NakedSignal
Talk to us

Free tool

Is this change real?

Two results from the same person never match exactly. Enter both, with the test's imprecision and the person's normal variation, to see whether the difference is likely a real change, including when the second result came from a different analyser.

()

Both results in the same unit; the test works on the ratio, so any unit will do.

Your laboratory's between-run CV for this test. The value shown is an example.

Filled from the analyte; change it if you have a better estimate.

If the second result came from a different analyser or method: how much higher (+) or lower (−) it reads than the first.

Result

Change between the results
Largest rise from variation alone
Largest fall from variation alone
Symmetric reference change value
Chance of a change this large from variation alone

Limits at 95% (Z = ), log-normal model. A result is a screening aid, not a diagnosis.

How the answer is worked out

Two results from the same person differ even when nothing has changed: the test has its own imprecision, and the body varies from day to day around its own set point. The reference change value is the largest difference those two sources of variation explain. A larger difference is likely to be a real change.

The classic formula is RCV = √2 × Z × √(CVA² + CVI²), with Z = 1.96 for any change and 1.65 for a change in one stated direction. Because many analytes are not symmetric around the set point, the calculator uses the log-normal limits, which allow a larger rise than fall: exp(±Z × √2 × σ) − 1, with σ = √ln(CVT² + 1) and CVT² = CVA² + CVI² [Díaz-Garzón Marco et al., 2020]. The chance shown is how often variation alone gives a change at least this large, under the same model.

When the second result came from a different analyser

If the laboratory changed analyser, reagent or method between the two results, part of the difference is the method, not the patient. Enter the method difference and the calculator removes it before testing. Without it, a method change can read as a change in the patient, or hide one. The analyser comparability lookup gives published differences between systems; your laboratory's own comparison is better still.

Where the biological variation comes from

  • Creatinine, serum or plasma: CVI 5.0%. Meta-analysis of published studies, outlier studies excluded (95% CI 4.7 to 5.4%) [Thöni et al., 2022].
  • Glucose, fasting plasma: CVI 5.0%. Meta-analysis of studies in healthy people appraised with the BIVAC checklist (95% CI 4.1 to 12.0%) [Ricós et al., 2020].
  • HbA1c: CVI 1.2%. Meta-analysis of studies in healthy people appraised with the BIVAC checklist (95% CI 0.3 to 2.5%) [Ricós et al., 2020].
  • Glycated albumin: CVI 1.4%. Meta-analysis of studies in healthy people appraised with the BIVAC checklist (95% CI 1.2 to 2.1%) [Ricós et al., 2020].

These estimates come from studies of healthy people; variation in disease can be larger. For any other analyte, choose “Other analyte” and enter a CVI from a biological variation database or study.

What it does not do

It assumes both results were measured with the stated imprecision and, unless you enter a method difference, on the same method. It does not replace clinical judgement, and it is not a diagnostic device.

The guide to the reference change value explains the formula term by term, how to choose Z, and where the calculation misleads.

Sources

  1. Díaz-Garzón Marco J, Fernández-Calle P, Ricós C (2020). Models to estimate biological variation components and interpretation of serial results: strengths and limitations. Advances in Laboratory Medicine. pmc.ncbi.nlm.nih.gov/articles/PMC10270238/ (opens in a new tab)
  2. Thöni S, Keller F, Denicolò S, et al. (2022). Biological variation and reference change value of the estimated glomerular filtration rate in humans: a systematic review and meta-analysis. Frontiers in Medicine. pmc.ncbi.nlm.nih.gov/articles/PMC9583397/ (opens in a new tab)
  3. Ricós C, Fernández-Calle P, Gonzalez-Lao E, et al. (2020). Critical appraisal and meta-analysis of biological variation studies on glycosylated albumin, glucose and HbA1c. Advances in Laboratory Medicine. pmc.ncbi.nlm.nih.gov/articles/PMC10197502/ (opens in a new tab)