that qualifies it.
5. Normalisation can misfit an individual body
Reporting per 1.73 m² supports comparison and staging, but drug dosing or assessment in people with very small or large body size may require conversion to an absolute value and clinical advice. The standard body surface area is another population reference carried into an individual decision.
This does not mean that every eGFR must be personalised with a more complicated model. It means the purpose must be stated. A value appropriate for CKD risk classification may not answer a dosing question in an atypical body. A good result is not universally fit for every downstream use.
The Kidney Health Australia eGFR calculator and guidance emphasise that clinical interpretation should accompany the output. Kidney Health Australia: eGFR calculator
When a decision is high consequence—using a nephrotoxic medicine, approving contrast, determining drug dose or considering donation—clinicians may need additional measurements, alternative filtration markers or specialist input. The authority granted to the estimate should be proportional to how closely the person and purpose match the equation’s assumptions.
6. The report should communicate uncertainty, not conceal it
Pathology reports can improve interpretation by labelling the result as estimated, showing units and reference information, and avoiding false precision. Clinicians can explain that 59 and 60 are not two different kinds of kidney but adjacent estimates around a useful boundary.
Patients should be able to see the trend, albuminuria and reason for repeat testing. “Your kidneys are failing” overstates one borderline result; “everything is normal” may understate persistent risk. A more accurate explanation is that the estimate is slightly below the classification boundary and needs confirmation and context.
Review must include source data. Age, sex, sample handling and creatinine result should be correct. It must also include model fit: unusual muscle mass, rapid change or other circumstances may weaken the estimate. Finally, it must include decision fit: the evidence needed for diagnosis may differ from evidence needed for medication dosing.
7. A practical framework for equation-based thresholds
When an estimated result crosses a clinical boundary, ask:
- What was measured directly, and what was calculated?
- Which equation and assumptions produced the estimate?
- Does the individual fit those assumptions?
- How much biological and laboratory variation exists near the line?
- Is chronicity established through time?
- Which other markers—especially albuminuria—change the interpretation?
- Is the result being used for screening, diagnosis, risk prediction or dosing?
- Would a high-consequence decision justify confirmation or specialist review?
These questions do not weaken evidence-based medicine. They prevent a useful population equation from acquiring unsupported individual precision.
Conclusion: the equation should begin judgement, not impersonate it
eGFR transformed kidney care by making a clinically important function visible through routine blood testing. The threshold at 60 supports consistent detection and follow-up. Without it, early kidney disease would be missed more often.
My judgement is that 59 and 60 should be treated as neighbouring estimates whose meaning comes from persistence, albuminuria, trend, body characteristics and clinical purpose. The category is useful because it prompts a structured response. It becomes harmful when one result is mistaken for a complete diagnosis or a permanent identity.
An equation can estimate the individual from group evidence. Responsible care keeps open the route back from the estimate to the person.
Discover more from Geoffrey Chen
Subscribe to get the latest posts sent to your email.